Digital Image Steganography Using Eight-Directional PVD ...
14
Research Article Digital Image Steganography Using Eight-Directional PVD against RS Analysis and PDH Analysis Gandharba Swain Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Andhra Pradesh 522502, India Correspondence should be addressed to Gandharba Swain; [email protected] Received 16 June 2017; Revised 19 August 2017; Accepted 7 September 2017; Published 26 September 2018 Academic Editor: Martin Reisslein Copyright © 2018 Gandharba Swain. is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. e least significant bit (LSB) substitution techniques are detected by RS analysis and the traditional pixel value differencing (PVD) approaches are detected by pixel difference histogram (PDH) analysis. e PVD steganography can escape from PDH analysis by using the edges in multiple directions. is paper proposes a steganography technique by exploiting the edges in eight directions and also using LSB substitution to resist from both RS analysis and PDH analysis. For every 3×3 pixel block the central pixel is embedded with 3 or 4 bits of data by modified LSB substitution technique. en this new value of the central pixel is utilized to calculate eight difference values with eight neighboring pixels. ese eight difference values are used to hide the data. ere are two types with regard to two different range tables. Type 1 uses 3 bit modified LSB substitution and range table 1. Type 2 uses 4 bit modified LSB substitution and range table 2. Type 1 and type 2 are also known as variant 1 and variant 2, respectively. Type 1 possesses higher PSNR and type 2 possesses higher hiding capacity. 1. Introduction e LSB substitution is one of the simplest and well- known approaches for data hiding. is simplest technique is detected by RS analysis. Wu and Tsai [1] exposed the fact that the edge regions in an image can conceal more amounts of data as compared to the smooth regions. Based on this principle they proposed pixel value differencing (PVD) steganography. e image should be partitioned into different blocks, each of size 1×2 pixels. For a block the difference value between the two pixels is computed and changed to a new value by adding data to it. e PVD technique with block size of 2×2 has been proposed to improve the embedding capacity [2, 3]. In blocks of size 2×2, edges in three directions are considered. Chang and Tseng [4] considered the values of 2, 3, and 4 neighboring pixels, to calculate the pixel value differences. But they did not address the fall in error problem (FIEP). Yang et al. [5] calculated varieties of pixel value differences in 4-pixel blocks for data hiding. Hong et al. [6] used diamond encoding with pixel value differencing to achieve better PSNR. LSB substitution techniques offer higher embedding capacity, but PVD techniques offer higher imperceptibility. us PVD and LSB techniques have been combined to obtain larger hiding capacity and better imperceptibility [7, 8]. Chen [9] proposed a PVD steganography using two reference tables to randomize the data embedding. Khodaei and Faez [10] proposed a hybrid approach by combining LSB substitution and PVD in 1×3 pixel block. It was further extended to 2×2 size block in [11] to achieve better performance. Based on pixel value differences adaptive LSB substitution has been performed in [12, 13]. e traditional PVD steganography techniques follow a static range table. Due to this some undesired steps are introduced in pixel difference histograms of the stego images. is step effects can be avoided by applying two tricks, (i) exploiting the edges in multiple directions and (ii) using adaptive range table. Luo et al. [14] also proposed an adaptive PVD steganography with three-pixel blocks, which does not suffer with the step effects. Swain [15] proposed two adaptive PVD steganography techniques using vertical, horizontal, and diagonal edges, which do not suffer from step effects. e first technique uses pixel blocks of size 2×2 and the second technique uses pixel blocks of size 3×3. In general, adaptive Hindawi Advances in Multimedia Volume 2018, Article ID 4847098, 13 pages https://doi.org/10.1155/2018/4847098
Transcript of Digital Image Steganography Using Eight-Directional PVD ...
Research ArticleDigital Image Steganography Using Eight-Directional PVDagainst RS Analysis and PDH Analysis
Gandharba Swain
Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation VaddeswaramAndhra Pradesh 522502 India
Correspondence should be addressed to Gandharba Swain gswain1234gmailcom
Received 16 June 2017 Revised 19 August 2017 Accepted 7 September 2017 Published 26 September 2018
Academic Editor Martin Reisslein
Copyright copy 2018 Gandharba Swain This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited
The least significant bit (LSB) substitution techniques are detected by RS analysis and the traditional pixel value differencing (PVD)approaches are detected by pixel difference histogram (PDH) analysis The PVD steganography can escape from PDH analysis byusing the edges in multiple directions This paper proposes a steganography technique by exploiting the edges in eight directionsand also using LSB substitution to resist from both RS analysis and PDH analysis For every 3times3 pixel block the central pixel isembedded with 3 or 4 bits of data by modified LSB substitution technique Then this new value of the central pixel is utilized tocalculate eight difference values with eight neighboring pixels These eight difference values are used to hide the data There aretwo types with regard to two different range tables Type 1 uses 3 bit modified LSB substitution and range table 1 Type 2 uses 4bit modified LSB substitution and range table 2 Type 1 and type 2 are also known as variant 1 and variant 2 respectively Type 1possesses higher PSNR and type 2 possesses higher hiding capacity
1 Introduction
The LSB substitution is one of the simplest and well-known approaches for data hiding This simplest techniqueis detected by RS analysis Wu and Tsai [1] exposed thefact that the edge regions in an image can conceal moreamounts of data as compared to the smooth regions Basedon this principle they proposed pixel value differencing(PVD) steganography The image should be partitioned intodifferent blocks each of size 1times2 pixels For a block thedifference value between the two pixels is computed andchanged to a new value by adding data to it The PVDtechnique with block size of 2times2 has been proposed toimprove the embedding capacity [2 3] In blocks of size2times2 edges in three directions are considered Chang andTseng [4] considered the values of 2 3 and 4 neighboringpixels to calculate the pixel value differences But they didnot address the fall in error problem (FIEP) Yang et al[5] calculated varieties of pixel value differences in 4-pixelblocks for data hidingHong et al [6] used diamond encodingwith pixel value differencing to achieve better PSNR LSBsubstitution techniques offer higher embedding capacity but
PVD techniques offer higher imperceptibility Thus PVDand LSB techniques have been combined to obtain largerhiding capacity and better imperceptibility [7 8] Chen [9]proposed a PVD steganography using two reference tablesto randomize the data embedding Khodaei and Faez [10]proposed a hybrid approach by combining LSB substitutionand PVD in 1times3 pixel block It was further extended to 2times2size block in [11] to achieve better performance Based onpixel value differences adaptive LSB substitution has beenperformed in [12 13]
The traditional PVD steganography techniques followa static range table Due to this some undesired steps areintroduced in pixel difference histograms of the stego imagesThis step effects can be avoided by applying two tricks (i)exploiting the edges in multiple directions and (ii) usingadaptive range table Luo et al [14] also proposed an adaptivePVD steganography with three-pixel blocks which does notsuffer with the step effects Swain [15] proposed two adaptivePVD steganography techniques using vertical horizontaland diagonal edges which do not suffer from step effects Thefirst technique uses pixel blocks of size 2times2 and the secondtechnique uses pixel blocks of size 3times3 In general adaptive
HindawiAdvances in MultimediaVolume 2018 Article ID 4847098 13 pageshttpsdoiorg10115520184847098
2 Advances in Multimedia
Table 1 Range table 1 (for type 1)
Range R1=[0 7] R
2=[8 15] R
3=[16 31] R
4=[32 63] R
5=[64 127] R
6=[128 255]
capacity 3 3 3 3 4 4
Table 2 Range Table 2 ( for type 2)
Range R1=[0 7] R
2=[8 15] R
3=[16 31] R
4=[32 63] R
5=[64 127] R
6=[128 255]
capacity 3 3 4 5 6 6
P1 P2 P3
P4
P6 P7 P8
Pc P5
(a)
P1 P
2 P
3
P4 P
c P
5
P6 P
7 P
8
(b)
Figure 1 (a) Original block (b) stego block
image steganography schemes possess lower hiding capacityThe edges can be predicted by some prediction functions andhiding capacity depends upon this prediction If we hide insmooth regions distortion will be more Based on the level ofcomplexity of the edge regions adaptive embedding can beapplied [16] In this way capacity can be increased and chanceof detection can be decreased Balasubramanian et al [17]published PVD schemes with 3times3 size pixel blocks to achievehigher hiding capacity To prevent the detection from pixeldifference histogram (PDH) analysis multidirectional edgeshave been exploited in [18] Darabkh et al [19] also proposedPVD steganography using eight-directional PVD which is anextension fromWuandTsairsquos original PVDAPVDsteganog-raphy technique which is an extension of Wu and Tsairsquos PVDapproach should be verified by PDH analysis Gulve and Joshi[20] used PVD with LSB in six-pixel blocks to achieve higherembedding capacity In one of the six pixels LSB substitutionwas applied Then five differences were calculated and basedon these differences PVD approach was applied
Zhang and Wang [21] proposed exploiting modificationdirection (EMD) steganography The main goal in it is thata group of secret bits is converted to a digit in (2n+1)-arynotational system where n is the size of pixel block Thissecret digit could be hidden in the pixel block by addingplusmn1 to only one pixel In this technique the hiding capacityis not good Kim [22] advanced the EMD technique usingbasis vector and (2n+x minus 1)-ary notational system where nand x are user defined values Shen and Huang [23] madethe hiding capacity of a block adaptive using PVD withEMD to achieve higher hiding capacity and better PSNRNguyen et al [24] used multiple bit planes and pixel blockcomplexity measure to perform adaptive embedding In thismethod more number of bits is hidden in high texturedregions as compared to low textured regions In [25] the
steganography algorithm uses a public key and a privatekey to generate pseudo random numbers that identify theembedding locations
This paper proposes a combination of modified LSBsubstitution and eight-directional PVD in 3times3 pixel blocksIt is judiciously designed in such a manner that it can neitherbe detected by RS analysis nor be detected by PDH analysisFurthermore Wu and Tsairsquos PVD principle is not used
2 Proposed Eight-Directional PVD Technique
The embedding algorithm operates on 3times3 pixel blocks Theblocks are formed in a nonoverlapping manner Let us denotea typical block as in Figure 1(a) The central pixel is Pc and itssurrounding pixels are P
1 P2 P3 P4 P5 P6 P7 and P
8 In
central pixel t bit LSB substitution is performed For type 1the t value is 3 and range table is Table 1 For type 2 the tvalue is 4 and the range table is Table 2
After performing t bit LSB substitution in Pc the stegovalue is P1015840c Suppose the integer (decimal) value of t LSB bitsin Pc is decold and the integer (decimal) value of t LSB bits inP1015840c is decnew Now find the deviation dev = decold - decnewFurthermore calculate a modified value of P1015840c as in
P1015840c
=
P1015840c + 2t if dev gt 2tminus1 and 0 le P1015840c + 2t le 255P1015840c minus 2t if dev lt minus2tminus1 and 0 le P1015840c minus 2t le 255P1015840c otherwise
(1)
Using this modified value P1015840c calculate eight difference valuesdi for i= 1 to 8 as in
di =10038161003816100381610038161003816P1015840
c minus Pi10038161003816100381610038161003816 (2)
Advances in Multimedia 3
The difference value d1belongs to one of the six ranges
Suppose the lower bound is denoted as L1and hiding capacity
as t1for the range of d
1 Similarly L
2 L3 L4 L5 L6 L7 and L
8
are the lower bounds and t2 t3 t4 t5 t6 t7 and t
8are the
hiding capacities of the ranges of d2 d3 d4 d5 d6 d7 and d
8
respectivelyNow take t
1 t2 t3 t4 t5 t6 t7 and t
8bits of data separately
and convert to their decimal values S1 S2 S3 S4 S5 S6 S7 and
S8 respectivelyNow for i=1 to 8 calculate new difference values d1015840i using
d1015840i = Li + Si (3)
For i = 1 to 8 a pair of new values for each Pi is calculated asin
P10158401015840i = P1015840c minus d1015840i (4)
P101584010158401015840i = P1015840c + d1015840i (5)
Now out of two new values P10158401015840i and P101584010158401015840i one value is chosenfor P1015840i using
P1015840i
=
P10158401015840i if 10038161003816100381610038161003816Pi minus P10158401015840i10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 255P101584010158401015840i otherwise
(6)
Thus the stego pixel block after hiding t t1 t2 t3 t4 t5 t6 t7
and t8bits of data in pixels Pc P1 P2 P3 P4 P5 P6 P7 and
P8 respectively is as shown in Figure 1(b)
The extraction algorithm operates on 3times3 pixel blocksas in embedding algorithm The blocks are formed in anonoverlapping manner A typical stego block is representedin Figure 1(b) Retrieve the t LSBs from P1015840c For i=1 to 8 findeight difference values as in
d1015840i =10038161003816100381610038161003816P1015840
c minus P1015840i10038161003816100381610038161003816 (7)
Each d1015840i belongs to a rangewhose embedding capacity is ti andlower bound is Li For i=1 to 8 find the decimal values of theembedded secret bit streams using
Si = d1015840i minus Li (8)
Now convert S1 S2 S3 S4 S5 S6 S7 and S
8into t1 t2 t3 t4
t5 t6 t7 and t
8binary bits respectively
The correctness of the extraction procedure is justifiedby the following mathematical discussions That means how(8) is derived from (7) is described below
As per (7) d1015840i = |P1015840c minus P1015840i |From (6) P1015840i is P
10158401015840
i if |Pi minus P10158401015840i | lt |Pi minus P101584010158401015840i | and 0 le P10158401015840i le255 Otherwise P1015840i is P101584010158401015840i So d1015840i can be defined as in
d1015840i
=
10038161003816100381610038161003816P1015840
c minus P10158401015840i10038161003816100381610038161003816 if 10038161003816100381610038161003816Pi minus P10158401015840i
10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 25510038161003816100381610038161003816P1015840
c minus P101584010158401015840i10038161003816100381610038161003816 otherwise
(9)
From (4) and (5) P10158401015840i = (P1015840c - d1015840i ) and P101584010158401015840i = (P1015840c + d1015840i )respectively Substituting these values of P10158401015840i and P101584010158401015840i in (9)(10) is obtained
d1015840i =
10038161003816100381610038161003816P1015840
c minus (P1015840c minus d1015840i)10038161003816100381610038161003816 if 10038161003816100381610038161003816Pi minus P10158401015840i
10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 25510038161003816100381610038161003816P1015840
c minus (P1015840c + d1015840i)10038161003816100381610038161003816 otherwise
(10)
Equation (10) can be simplified as
d1015840i =
10038161003816100381610038161003816d1015840
i10038161003816100381610038161003816 if 10038161003816100381610038161003816Pi minus P10158401015840i
10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 25510038161003816100381610038161003816minusd1015840
i10038161003816100381610038161003816 otherwise
(11)
From (3) d1015840i = Li+ Si Substituting d1015840i on right hand side of(11) (12) is obtainedd1015840i
=
Li + Si if 10038161003816100381610038161003816Pi minus P10158401015840i10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 255Li + Si otherwise
(12)
From (12) it can be concluded that d1015840i = Li+ Si or Si= d1015840i minusLi This is nothing but (8) used in extraction procedureThuscorrectness of (8) is justified
3 Examples for Embedding and Extraction
Figure 2 represents an example of embedding for type 1with t=3 of the proposed technique It is self-explanatory To
understand this example it can be traversed from left to rightand top to bottom
Figure 3 represents an example of extraction for type 1with t=3 of the proposed technique It is self-explanatory Tounderstand this example it can be traversed from left to rightand top to bottom
4 Results and Discussion
The embedding and extraction algorithms are simulated inMATLAB Standard RGB color images from SIPI image database are used for testing The R G and B components (each8 bits) of a pixel are treated as a single unit For eight sampleimages the results are discussed and averaged Figure 4 showsthe original images Figure 5 shows the stego images of type 1and Figure 6 shows the stego images of type 2 In each of thesestego images seven lakhs (700000) bits of data is hiddenThe evaluation parameters are (i) hiding capacity (ii) bit rate(BPB) (iii) PSNR and quality index (Q) The hiding capacityis the total number of bits that an image can conceal The bits
4 Advances in Multimedia
Secret data to be embedded 000 100 100 010 001 010 100 110 101 t = 3
192 190 200
192 190 188
192 190 200
Original block
192 190 200
192 184 188
192 190 200
[0 7] 3
[8 15] 3
[16 31] 3
[32 63] 3
[64 127] 4
[128 255] 4
Range table 1
196 188 202
193 184 186
196 190 205
Stego block
Value of Pc is 190
190 = (10111)2
3 bits of data =
(10111)2 = 184
t bit LSB substitution at Pc
After t bit LSB substitution at Pc
Pc is 184
d1 = |184 minus 192| = 8
d2 = |184 minus 190| = 6
d3 = |184 minus 200| = 16
d4 = |184 minus 192| = 8
d5 = |184 minus 188| = 4
d6 = |184 minus 192| = 8
d7 = |184 minus 190| = 6
d8 = |184 minus 200| = 16
t1 = 3
t2 = 3
t3 = 3
t4 = 3
t5 = 3
t6 = 3
t7 = 3
t8 = 3
L1 = 8
L2 = 0
L3 = 16
L4 = 8
L5 = 0
L6 = 8
L7 = 0
L8 = 16
S1 = 4
S2 = 4
S3 = 2
S4 = 1
S5 = 2
S6 = 4
S7 = 6
S8 = 5
d1 = 8 + 4 = 12
d2 = 0 + 4 = 4
d4 = 8 + 1 = 9
d5 = 0 + 2 = 2
d6 = 8 + 4 = 12
d7 = 0 + 6 = 6
d3 = 16 + 2 = 18
d8 = 16 + 5 = 21
P1 = 184 minus 12 = 172
P3 = 184 minus 18 = 166
P6 = 184 minus 12 = 172
P8 = 184 minus 21 = 163
P2 = 184 minus 4 = 180
P4 = 184 minus 9 = 175
P5 = 184 minus 2 = 182
P7 = 184 minus 6 = 178
P1 = 184 + 12 = 196
P3 = 184 + 18 = 202
P6 = 184 + 12 = 196
P8 = 184 + 21 = 205
P2 = 184 + 4 = 188
P4 = 184 + 9 = 193
P5 = 184 + 2 = 186
P7 = 184 + 6 = 190
P1 = 196 P
2 = 188 P
3 = 202
P5 = 186
P8 = 205P
7 = 190
P4 = 193
P6 = 196
Pc = 184
Figure 2 Example of embedding
Advances in Multimedia 5
196 188 202
193 184 186
196 190 205
Stego block
[0 7] 3
[8 15] 3
[16 31] 3
[32 63] 3
[64 127] 4
[128 255] 4
Range table 1
4 4 2 1 2 4 6 5
Extracted data 000 100 100 010 001 010 100 110 101
|184 minus 196| = 12
|184 minus 188| = 4
|184 minus 202| = 18
|184 minus 193| = 9
|184 minus 186| = 2
|184 minus 196| = 12
|184 minus 190| = 6
|184 minus 205| = 21Pc is 184
184 = (10111)2
t1 = 3
t2 = 3
t3 = 3
t4 = 3
t5 = 3
t6 = 3
t7 = 3
t8 = 3
L1 = 8
L2 = 0
L3 = 16
L4 = 8
L5 = 0
L6 = 8
L7 = 0
L8 = 16
S1 = 12 minus 8 = 4
S3 = 18 minus 6 = 2
S6 = 12 minus 8 = 4
S8 = 21 minus 16 = 5
S2 = 4 minus 0 = 4
S4 = 9 minus 8 = 1
S5 = 2 minus 0 = 2
S7 = 6 minus 0 = 6
Si rarr
d1 =
d2 =
d3 =
d4 =
d5 =
d6 =
d7 =
d8 =
Figure 3 Example of extraction
(a) Pot (b) House (c) Boat (d) Jet
(e) Peppers (f) Tiffany (g) Baboon (h) Lena
Figure 4 Original images
per byte (BPB) indicate the average hiding capacity per a byteof the original image Equation (13) is used to calculate BPB
BPB = hiding capacityimage size in bytes (13)
Distortion in stego image is measured in terms of PSNREquation (14) is used to calculate it
PSNR = 10 times log10
255 times 255MSE
(14)
where MSE denotes the mean square error and can becomputed using
MSE = 1m times n
msumi=1
nsumj=1(pij minus qij)
2
(15)
where pij is a pixel of cover image and qij is the correspondingstego pixel
Thequality index (Q) is another parameter tomeasure thestructural similarity between original image and stego image
6 Advances in Multimedia
PSNR = 4115(a)
PSNR = 3969(b)
PSNR = 3725(c)
PSNR = 4221(d)
PSNR = 3781(e)
PSNR = 4146(f)
PSNR = 3426(g)
PSNR = 4262(h)
Figure 5 Stego images of proposed type 1 (t=3 range table 1)
PSNR = 3711(a)
PSNR = 3839(b)
PSNR = 3408(c)
PSNR = 4038(d)
PSNR = 3469(e)
PSNR = 4053(f)
PSNR = 3209(g)
PSNR = 4048(h)
Figure 6 Stego images of proposed type 2 (t=4 range table 2)
Advances in Multimedia 7
Table 3 Results of Pradhan et alrsquos seven-way PVD technique [18]
Images 512times512times3 Seven-way PVD [18] Seven-way+ one-way PVD [18]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4237 1795551 09996 228 4240 1803635 09996 229House 3897 1972223 09986 250 3898 1977403 09986 251Boat 3789 1972086 09988 250 3791 1991005 09988 253Jet 4209 1906254 09989 242 4198 1909595 09989 242Peppers 4042 1778072 09993 226 4020 1806166 09992 229Tiffany 4123 1400756 09986 178 4130 1450799 09986 184Baboon 3379 2226806 09957 283 3377 2243218 09957 285Lena 4173 1896662 09993 241 4173 1901149 09993 241Average 3981 1868551 09986 237 3978 1885371 09985 239
Table 4 Results of Khodaei and Faezrsquos technique
Images 512times512times3 Khodaei and Faez type 1 [10] Khodaei and Faez type 2 [10]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4121 2366001 09995 300 3783 2387494 09990 303House 3916 2387183 09986 303 3875 2470824 09985 314Boat 4005 2391994 09993 304 3749 2504613 09987 318Jet 4089 2374048 09986 301 4064 2418419 09985 307Peppers 3932 2372858 09991 301 3791 2435223 09987 309Tiffany 3973 2372396 09980 301 4025 2416944 09982 307Baboon 3777 2443361 09982 310 3449 2662080 09963 338Lena 4278 2375248 09995 302 4125 2434603 09993 309Average 4011 2385386 09988 303 3857 2466275 09984 313
Equation (16) is used to compute itThemaximum value of Qcan be 1 if original image p and stego image q are the same
Q =4120590xyp q
(1205902x + 1205902y) [(p)2 + (q)2] (16)
where p stands for the mean pixel value of the original imageq stands for the mean pixel value of the stego image1205902x standsfor the standard deviation of pixel values of original image1205902y stands for the standard deviation of pixel values of stegoimage and 120590xy is the covariance between original image andstego image Equations (17) (18) (19) (20) and (21) are usedto calculate these values
p = 1m times n
msumi=1
nsumj=1pij (17)
q = 1m times n
msumi=1
nsumj=1qij (18)
1205902x =1
m times n minus 1msumi=1
nsumj=1(pij minus p)
2
(19)
1205902y =1
m times n minus 1msumi=1
nsumj=1(qij minus q)
2
(20)
120590xy =1
m times n minus 1msumi=1
nsumj=1(pij minus p) (qij minus q) (21)
Table 3 notes the results of seven-way PVD [18] whichis an extension of Wu and Tsairsquos [1] technique Table 4 notes
the results of Khodaei and Faezrsquos technique [10] and Table 5notes the results of the proposed technique The PSNR andcapacity of the proposed technique (both type 1 and type2) are better than seven-way PVD The bit rate and Q arealso better Although as compared to Khodaei and Faezrsquostechnique the PSNR is not improved but the hiding capacityis improved The major advantage over Khodaei and Faezrsquostechnique is that the proposed technique is resistant to PDHanalysis but Khodaei and Faezrsquos technique is detectable byPDH analysis This is made possible by exploiting the edgesin eight directions
Now let us observe the results of proposed technique andcompare type 1 and type 2 We can observe that the capacityis higher in type 2 as compared to type 1 This is because ofthe 4 LSBs substitution in type 2 at the central pixel ie the tvalue is 4 On the other hand the PSNR of type 1 is better thanthat of type 2
Table 6 represents the comparison of the proposedtechnique with the adaptive PVD technique in [26] andhybrid techniques in [27 28] The comparison parametersare average PSNR and average hiding capacity values Thetwo variants of adaptive PVD technique possess lesser hidingcapacity as compared to the proposed eight-directional PVDAs the hiding capacity of adaptive PVD [26] is very littleso PSNR is larger The hybrid technique proposed in [27] isa combination of PVD LSB and EMD The hiding capacityof proposed technique is greater than all the four variantsof the hybrid technique proposed in [27] The technique in[28] uses PVD with modified LSB (MLSB) substitution Theresults of this technique (type 1) are as good as the results of
8 Advances in Multimedia
Table 5 Results of proposed eight-directional PVD technique
Images 512times512times3 Proposed type 1 (t=3 Range Table 1) Proposed type 2 (t=4 Range Table 2)PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4115 2354240 09996 299 3711 2475977 09988 315House 3969 2358575 09988 300 3839 2625804 09984 334Boat 3725 2364685 09987 301 3408 2659795 09973 338Jet 4221 2356828 09989 300 4038 2538801 09985 323Peppers 3781 2356645 09988 300 3469 2544392 09975 324Tiffany 4146 2353505 09987 299 4053 2511139 09984 319Baboon 3426 2392573 09961 304 3209 2939376 09937 374Lena 4262 2353892 09995 299 4048 2533551 09992 322Average 3955 2361368 09986 300 3722 2603604 09977 331
Table 6 Comparison of the proposed technique with techniques in [26ndash28]
Technique PSNR Capacity Amount of hidden data in stego image to compute PSNR2times3 adaptive PVD [26] 5093 1445645 One lakh and forty thousand (140000) bits3times2 adaptive PVD [26] 5093 14356233 PVD +3 LSB +EMD type 1 [27] 4107 1686041
Seven lakhs (700000) bits3 PVD +3 LSB +EMD type 2 [27] 3895 17988347 PVD +3 LSB +EMD type 1 [27] 4028 17108887 PVD +3 LSB +EMD type 2 [27] 3726 1823282Proposed eight-directional PVD type 1 3955 2361368 Seven lakhs (700000) bitsProposed eight-directional PVD type 2 3722 2603604Five-directional PVD+MLSB type 1 [28] 3762 2399853
Eight lakhs and forty thousand (840000) bitsFive-directional PVD+MLSB type 2 [28] 3623 2514038Eight-directional PVD+MLSB type 1 [28] 3749 2394086Eight-directional PVD+MLSB type 2 [28] 3560 2603604
the proposed technique (type 1) Similarly the results of thistechnique (type 2) are as good as the results of the proposedtechnique (type 2)
5 Security Analysis
The LSB substitution techniques suffer from RS analysis andPVD techniques suffer from PDH analysis The proposedtechnique uses the concepts like modified LSB substitutionandPVD so itmust be analyzed by both RS analysis andPDHanalysis
The PDH analysis is performed by calculating the differ-ence value between every pair of pixels A pair comprises twoconsecutive pixels This difference values will be from -255 to+255 including 0 The frequency of each of these differencevalues is counted A graph is plotted with the pixel differencevalue on X-axis and frequency on Y-axis The curve obtainedis called the PDHThePDHof original imagewill be a smoothcurve If the PDH of stego image is also a smooth curve thensteganography is not detected If the PDH of stego imageshows step effects then steganography is identified
Figure 7 is the PDH analysis of type 1 and Figure 8 isthe PDH analysis of type 2 of the proposed technique Theanalysis is carried over all the eight test images For eachimage the PDH of the original image is represented by solidline and the PDH of the stego image is represented by dottedlineThe solid line curves will be obviously free of step effectsas they are of the original images The dotted line curves in
all the sixteen cases do not show any step effects This justifiesthat the proposed technique is resistant to PDH analysis
The RS analysis is based on statistical measures It is donein the following manner Define a function F
1 2nlarrrarr2n+1
It defines two transformations (i) from value 2n to value2n + 1 and (ii) from value 2n + 1 to value 2n Similarlydefine another function F
minus1 2n larrrarr2n -1 It defines other
two transformations (i) from value 2n to value 2n ndash 1 and(ii) from value 2n - 1 to value 2n The image say M isdivided into a number of equal size blocks Suppose such ablock is G whose pixels are X
1X2X3 Xn Then use the
function f(X1X2X3 Xn)=sumnminus1i=1 |Xi+1minusXi| tomeasure the
smoothness of G Then apply F1to all the blocks of M and
define the two parameters Rm and Sm as in (22) and (23)Similarly apply F
minus1to all the blocks of M and define the two
parameters Rminusm and S
minusm as in (24) and (25)
Rm
= No of blocks satisfying the condition f (F1(G)) gt f (G)
Total number of blocks
(22)
Sm
= No of blocks satisfying the condition f (F1(G)) lt f (G)
Total number of blocks
(23)
Rminusm
= No of blocks satisfying the condition f (Fminus1(G)) gt f (G)
Total number of blocks
(24)
Advances in Multimedia 9
0
1
2
Occ
urre
nces
3 times104
Lena coverLena stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3 times104
Baboon coverBaboon stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
Tiffany coverTiffany stego
times104
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Pot coverPot stego
times104
Figure 7 PDH analysis for type 1 of the proposed technique
10 Advances in Multimedia
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Lena coverLena stego
times104
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Baboon coverBaboon stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Tiffany coverTiffany stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Pot coverPot stego
times104
Figure 8 PDH analysis for type 2 of the proposed technique
Advances in Multimedia 11
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(a) Type 1 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(b) Type 1 Baboon
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(c) Type 2 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(d) Type 2 Baboon
Figure 9 RS analysis for type 1 and type 2 of the proposed technique
Sminusm
= No of blocks satisfying the condition f (Fminus1(G)) lt f (G)
Total number of blocks
(25)
RS analysis is performed by using these four parametersIf the condition R
119898asymp Rminus119898gt Sm asymp S
minusm is true then RSanalysis fails to detect the steganography technique But if the
condition Rminus119898
- Sminusm gt Rm - Sm is true then the RS analysis
succeeds in detecting the steganography techniqueFigure 9 represents RS analysis for the proposed tech-
nique Figures 9(a) and 9(b) stand for RS analysis over Lenaand Baboon images respectively of type 1 Similarly Figures9(c) and 9(d) stand for RS analysis over Lena and Baboonimages respectively of type 2 In all the four cases the
12 Advances in Multimedia
condition R119898asymp Rminus119898gt Sm asymp Sminusm is true so we can conclude
that RS analysis could not detect the proposed techniqueLena image has more smooth regions and Baboon imagehas more edge regions all the remaining six images are inbetween these two in terms of smoothness So it is sufficientto draw a conclusion based on these two images
6 Conclusion
The traditional PVD steganography techniques follow a staticrange table Due to this some undesired steps are introducedin pixel difference histograms of the stego images This stepeffect could be avoided by exploiting edges in eight differentdirections This paper proposes a steganography techniqueusing a judicious combination of LSB substitution and eight-directional PVD The LSB substitution is performed only inone pixel out of nine pixels in a block so RS analysis cannotdetect it By using the edges in multiple directions the PDHanalysis cannot detect itThere exist two types of the proposedtechnique with regard to two different quantization tablesThe type 1 provides higher PSNR and the type 2 provideshigher hiding capacity This proposed technique performsbetter in terms of hiding capacity and PSNR over othercompeting PVD techniques
Conflicts of Interest
The author declares that there are no conflicts of interest
References
[1] D-CWu andW-H Tsai ldquoA steganographicmethod for imagesby pixel-value differencingrdquo Pattern Recognition Letters vol 24no 9-10 pp 1613ndash1626 2003
[2] K C Chang C P Chang P S Huang and T M Tu ldquoAnovel image steganographic method using tri-way pixel-valuedifferencingrdquo Journal of Multimedia vol 3 no 2 pp 37ndash442008
[3] Y-P Lee J-C Lee W-K Chen K-C Chang I-J Su and C-P Chang ldquoHigh-payload image hiding with quality recoveryusing tri-way pixel-value differencingrdquo Information Sciencesvol 191 pp 214ndash225 2012
[4] C-C Chang and H-W Tseng ldquoA steganographic method fordigital images using side matchrdquo Pattern Recognition Lettersvol 25 no 12 pp 1431ndash1437 2004
[5] C-H Yang C-Y Weng H-K Tso and S-J Wang ldquoA datahiding scheme using the varieties of pixel-value differencing inmultimedia imagesrdquo The Journal of Systems and Software vol84 no 4 pp 669ndash678 2011
[6] W Hong T-S Chen and C-W Luo ldquoData embedding usingpixel value differencing and diamond encoding with multiple-base notational systemrdquo The Journal of Systems and Softwarevol 85 no 5 pp 1166ndash1175 2012
[7] H C Wu N I Wu C S Tsai and M-S Hwang ldquoImagesteganographic scheme based on pixel-value differencing andLSB replacementmethodsrdquo IEE ProceedingsmdashVision Image andSignal Processing vol 152 no 5 pp 611ndash615 2005
[8] C-H Yang C-Y Weng S-J Wang and H-M Sun ldquoVariedPVD + LSB evading detection programs to spatial domain in
data embedding systemsrdquo The Journal of Systems and Softwarevol 83 no 10 pp 1635ndash1643 2010
[9] J Chen ldquoA PVD-based data hiding method with histogrampreserving using pixel pair matchingrdquo Signal Processing ImageCommunication vol 29 no 3 pp 375ndash384 2014
[10] MKhodaei andK Faez ldquoNew adaptive steganographicmethodusing least-significant-bit substitution and pixel-value differ-encingrdquo IET Image Processing vol 6 no 6 pp 677ndash686 2012
[11] G Swain ldquoA steganographic method combining LSB substitu-tion and PVD in a blockrdquo Procedia Computer Science vol 85pp 39ndash44 2016
[12] X Liao Q Wen and J Zhang ldquoA steganographic methodfor digital images with four-pixel differencing and modifiedLSB substitutionrdquo Journal of Visual Communication and ImageRepresentation vol 22 no 1 pp 1ndash8 2011
[13] G Swain ldquoDigital image steganography using nine-pixel dif-ferencing and modified LSB substitutionrdquo Indian Journal ofScience and Technology vol 7 no 9 pp 1444ndash1450 2014httpwwwindjstorgindexphpindjstarticle
[14] W Luo F Huang and J Huang ldquoA more secure steganographybased on adaptive pixel-value differencing schemerdquoMultimediaTools and Applications vol 52 no 2-3 pp 407ndash430 2011
[15] G Swain ldquoAdaptive pixel value differencing steganographyusing both vertical and horizontal edgesrdquoMultimedia Tools andApplications vol 75 no 21 pp 13541ndash13556 2016
[16] S Chakraborty A S Jalal and C Bhatnagar ldquoLSB based nonblind predictive edge adaptive image steganographyrdquoMultime-dia Tools and Applications vol 76 no 6 pp 7973ndash7987 2017
[17] C Balasubramanian S Selvakumar and S Geetha ldquoHighpayload image steganography with reduced distortion usingoctonary pixel pairing schemerdquoMultimedia Tools and Applica-tions vol 73 no 3 pp 2223ndash2245 2014
[18] A Pradhan K R Sekhar and G Swain ldquoDigital imagesteganography based on seven way pixel value differencingrdquoIndian Journal of Science and Technology vol 9 no 37 pp 1ndash112016
[19] K A Darabkh A K Al-Dhamari and I F Jafar ldquoA newsteganographic algorithm based on multi directional PVD andmodified LSBrdquo Information Technology and Control vol 46 no1 pp 16ndash36 2017
[20] A K Gulve and M S Joshi ldquoA high capacity secured imagesteganography method with five pixel pair differencing andLSB substitutionrdquo International Journal of Image Graphics andSignal Processing vol 7 no 5 pp 66ndash74 2015
[21] X Zhang and S Wang ldquoEfficient steganographic embeddingby exploiting modification directionrdquo IEEE CommunicationsLetters vol 10 no 11 pp 781ndash783 2006
[22] C Kim ldquoData hiding by an improved exploiting modificationdirectionrdquoMultimedia Tools and Applications vol 69 no 3 pp569ndash584 2014
[23] S Y Shen and L H Huang ldquoA data hiding scheme usingpixel value differencing and improving exploiting modificationdirectionsrdquo Computers amp Security vol 48 pp 131ndash141 2015
[24] T D Nguyen S Arch-Int and N Arch-Int ldquoAn adaptivemulti bit-plane image steganography using block data-hidingrdquoMultimediaTools andApplications vol 75 no 14 pp 8319ndash83452016
[25] A Soria-Lorente and S Berres ldquoA secure steganographicalgorithm based on frequency domain for the transmission ofhidden informationrdquo Security and Communication Networksvol 2017 Article ID 5397028 14 pages 2017
Advances in Multimedia 13
[26] A Pradhan K R Sekhar and G Swain ldquoAdaptive PVDsteganography using horizontal vertical and diagonal edges insix-pixel blocksrdquo Security and Communication Networks vol2017 Article ID 1924618 13 pages 2017
[27] A Pradhan K R ekhar andG Swain ldquoDigital image steganog-raphy using LSB substitution PVD and EMDrdquo MathematicalProblems in Engineering 2018
[28] G Swain ldquoHigh capacity image steganography using modifiedLSB substitution and PVD against pixel difference histogramanalysisrdquo Security and Communication Networks 2018
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
2 Advances in Multimedia
Table 1 Range table 1 (for type 1)
Range R1=[0 7] R
2=[8 15] R
3=[16 31] R
4=[32 63] R
5=[64 127] R
6=[128 255]
capacity 3 3 3 3 4 4
Table 2 Range Table 2 ( for type 2)
Range R1=[0 7] R
2=[8 15] R
3=[16 31] R
4=[32 63] R
5=[64 127] R
6=[128 255]
capacity 3 3 4 5 6 6
P1 P2 P3
P4
P6 P7 P8
Pc P5
(a)
P1 P
2 P
3
P4 P
c P
5
P6 P
7 P
8
(b)
Figure 1 (a) Original block (b) stego block
image steganography schemes possess lower hiding capacityThe edges can be predicted by some prediction functions andhiding capacity depends upon this prediction If we hide insmooth regions distortion will be more Based on the level ofcomplexity of the edge regions adaptive embedding can beapplied [16] In this way capacity can be increased and chanceof detection can be decreased Balasubramanian et al [17]published PVD schemes with 3times3 size pixel blocks to achievehigher hiding capacity To prevent the detection from pixeldifference histogram (PDH) analysis multidirectional edgeshave been exploited in [18] Darabkh et al [19] also proposedPVD steganography using eight-directional PVD which is anextension fromWuandTsairsquos original PVDAPVDsteganog-raphy technique which is an extension of Wu and Tsairsquos PVDapproach should be verified by PDH analysis Gulve and Joshi[20] used PVD with LSB in six-pixel blocks to achieve higherembedding capacity In one of the six pixels LSB substitutionwas applied Then five differences were calculated and basedon these differences PVD approach was applied
Zhang and Wang [21] proposed exploiting modificationdirection (EMD) steganography The main goal in it is thata group of secret bits is converted to a digit in (2n+1)-arynotational system where n is the size of pixel block Thissecret digit could be hidden in the pixel block by addingplusmn1 to only one pixel In this technique the hiding capacityis not good Kim [22] advanced the EMD technique usingbasis vector and (2n+x minus 1)-ary notational system where nand x are user defined values Shen and Huang [23] madethe hiding capacity of a block adaptive using PVD withEMD to achieve higher hiding capacity and better PSNRNguyen et al [24] used multiple bit planes and pixel blockcomplexity measure to perform adaptive embedding In thismethod more number of bits is hidden in high texturedregions as compared to low textured regions In [25] the
steganography algorithm uses a public key and a privatekey to generate pseudo random numbers that identify theembedding locations
This paper proposes a combination of modified LSBsubstitution and eight-directional PVD in 3times3 pixel blocksIt is judiciously designed in such a manner that it can neitherbe detected by RS analysis nor be detected by PDH analysisFurthermore Wu and Tsairsquos PVD principle is not used
2 Proposed Eight-Directional PVD Technique
The embedding algorithm operates on 3times3 pixel blocks Theblocks are formed in a nonoverlapping manner Let us denotea typical block as in Figure 1(a) The central pixel is Pc and itssurrounding pixels are P
1 P2 P3 P4 P5 P6 P7 and P
8 In
central pixel t bit LSB substitution is performed For type 1the t value is 3 and range table is Table 1 For type 2 the tvalue is 4 and the range table is Table 2
After performing t bit LSB substitution in Pc the stegovalue is P1015840c Suppose the integer (decimal) value of t LSB bitsin Pc is decold and the integer (decimal) value of t LSB bits inP1015840c is decnew Now find the deviation dev = decold - decnewFurthermore calculate a modified value of P1015840c as in
P1015840c
=
P1015840c + 2t if dev gt 2tminus1 and 0 le P1015840c + 2t le 255P1015840c minus 2t if dev lt minus2tminus1 and 0 le P1015840c minus 2t le 255P1015840c otherwise
(1)
Using this modified value P1015840c calculate eight difference valuesdi for i= 1 to 8 as in
di =10038161003816100381610038161003816P1015840
c minus Pi10038161003816100381610038161003816 (2)
Advances in Multimedia 3
The difference value d1belongs to one of the six ranges
Suppose the lower bound is denoted as L1and hiding capacity
as t1for the range of d
1 Similarly L
2 L3 L4 L5 L6 L7 and L
8
are the lower bounds and t2 t3 t4 t5 t6 t7 and t
8are the
hiding capacities of the ranges of d2 d3 d4 d5 d6 d7 and d
8
respectivelyNow take t
1 t2 t3 t4 t5 t6 t7 and t
8bits of data separately
and convert to their decimal values S1 S2 S3 S4 S5 S6 S7 and
S8 respectivelyNow for i=1 to 8 calculate new difference values d1015840i using
d1015840i = Li + Si (3)
For i = 1 to 8 a pair of new values for each Pi is calculated asin
P10158401015840i = P1015840c minus d1015840i (4)
P101584010158401015840i = P1015840c + d1015840i (5)
Now out of two new values P10158401015840i and P101584010158401015840i one value is chosenfor P1015840i using
P1015840i
=
P10158401015840i if 10038161003816100381610038161003816Pi minus P10158401015840i10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 255P101584010158401015840i otherwise
(6)
Thus the stego pixel block after hiding t t1 t2 t3 t4 t5 t6 t7
and t8bits of data in pixels Pc P1 P2 P3 P4 P5 P6 P7 and
P8 respectively is as shown in Figure 1(b)
The extraction algorithm operates on 3times3 pixel blocksas in embedding algorithm The blocks are formed in anonoverlapping manner A typical stego block is representedin Figure 1(b) Retrieve the t LSBs from P1015840c For i=1 to 8 findeight difference values as in
d1015840i =10038161003816100381610038161003816P1015840
c minus P1015840i10038161003816100381610038161003816 (7)
Each d1015840i belongs to a rangewhose embedding capacity is ti andlower bound is Li For i=1 to 8 find the decimal values of theembedded secret bit streams using
Si = d1015840i minus Li (8)
Now convert S1 S2 S3 S4 S5 S6 S7 and S
8into t1 t2 t3 t4
t5 t6 t7 and t
8binary bits respectively
The correctness of the extraction procedure is justifiedby the following mathematical discussions That means how(8) is derived from (7) is described below
As per (7) d1015840i = |P1015840c minus P1015840i |From (6) P1015840i is P
10158401015840
i if |Pi minus P10158401015840i | lt |Pi minus P101584010158401015840i | and 0 le P10158401015840i le255 Otherwise P1015840i is P101584010158401015840i So d1015840i can be defined as in
d1015840i
=
10038161003816100381610038161003816P1015840
c minus P10158401015840i10038161003816100381610038161003816 if 10038161003816100381610038161003816Pi minus P10158401015840i
10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 25510038161003816100381610038161003816P1015840
c minus P101584010158401015840i10038161003816100381610038161003816 otherwise
(9)
From (4) and (5) P10158401015840i = (P1015840c - d1015840i ) and P101584010158401015840i = (P1015840c + d1015840i )respectively Substituting these values of P10158401015840i and P101584010158401015840i in (9)(10) is obtained
d1015840i =
10038161003816100381610038161003816P1015840
c minus (P1015840c minus d1015840i)10038161003816100381610038161003816 if 10038161003816100381610038161003816Pi minus P10158401015840i
10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 25510038161003816100381610038161003816P1015840
c minus (P1015840c + d1015840i)10038161003816100381610038161003816 otherwise
(10)
Equation (10) can be simplified as
d1015840i =
10038161003816100381610038161003816d1015840
i10038161003816100381610038161003816 if 10038161003816100381610038161003816Pi minus P10158401015840i
10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 25510038161003816100381610038161003816minusd1015840
i10038161003816100381610038161003816 otherwise
(11)
From (3) d1015840i = Li+ Si Substituting d1015840i on right hand side of(11) (12) is obtainedd1015840i
=
Li + Si if 10038161003816100381610038161003816Pi minus P10158401015840i10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 255Li + Si otherwise
(12)
From (12) it can be concluded that d1015840i = Li+ Si or Si= d1015840i minusLi This is nothing but (8) used in extraction procedureThuscorrectness of (8) is justified
3 Examples for Embedding and Extraction
Figure 2 represents an example of embedding for type 1with t=3 of the proposed technique It is self-explanatory To
understand this example it can be traversed from left to rightand top to bottom
Figure 3 represents an example of extraction for type 1with t=3 of the proposed technique It is self-explanatory Tounderstand this example it can be traversed from left to rightand top to bottom
4 Results and Discussion
The embedding and extraction algorithms are simulated inMATLAB Standard RGB color images from SIPI image database are used for testing The R G and B components (each8 bits) of a pixel are treated as a single unit For eight sampleimages the results are discussed and averaged Figure 4 showsthe original images Figure 5 shows the stego images of type 1and Figure 6 shows the stego images of type 2 In each of thesestego images seven lakhs (700000) bits of data is hiddenThe evaluation parameters are (i) hiding capacity (ii) bit rate(BPB) (iii) PSNR and quality index (Q) The hiding capacityis the total number of bits that an image can conceal The bits
4 Advances in Multimedia
Secret data to be embedded 000 100 100 010 001 010 100 110 101 t = 3
192 190 200
192 190 188
192 190 200
Original block
192 190 200
192 184 188
192 190 200
[0 7] 3
[8 15] 3
[16 31] 3
[32 63] 3
[64 127] 4
[128 255] 4
Range table 1
196 188 202
193 184 186
196 190 205
Stego block
Value of Pc is 190
190 = (10111)2
3 bits of data =
(10111)2 = 184
t bit LSB substitution at Pc
After t bit LSB substitution at Pc
Pc is 184
d1 = |184 minus 192| = 8
d2 = |184 minus 190| = 6
d3 = |184 minus 200| = 16
d4 = |184 minus 192| = 8
d5 = |184 minus 188| = 4
d6 = |184 minus 192| = 8
d7 = |184 minus 190| = 6
d8 = |184 minus 200| = 16
t1 = 3
t2 = 3
t3 = 3
t4 = 3
t5 = 3
t6 = 3
t7 = 3
t8 = 3
L1 = 8
L2 = 0
L3 = 16
L4 = 8
L5 = 0
L6 = 8
L7 = 0
L8 = 16
S1 = 4
S2 = 4
S3 = 2
S4 = 1
S5 = 2
S6 = 4
S7 = 6
S8 = 5
d1 = 8 + 4 = 12
d2 = 0 + 4 = 4
d4 = 8 + 1 = 9
d5 = 0 + 2 = 2
d6 = 8 + 4 = 12
d7 = 0 + 6 = 6
d3 = 16 + 2 = 18
d8 = 16 + 5 = 21
P1 = 184 minus 12 = 172
P3 = 184 minus 18 = 166
P6 = 184 minus 12 = 172
P8 = 184 minus 21 = 163
P2 = 184 minus 4 = 180
P4 = 184 minus 9 = 175
P5 = 184 minus 2 = 182
P7 = 184 minus 6 = 178
P1 = 184 + 12 = 196
P3 = 184 + 18 = 202
P6 = 184 + 12 = 196
P8 = 184 + 21 = 205
P2 = 184 + 4 = 188
P4 = 184 + 9 = 193
P5 = 184 + 2 = 186
P7 = 184 + 6 = 190
P1 = 196 P
2 = 188 P
3 = 202
P5 = 186
P8 = 205P
7 = 190
P4 = 193
P6 = 196
Pc = 184
Figure 2 Example of embedding
Advances in Multimedia 5
196 188 202
193 184 186
196 190 205
Stego block
[0 7] 3
[8 15] 3
[16 31] 3
[32 63] 3
[64 127] 4
[128 255] 4
Range table 1
4 4 2 1 2 4 6 5
Extracted data 000 100 100 010 001 010 100 110 101
|184 minus 196| = 12
|184 minus 188| = 4
|184 minus 202| = 18
|184 minus 193| = 9
|184 minus 186| = 2
|184 minus 196| = 12
|184 minus 190| = 6
|184 minus 205| = 21Pc is 184
184 = (10111)2
t1 = 3
t2 = 3
t3 = 3
t4 = 3
t5 = 3
t6 = 3
t7 = 3
t8 = 3
L1 = 8
L2 = 0
L3 = 16
L4 = 8
L5 = 0
L6 = 8
L7 = 0
L8 = 16
S1 = 12 minus 8 = 4
S3 = 18 minus 6 = 2
S6 = 12 minus 8 = 4
S8 = 21 minus 16 = 5
S2 = 4 minus 0 = 4
S4 = 9 minus 8 = 1
S5 = 2 minus 0 = 2
S7 = 6 minus 0 = 6
Si rarr
d1 =
d2 =
d3 =
d4 =
d5 =
d6 =
d7 =
d8 =
Figure 3 Example of extraction
(a) Pot (b) House (c) Boat (d) Jet
(e) Peppers (f) Tiffany (g) Baboon (h) Lena
Figure 4 Original images
per byte (BPB) indicate the average hiding capacity per a byteof the original image Equation (13) is used to calculate BPB
BPB = hiding capacityimage size in bytes (13)
Distortion in stego image is measured in terms of PSNREquation (14) is used to calculate it
PSNR = 10 times log10
255 times 255MSE
(14)
where MSE denotes the mean square error and can becomputed using
MSE = 1m times n
msumi=1
nsumj=1(pij minus qij)
2
(15)
where pij is a pixel of cover image and qij is the correspondingstego pixel
Thequality index (Q) is another parameter tomeasure thestructural similarity between original image and stego image
6 Advances in Multimedia
PSNR = 4115(a)
PSNR = 3969(b)
PSNR = 3725(c)
PSNR = 4221(d)
PSNR = 3781(e)
PSNR = 4146(f)
PSNR = 3426(g)
PSNR = 4262(h)
Figure 5 Stego images of proposed type 1 (t=3 range table 1)
PSNR = 3711(a)
PSNR = 3839(b)
PSNR = 3408(c)
PSNR = 4038(d)
PSNR = 3469(e)
PSNR = 4053(f)
PSNR = 3209(g)
PSNR = 4048(h)
Figure 6 Stego images of proposed type 2 (t=4 range table 2)
Advances in Multimedia 7
Table 3 Results of Pradhan et alrsquos seven-way PVD technique [18]
Images 512times512times3 Seven-way PVD [18] Seven-way+ one-way PVD [18]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4237 1795551 09996 228 4240 1803635 09996 229House 3897 1972223 09986 250 3898 1977403 09986 251Boat 3789 1972086 09988 250 3791 1991005 09988 253Jet 4209 1906254 09989 242 4198 1909595 09989 242Peppers 4042 1778072 09993 226 4020 1806166 09992 229Tiffany 4123 1400756 09986 178 4130 1450799 09986 184Baboon 3379 2226806 09957 283 3377 2243218 09957 285Lena 4173 1896662 09993 241 4173 1901149 09993 241Average 3981 1868551 09986 237 3978 1885371 09985 239
Table 4 Results of Khodaei and Faezrsquos technique
Images 512times512times3 Khodaei and Faez type 1 [10] Khodaei and Faez type 2 [10]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4121 2366001 09995 300 3783 2387494 09990 303House 3916 2387183 09986 303 3875 2470824 09985 314Boat 4005 2391994 09993 304 3749 2504613 09987 318Jet 4089 2374048 09986 301 4064 2418419 09985 307Peppers 3932 2372858 09991 301 3791 2435223 09987 309Tiffany 3973 2372396 09980 301 4025 2416944 09982 307Baboon 3777 2443361 09982 310 3449 2662080 09963 338Lena 4278 2375248 09995 302 4125 2434603 09993 309Average 4011 2385386 09988 303 3857 2466275 09984 313
Equation (16) is used to compute itThemaximum value of Qcan be 1 if original image p and stego image q are the same
Q =4120590xyp q
(1205902x + 1205902y) [(p)2 + (q)2] (16)
where p stands for the mean pixel value of the original imageq stands for the mean pixel value of the stego image1205902x standsfor the standard deviation of pixel values of original image1205902y stands for the standard deviation of pixel values of stegoimage and 120590xy is the covariance between original image andstego image Equations (17) (18) (19) (20) and (21) are usedto calculate these values
p = 1m times n
msumi=1
nsumj=1pij (17)
q = 1m times n
msumi=1
nsumj=1qij (18)
1205902x =1
m times n minus 1msumi=1
nsumj=1(pij minus p)
2
(19)
1205902y =1
m times n minus 1msumi=1
nsumj=1(qij minus q)
2
(20)
120590xy =1
m times n minus 1msumi=1
nsumj=1(pij minus p) (qij minus q) (21)
Table 3 notes the results of seven-way PVD [18] whichis an extension of Wu and Tsairsquos [1] technique Table 4 notes
the results of Khodaei and Faezrsquos technique [10] and Table 5notes the results of the proposed technique The PSNR andcapacity of the proposed technique (both type 1 and type2) are better than seven-way PVD The bit rate and Q arealso better Although as compared to Khodaei and Faezrsquostechnique the PSNR is not improved but the hiding capacityis improved The major advantage over Khodaei and Faezrsquostechnique is that the proposed technique is resistant to PDHanalysis but Khodaei and Faezrsquos technique is detectable byPDH analysis This is made possible by exploiting the edgesin eight directions
Now let us observe the results of proposed technique andcompare type 1 and type 2 We can observe that the capacityis higher in type 2 as compared to type 1 This is because ofthe 4 LSBs substitution in type 2 at the central pixel ie the tvalue is 4 On the other hand the PSNR of type 1 is better thanthat of type 2
Table 6 represents the comparison of the proposedtechnique with the adaptive PVD technique in [26] andhybrid techniques in [27 28] The comparison parametersare average PSNR and average hiding capacity values Thetwo variants of adaptive PVD technique possess lesser hidingcapacity as compared to the proposed eight-directional PVDAs the hiding capacity of adaptive PVD [26] is very littleso PSNR is larger The hybrid technique proposed in [27] isa combination of PVD LSB and EMD The hiding capacityof proposed technique is greater than all the four variantsof the hybrid technique proposed in [27] The technique in[28] uses PVD with modified LSB (MLSB) substitution Theresults of this technique (type 1) are as good as the results of
8 Advances in Multimedia
Table 5 Results of proposed eight-directional PVD technique
Images 512times512times3 Proposed type 1 (t=3 Range Table 1) Proposed type 2 (t=4 Range Table 2)PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4115 2354240 09996 299 3711 2475977 09988 315House 3969 2358575 09988 300 3839 2625804 09984 334Boat 3725 2364685 09987 301 3408 2659795 09973 338Jet 4221 2356828 09989 300 4038 2538801 09985 323Peppers 3781 2356645 09988 300 3469 2544392 09975 324Tiffany 4146 2353505 09987 299 4053 2511139 09984 319Baboon 3426 2392573 09961 304 3209 2939376 09937 374Lena 4262 2353892 09995 299 4048 2533551 09992 322Average 3955 2361368 09986 300 3722 2603604 09977 331
Table 6 Comparison of the proposed technique with techniques in [26ndash28]
Technique PSNR Capacity Amount of hidden data in stego image to compute PSNR2times3 adaptive PVD [26] 5093 1445645 One lakh and forty thousand (140000) bits3times2 adaptive PVD [26] 5093 14356233 PVD +3 LSB +EMD type 1 [27] 4107 1686041
Seven lakhs (700000) bits3 PVD +3 LSB +EMD type 2 [27] 3895 17988347 PVD +3 LSB +EMD type 1 [27] 4028 17108887 PVD +3 LSB +EMD type 2 [27] 3726 1823282Proposed eight-directional PVD type 1 3955 2361368 Seven lakhs (700000) bitsProposed eight-directional PVD type 2 3722 2603604Five-directional PVD+MLSB type 1 [28] 3762 2399853
Eight lakhs and forty thousand (840000) bitsFive-directional PVD+MLSB type 2 [28] 3623 2514038Eight-directional PVD+MLSB type 1 [28] 3749 2394086Eight-directional PVD+MLSB type 2 [28] 3560 2603604
the proposed technique (type 1) Similarly the results of thistechnique (type 2) are as good as the results of the proposedtechnique (type 2)
5 Security Analysis
The LSB substitution techniques suffer from RS analysis andPVD techniques suffer from PDH analysis The proposedtechnique uses the concepts like modified LSB substitutionandPVD so itmust be analyzed by both RS analysis andPDHanalysis
The PDH analysis is performed by calculating the differ-ence value between every pair of pixels A pair comprises twoconsecutive pixels This difference values will be from -255 to+255 including 0 The frequency of each of these differencevalues is counted A graph is plotted with the pixel differencevalue on X-axis and frequency on Y-axis The curve obtainedis called the PDHThePDHof original imagewill be a smoothcurve If the PDH of stego image is also a smooth curve thensteganography is not detected If the PDH of stego imageshows step effects then steganography is identified
Figure 7 is the PDH analysis of type 1 and Figure 8 isthe PDH analysis of type 2 of the proposed technique Theanalysis is carried over all the eight test images For eachimage the PDH of the original image is represented by solidline and the PDH of the stego image is represented by dottedlineThe solid line curves will be obviously free of step effectsas they are of the original images The dotted line curves in
all the sixteen cases do not show any step effects This justifiesthat the proposed technique is resistant to PDH analysis
The RS analysis is based on statistical measures It is donein the following manner Define a function F
1 2nlarrrarr2n+1
It defines two transformations (i) from value 2n to value2n + 1 and (ii) from value 2n + 1 to value 2n Similarlydefine another function F
minus1 2n larrrarr2n -1 It defines other
two transformations (i) from value 2n to value 2n ndash 1 and(ii) from value 2n - 1 to value 2n The image say M isdivided into a number of equal size blocks Suppose such ablock is G whose pixels are X
1X2X3 Xn Then use the
function f(X1X2X3 Xn)=sumnminus1i=1 |Xi+1minusXi| tomeasure the
smoothness of G Then apply F1to all the blocks of M and
define the two parameters Rm and Sm as in (22) and (23)Similarly apply F
minus1to all the blocks of M and define the two
parameters Rminusm and S
minusm as in (24) and (25)
Rm
= No of blocks satisfying the condition f (F1(G)) gt f (G)
Total number of blocks
(22)
Sm
= No of blocks satisfying the condition f (F1(G)) lt f (G)
Total number of blocks
(23)
Rminusm
= No of blocks satisfying the condition f (Fminus1(G)) gt f (G)
Total number of blocks
(24)
Advances in Multimedia 9
0
1
2
Occ
urre
nces
3 times104
Lena coverLena stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3 times104
Baboon coverBaboon stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
Tiffany coverTiffany stego
times104
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Pot coverPot stego
times104
Figure 7 PDH analysis for type 1 of the proposed technique
10 Advances in Multimedia
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Lena coverLena stego
times104
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Baboon coverBaboon stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Tiffany coverTiffany stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Pot coverPot stego
times104
Figure 8 PDH analysis for type 2 of the proposed technique
Advances in Multimedia 11
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(a) Type 1 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(b) Type 1 Baboon
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(c) Type 2 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(d) Type 2 Baboon
Figure 9 RS analysis for type 1 and type 2 of the proposed technique
Sminusm
= No of blocks satisfying the condition f (Fminus1(G)) lt f (G)
Total number of blocks
(25)
RS analysis is performed by using these four parametersIf the condition R
119898asymp Rminus119898gt Sm asymp S
minusm is true then RSanalysis fails to detect the steganography technique But if the
condition Rminus119898
- Sminusm gt Rm - Sm is true then the RS analysis
succeeds in detecting the steganography techniqueFigure 9 represents RS analysis for the proposed tech-
nique Figures 9(a) and 9(b) stand for RS analysis over Lenaand Baboon images respectively of type 1 Similarly Figures9(c) and 9(d) stand for RS analysis over Lena and Baboonimages respectively of type 2 In all the four cases the
12 Advances in Multimedia
condition R119898asymp Rminus119898gt Sm asymp Sminusm is true so we can conclude
that RS analysis could not detect the proposed techniqueLena image has more smooth regions and Baboon imagehas more edge regions all the remaining six images are inbetween these two in terms of smoothness So it is sufficientto draw a conclusion based on these two images
6 Conclusion
The traditional PVD steganography techniques follow a staticrange table Due to this some undesired steps are introducedin pixel difference histograms of the stego images This stepeffect could be avoided by exploiting edges in eight differentdirections This paper proposes a steganography techniqueusing a judicious combination of LSB substitution and eight-directional PVD The LSB substitution is performed only inone pixel out of nine pixels in a block so RS analysis cannotdetect it By using the edges in multiple directions the PDHanalysis cannot detect itThere exist two types of the proposedtechnique with regard to two different quantization tablesThe type 1 provides higher PSNR and the type 2 provideshigher hiding capacity This proposed technique performsbetter in terms of hiding capacity and PSNR over othercompeting PVD techniques
Conflicts of Interest
The author declares that there are no conflicts of interest
References
[1] D-CWu andW-H Tsai ldquoA steganographicmethod for imagesby pixel-value differencingrdquo Pattern Recognition Letters vol 24no 9-10 pp 1613ndash1626 2003
[2] K C Chang C P Chang P S Huang and T M Tu ldquoAnovel image steganographic method using tri-way pixel-valuedifferencingrdquo Journal of Multimedia vol 3 no 2 pp 37ndash442008
[3] Y-P Lee J-C Lee W-K Chen K-C Chang I-J Su and C-P Chang ldquoHigh-payload image hiding with quality recoveryusing tri-way pixel-value differencingrdquo Information Sciencesvol 191 pp 214ndash225 2012
[4] C-C Chang and H-W Tseng ldquoA steganographic method fordigital images using side matchrdquo Pattern Recognition Lettersvol 25 no 12 pp 1431ndash1437 2004
[5] C-H Yang C-Y Weng H-K Tso and S-J Wang ldquoA datahiding scheme using the varieties of pixel-value differencing inmultimedia imagesrdquo The Journal of Systems and Software vol84 no 4 pp 669ndash678 2011
[6] W Hong T-S Chen and C-W Luo ldquoData embedding usingpixel value differencing and diamond encoding with multiple-base notational systemrdquo The Journal of Systems and Softwarevol 85 no 5 pp 1166ndash1175 2012
[7] H C Wu N I Wu C S Tsai and M-S Hwang ldquoImagesteganographic scheme based on pixel-value differencing andLSB replacementmethodsrdquo IEE ProceedingsmdashVision Image andSignal Processing vol 152 no 5 pp 611ndash615 2005
[8] C-H Yang C-Y Weng S-J Wang and H-M Sun ldquoVariedPVD + LSB evading detection programs to spatial domain in
data embedding systemsrdquo The Journal of Systems and Softwarevol 83 no 10 pp 1635ndash1643 2010
[9] J Chen ldquoA PVD-based data hiding method with histogrampreserving using pixel pair matchingrdquo Signal Processing ImageCommunication vol 29 no 3 pp 375ndash384 2014
[10] MKhodaei andK Faez ldquoNew adaptive steganographicmethodusing least-significant-bit substitution and pixel-value differ-encingrdquo IET Image Processing vol 6 no 6 pp 677ndash686 2012
[11] G Swain ldquoA steganographic method combining LSB substitu-tion and PVD in a blockrdquo Procedia Computer Science vol 85pp 39ndash44 2016
[12] X Liao Q Wen and J Zhang ldquoA steganographic methodfor digital images with four-pixel differencing and modifiedLSB substitutionrdquo Journal of Visual Communication and ImageRepresentation vol 22 no 1 pp 1ndash8 2011
[13] G Swain ldquoDigital image steganography using nine-pixel dif-ferencing and modified LSB substitutionrdquo Indian Journal ofScience and Technology vol 7 no 9 pp 1444ndash1450 2014httpwwwindjstorgindexphpindjstarticle
[14] W Luo F Huang and J Huang ldquoA more secure steganographybased on adaptive pixel-value differencing schemerdquoMultimediaTools and Applications vol 52 no 2-3 pp 407ndash430 2011
[15] G Swain ldquoAdaptive pixel value differencing steganographyusing both vertical and horizontal edgesrdquoMultimedia Tools andApplications vol 75 no 21 pp 13541ndash13556 2016
[16] S Chakraborty A S Jalal and C Bhatnagar ldquoLSB based nonblind predictive edge adaptive image steganographyrdquoMultime-dia Tools and Applications vol 76 no 6 pp 7973ndash7987 2017
[17] C Balasubramanian S Selvakumar and S Geetha ldquoHighpayload image steganography with reduced distortion usingoctonary pixel pairing schemerdquoMultimedia Tools and Applica-tions vol 73 no 3 pp 2223ndash2245 2014
[18] A Pradhan K R Sekhar and G Swain ldquoDigital imagesteganography based on seven way pixel value differencingrdquoIndian Journal of Science and Technology vol 9 no 37 pp 1ndash112016
[19] K A Darabkh A K Al-Dhamari and I F Jafar ldquoA newsteganographic algorithm based on multi directional PVD andmodified LSBrdquo Information Technology and Control vol 46 no1 pp 16ndash36 2017
[20] A K Gulve and M S Joshi ldquoA high capacity secured imagesteganography method with five pixel pair differencing andLSB substitutionrdquo International Journal of Image Graphics andSignal Processing vol 7 no 5 pp 66ndash74 2015
[21] X Zhang and S Wang ldquoEfficient steganographic embeddingby exploiting modification directionrdquo IEEE CommunicationsLetters vol 10 no 11 pp 781ndash783 2006
[22] C Kim ldquoData hiding by an improved exploiting modificationdirectionrdquoMultimedia Tools and Applications vol 69 no 3 pp569ndash584 2014
[23] S Y Shen and L H Huang ldquoA data hiding scheme usingpixel value differencing and improving exploiting modificationdirectionsrdquo Computers amp Security vol 48 pp 131ndash141 2015
[24] T D Nguyen S Arch-Int and N Arch-Int ldquoAn adaptivemulti bit-plane image steganography using block data-hidingrdquoMultimediaTools andApplications vol 75 no 14 pp 8319ndash83452016
[25] A Soria-Lorente and S Berres ldquoA secure steganographicalgorithm based on frequency domain for the transmission ofhidden informationrdquo Security and Communication Networksvol 2017 Article ID 5397028 14 pages 2017
Advances in Multimedia 13
[26] A Pradhan K R Sekhar and G Swain ldquoAdaptive PVDsteganography using horizontal vertical and diagonal edges insix-pixel blocksrdquo Security and Communication Networks vol2017 Article ID 1924618 13 pages 2017
[27] A Pradhan K R ekhar andG Swain ldquoDigital image steganog-raphy using LSB substitution PVD and EMDrdquo MathematicalProblems in Engineering 2018
[28] G Swain ldquoHigh capacity image steganography using modifiedLSB substitution and PVD against pixel difference histogramanalysisrdquo Security and Communication Networks 2018
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Advances in Multimedia 3
The difference value d1belongs to one of the six ranges
Suppose the lower bound is denoted as L1and hiding capacity
as t1for the range of d
1 Similarly L
2 L3 L4 L5 L6 L7 and L
8
are the lower bounds and t2 t3 t4 t5 t6 t7 and t
8are the
hiding capacities of the ranges of d2 d3 d4 d5 d6 d7 and d
8
respectivelyNow take t
1 t2 t3 t4 t5 t6 t7 and t
8bits of data separately
and convert to their decimal values S1 S2 S3 S4 S5 S6 S7 and
S8 respectivelyNow for i=1 to 8 calculate new difference values d1015840i using
d1015840i = Li + Si (3)
For i = 1 to 8 a pair of new values for each Pi is calculated asin
P10158401015840i = P1015840c minus d1015840i (4)
P101584010158401015840i = P1015840c + d1015840i (5)
Now out of two new values P10158401015840i and P101584010158401015840i one value is chosenfor P1015840i using
P1015840i
=
P10158401015840i if 10038161003816100381610038161003816Pi minus P10158401015840i10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 255P101584010158401015840i otherwise
(6)
Thus the stego pixel block after hiding t t1 t2 t3 t4 t5 t6 t7
and t8bits of data in pixels Pc P1 P2 P3 P4 P5 P6 P7 and
P8 respectively is as shown in Figure 1(b)
The extraction algorithm operates on 3times3 pixel blocksas in embedding algorithm The blocks are formed in anonoverlapping manner A typical stego block is representedin Figure 1(b) Retrieve the t LSBs from P1015840c For i=1 to 8 findeight difference values as in
d1015840i =10038161003816100381610038161003816P1015840
c minus P1015840i10038161003816100381610038161003816 (7)
Each d1015840i belongs to a rangewhose embedding capacity is ti andlower bound is Li For i=1 to 8 find the decimal values of theembedded secret bit streams using
Si = d1015840i minus Li (8)
Now convert S1 S2 S3 S4 S5 S6 S7 and S
8into t1 t2 t3 t4
t5 t6 t7 and t
8binary bits respectively
The correctness of the extraction procedure is justifiedby the following mathematical discussions That means how(8) is derived from (7) is described below
As per (7) d1015840i = |P1015840c minus P1015840i |From (6) P1015840i is P
10158401015840
i if |Pi minus P10158401015840i | lt |Pi minus P101584010158401015840i | and 0 le P10158401015840i le255 Otherwise P1015840i is P101584010158401015840i So d1015840i can be defined as in
d1015840i
=
10038161003816100381610038161003816P1015840
c minus P10158401015840i10038161003816100381610038161003816 if 10038161003816100381610038161003816Pi minus P10158401015840i
10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 25510038161003816100381610038161003816P1015840
c minus P101584010158401015840i10038161003816100381610038161003816 otherwise
(9)
From (4) and (5) P10158401015840i = (P1015840c - d1015840i ) and P101584010158401015840i = (P1015840c + d1015840i )respectively Substituting these values of P10158401015840i and P101584010158401015840i in (9)(10) is obtained
d1015840i =
10038161003816100381610038161003816P1015840
c minus (P1015840c minus d1015840i)10038161003816100381610038161003816 if 10038161003816100381610038161003816Pi minus P10158401015840i
10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 25510038161003816100381610038161003816P1015840
c minus (P1015840c + d1015840i)10038161003816100381610038161003816 otherwise
(10)
Equation (10) can be simplified as
d1015840i =
10038161003816100381610038161003816d1015840
i10038161003816100381610038161003816 if 10038161003816100381610038161003816Pi minus P10158401015840i
10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 25510038161003816100381610038161003816minusd1015840
i10038161003816100381610038161003816 otherwise
(11)
From (3) d1015840i = Li+ Si Substituting d1015840i on right hand side of(11) (12) is obtainedd1015840i
=
Li + Si if 10038161003816100381610038161003816Pi minus P10158401015840i10038161003816100381610038161003816 lt10038161003816100381610038161003816Pi minus P101584010158401015840i
10038161003816100381610038161003816 and 0 le P10158401015840
i le 255Li + Si otherwise
(12)
From (12) it can be concluded that d1015840i = Li+ Si or Si= d1015840i minusLi This is nothing but (8) used in extraction procedureThuscorrectness of (8) is justified
3 Examples for Embedding and Extraction
Figure 2 represents an example of embedding for type 1with t=3 of the proposed technique It is self-explanatory To
understand this example it can be traversed from left to rightand top to bottom
Figure 3 represents an example of extraction for type 1with t=3 of the proposed technique It is self-explanatory Tounderstand this example it can be traversed from left to rightand top to bottom
4 Results and Discussion
The embedding and extraction algorithms are simulated inMATLAB Standard RGB color images from SIPI image database are used for testing The R G and B components (each8 bits) of a pixel are treated as a single unit For eight sampleimages the results are discussed and averaged Figure 4 showsthe original images Figure 5 shows the stego images of type 1and Figure 6 shows the stego images of type 2 In each of thesestego images seven lakhs (700000) bits of data is hiddenThe evaluation parameters are (i) hiding capacity (ii) bit rate(BPB) (iii) PSNR and quality index (Q) The hiding capacityis the total number of bits that an image can conceal The bits
4 Advances in Multimedia
Secret data to be embedded 000 100 100 010 001 010 100 110 101 t = 3
192 190 200
192 190 188
192 190 200
Original block
192 190 200
192 184 188
192 190 200
[0 7] 3
[8 15] 3
[16 31] 3
[32 63] 3
[64 127] 4
[128 255] 4
Range table 1
196 188 202
193 184 186
196 190 205
Stego block
Value of Pc is 190
190 = (10111)2
3 bits of data =
(10111)2 = 184
t bit LSB substitution at Pc
After t bit LSB substitution at Pc
Pc is 184
d1 = |184 minus 192| = 8
d2 = |184 minus 190| = 6
d3 = |184 minus 200| = 16
d4 = |184 minus 192| = 8
d5 = |184 minus 188| = 4
d6 = |184 minus 192| = 8
d7 = |184 minus 190| = 6
d8 = |184 minus 200| = 16
t1 = 3
t2 = 3
t3 = 3
t4 = 3
t5 = 3
t6 = 3
t7 = 3
t8 = 3
L1 = 8
L2 = 0
L3 = 16
L4 = 8
L5 = 0
L6 = 8
L7 = 0
L8 = 16
S1 = 4
S2 = 4
S3 = 2
S4 = 1
S5 = 2
S6 = 4
S7 = 6
S8 = 5
d1 = 8 + 4 = 12
d2 = 0 + 4 = 4
d4 = 8 + 1 = 9
d5 = 0 + 2 = 2
d6 = 8 + 4 = 12
d7 = 0 + 6 = 6
d3 = 16 + 2 = 18
d8 = 16 + 5 = 21
P1 = 184 minus 12 = 172
P3 = 184 minus 18 = 166
P6 = 184 minus 12 = 172
P8 = 184 minus 21 = 163
P2 = 184 minus 4 = 180
P4 = 184 minus 9 = 175
P5 = 184 minus 2 = 182
P7 = 184 minus 6 = 178
P1 = 184 + 12 = 196
P3 = 184 + 18 = 202
P6 = 184 + 12 = 196
P8 = 184 + 21 = 205
P2 = 184 + 4 = 188
P4 = 184 + 9 = 193
P5 = 184 + 2 = 186
P7 = 184 + 6 = 190
P1 = 196 P
2 = 188 P
3 = 202
P5 = 186
P8 = 205P
7 = 190
P4 = 193
P6 = 196
Pc = 184
Figure 2 Example of embedding
Advances in Multimedia 5
196 188 202
193 184 186
196 190 205
Stego block
[0 7] 3
[8 15] 3
[16 31] 3
[32 63] 3
[64 127] 4
[128 255] 4
Range table 1
4 4 2 1 2 4 6 5
Extracted data 000 100 100 010 001 010 100 110 101
|184 minus 196| = 12
|184 minus 188| = 4
|184 minus 202| = 18
|184 minus 193| = 9
|184 minus 186| = 2
|184 minus 196| = 12
|184 minus 190| = 6
|184 minus 205| = 21Pc is 184
184 = (10111)2
t1 = 3
t2 = 3
t3 = 3
t4 = 3
t5 = 3
t6 = 3
t7 = 3
t8 = 3
L1 = 8
L2 = 0
L3 = 16
L4 = 8
L5 = 0
L6 = 8
L7 = 0
L8 = 16
S1 = 12 minus 8 = 4
S3 = 18 minus 6 = 2
S6 = 12 minus 8 = 4
S8 = 21 minus 16 = 5
S2 = 4 minus 0 = 4
S4 = 9 minus 8 = 1
S5 = 2 minus 0 = 2
S7 = 6 minus 0 = 6
Si rarr
d1 =
d2 =
d3 =
d4 =
d5 =
d6 =
d7 =
d8 =
Figure 3 Example of extraction
(a) Pot (b) House (c) Boat (d) Jet
(e) Peppers (f) Tiffany (g) Baboon (h) Lena
Figure 4 Original images
per byte (BPB) indicate the average hiding capacity per a byteof the original image Equation (13) is used to calculate BPB
BPB = hiding capacityimage size in bytes (13)
Distortion in stego image is measured in terms of PSNREquation (14) is used to calculate it
PSNR = 10 times log10
255 times 255MSE
(14)
where MSE denotes the mean square error and can becomputed using
MSE = 1m times n
msumi=1
nsumj=1(pij minus qij)
2
(15)
where pij is a pixel of cover image and qij is the correspondingstego pixel
Thequality index (Q) is another parameter tomeasure thestructural similarity between original image and stego image
6 Advances in Multimedia
PSNR = 4115(a)
PSNR = 3969(b)
PSNR = 3725(c)
PSNR = 4221(d)
PSNR = 3781(e)
PSNR = 4146(f)
PSNR = 3426(g)
PSNR = 4262(h)
Figure 5 Stego images of proposed type 1 (t=3 range table 1)
PSNR = 3711(a)
PSNR = 3839(b)
PSNR = 3408(c)
PSNR = 4038(d)
PSNR = 3469(e)
PSNR = 4053(f)
PSNR = 3209(g)
PSNR = 4048(h)
Figure 6 Stego images of proposed type 2 (t=4 range table 2)
Advances in Multimedia 7
Table 3 Results of Pradhan et alrsquos seven-way PVD technique [18]
Images 512times512times3 Seven-way PVD [18] Seven-way+ one-way PVD [18]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4237 1795551 09996 228 4240 1803635 09996 229House 3897 1972223 09986 250 3898 1977403 09986 251Boat 3789 1972086 09988 250 3791 1991005 09988 253Jet 4209 1906254 09989 242 4198 1909595 09989 242Peppers 4042 1778072 09993 226 4020 1806166 09992 229Tiffany 4123 1400756 09986 178 4130 1450799 09986 184Baboon 3379 2226806 09957 283 3377 2243218 09957 285Lena 4173 1896662 09993 241 4173 1901149 09993 241Average 3981 1868551 09986 237 3978 1885371 09985 239
Table 4 Results of Khodaei and Faezrsquos technique
Images 512times512times3 Khodaei and Faez type 1 [10] Khodaei and Faez type 2 [10]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4121 2366001 09995 300 3783 2387494 09990 303House 3916 2387183 09986 303 3875 2470824 09985 314Boat 4005 2391994 09993 304 3749 2504613 09987 318Jet 4089 2374048 09986 301 4064 2418419 09985 307Peppers 3932 2372858 09991 301 3791 2435223 09987 309Tiffany 3973 2372396 09980 301 4025 2416944 09982 307Baboon 3777 2443361 09982 310 3449 2662080 09963 338Lena 4278 2375248 09995 302 4125 2434603 09993 309Average 4011 2385386 09988 303 3857 2466275 09984 313
Equation (16) is used to compute itThemaximum value of Qcan be 1 if original image p and stego image q are the same
Q =4120590xyp q
(1205902x + 1205902y) [(p)2 + (q)2] (16)
where p stands for the mean pixel value of the original imageq stands for the mean pixel value of the stego image1205902x standsfor the standard deviation of pixel values of original image1205902y stands for the standard deviation of pixel values of stegoimage and 120590xy is the covariance between original image andstego image Equations (17) (18) (19) (20) and (21) are usedto calculate these values
p = 1m times n
msumi=1
nsumj=1pij (17)
q = 1m times n
msumi=1
nsumj=1qij (18)
1205902x =1
m times n minus 1msumi=1
nsumj=1(pij minus p)
2
(19)
1205902y =1
m times n minus 1msumi=1
nsumj=1(qij minus q)
2
(20)
120590xy =1
m times n minus 1msumi=1
nsumj=1(pij minus p) (qij minus q) (21)
Table 3 notes the results of seven-way PVD [18] whichis an extension of Wu and Tsairsquos [1] technique Table 4 notes
the results of Khodaei and Faezrsquos technique [10] and Table 5notes the results of the proposed technique The PSNR andcapacity of the proposed technique (both type 1 and type2) are better than seven-way PVD The bit rate and Q arealso better Although as compared to Khodaei and Faezrsquostechnique the PSNR is not improved but the hiding capacityis improved The major advantage over Khodaei and Faezrsquostechnique is that the proposed technique is resistant to PDHanalysis but Khodaei and Faezrsquos technique is detectable byPDH analysis This is made possible by exploiting the edgesin eight directions
Now let us observe the results of proposed technique andcompare type 1 and type 2 We can observe that the capacityis higher in type 2 as compared to type 1 This is because ofthe 4 LSBs substitution in type 2 at the central pixel ie the tvalue is 4 On the other hand the PSNR of type 1 is better thanthat of type 2
Table 6 represents the comparison of the proposedtechnique with the adaptive PVD technique in [26] andhybrid techniques in [27 28] The comparison parametersare average PSNR and average hiding capacity values Thetwo variants of adaptive PVD technique possess lesser hidingcapacity as compared to the proposed eight-directional PVDAs the hiding capacity of adaptive PVD [26] is very littleso PSNR is larger The hybrid technique proposed in [27] isa combination of PVD LSB and EMD The hiding capacityof proposed technique is greater than all the four variantsof the hybrid technique proposed in [27] The technique in[28] uses PVD with modified LSB (MLSB) substitution Theresults of this technique (type 1) are as good as the results of
8 Advances in Multimedia
Table 5 Results of proposed eight-directional PVD technique
Images 512times512times3 Proposed type 1 (t=3 Range Table 1) Proposed type 2 (t=4 Range Table 2)PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4115 2354240 09996 299 3711 2475977 09988 315House 3969 2358575 09988 300 3839 2625804 09984 334Boat 3725 2364685 09987 301 3408 2659795 09973 338Jet 4221 2356828 09989 300 4038 2538801 09985 323Peppers 3781 2356645 09988 300 3469 2544392 09975 324Tiffany 4146 2353505 09987 299 4053 2511139 09984 319Baboon 3426 2392573 09961 304 3209 2939376 09937 374Lena 4262 2353892 09995 299 4048 2533551 09992 322Average 3955 2361368 09986 300 3722 2603604 09977 331
Table 6 Comparison of the proposed technique with techniques in [26ndash28]
Technique PSNR Capacity Amount of hidden data in stego image to compute PSNR2times3 adaptive PVD [26] 5093 1445645 One lakh and forty thousand (140000) bits3times2 adaptive PVD [26] 5093 14356233 PVD +3 LSB +EMD type 1 [27] 4107 1686041
Seven lakhs (700000) bits3 PVD +3 LSB +EMD type 2 [27] 3895 17988347 PVD +3 LSB +EMD type 1 [27] 4028 17108887 PVD +3 LSB +EMD type 2 [27] 3726 1823282Proposed eight-directional PVD type 1 3955 2361368 Seven lakhs (700000) bitsProposed eight-directional PVD type 2 3722 2603604Five-directional PVD+MLSB type 1 [28] 3762 2399853
Eight lakhs and forty thousand (840000) bitsFive-directional PVD+MLSB type 2 [28] 3623 2514038Eight-directional PVD+MLSB type 1 [28] 3749 2394086Eight-directional PVD+MLSB type 2 [28] 3560 2603604
the proposed technique (type 1) Similarly the results of thistechnique (type 2) are as good as the results of the proposedtechnique (type 2)
5 Security Analysis
The LSB substitution techniques suffer from RS analysis andPVD techniques suffer from PDH analysis The proposedtechnique uses the concepts like modified LSB substitutionandPVD so itmust be analyzed by both RS analysis andPDHanalysis
The PDH analysis is performed by calculating the differ-ence value between every pair of pixels A pair comprises twoconsecutive pixels This difference values will be from -255 to+255 including 0 The frequency of each of these differencevalues is counted A graph is plotted with the pixel differencevalue on X-axis and frequency on Y-axis The curve obtainedis called the PDHThePDHof original imagewill be a smoothcurve If the PDH of stego image is also a smooth curve thensteganography is not detected If the PDH of stego imageshows step effects then steganography is identified
Figure 7 is the PDH analysis of type 1 and Figure 8 isthe PDH analysis of type 2 of the proposed technique Theanalysis is carried over all the eight test images For eachimage the PDH of the original image is represented by solidline and the PDH of the stego image is represented by dottedlineThe solid line curves will be obviously free of step effectsas they are of the original images The dotted line curves in
all the sixteen cases do not show any step effects This justifiesthat the proposed technique is resistant to PDH analysis
The RS analysis is based on statistical measures It is donein the following manner Define a function F
1 2nlarrrarr2n+1
It defines two transformations (i) from value 2n to value2n + 1 and (ii) from value 2n + 1 to value 2n Similarlydefine another function F
minus1 2n larrrarr2n -1 It defines other
two transformations (i) from value 2n to value 2n ndash 1 and(ii) from value 2n - 1 to value 2n The image say M isdivided into a number of equal size blocks Suppose such ablock is G whose pixels are X
1X2X3 Xn Then use the
function f(X1X2X3 Xn)=sumnminus1i=1 |Xi+1minusXi| tomeasure the
smoothness of G Then apply F1to all the blocks of M and
define the two parameters Rm and Sm as in (22) and (23)Similarly apply F
minus1to all the blocks of M and define the two
parameters Rminusm and S
minusm as in (24) and (25)
Rm
= No of blocks satisfying the condition f (F1(G)) gt f (G)
Total number of blocks
(22)
Sm
= No of blocks satisfying the condition f (F1(G)) lt f (G)
Total number of blocks
(23)
Rminusm
= No of blocks satisfying the condition f (Fminus1(G)) gt f (G)
Total number of blocks
(24)
Advances in Multimedia 9
0
1
2
Occ
urre
nces
3 times104
Lena coverLena stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3 times104
Baboon coverBaboon stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
Tiffany coverTiffany stego
times104
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Pot coverPot stego
times104
Figure 7 PDH analysis for type 1 of the proposed technique
10 Advances in Multimedia
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Lena coverLena stego
times104
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Baboon coverBaboon stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Tiffany coverTiffany stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Pot coverPot stego
times104
Figure 8 PDH analysis for type 2 of the proposed technique
Advances in Multimedia 11
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(a) Type 1 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(b) Type 1 Baboon
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(c) Type 2 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(d) Type 2 Baboon
Figure 9 RS analysis for type 1 and type 2 of the proposed technique
Sminusm
= No of blocks satisfying the condition f (Fminus1(G)) lt f (G)
Total number of blocks
(25)
RS analysis is performed by using these four parametersIf the condition R
119898asymp Rminus119898gt Sm asymp S
minusm is true then RSanalysis fails to detect the steganography technique But if the
condition Rminus119898
- Sminusm gt Rm - Sm is true then the RS analysis
succeeds in detecting the steganography techniqueFigure 9 represents RS analysis for the proposed tech-
nique Figures 9(a) and 9(b) stand for RS analysis over Lenaand Baboon images respectively of type 1 Similarly Figures9(c) and 9(d) stand for RS analysis over Lena and Baboonimages respectively of type 2 In all the four cases the
12 Advances in Multimedia
condition R119898asymp Rminus119898gt Sm asymp Sminusm is true so we can conclude
that RS analysis could not detect the proposed techniqueLena image has more smooth regions and Baboon imagehas more edge regions all the remaining six images are inbetween these two in terms of smoothness So it is sufficientto draw a conclusion based on these two images
6 Conclusion
The traditional PVD steganography techniques follow a staticrange table Due to this some undesired steps are introducedin pixel difference histograms of the stego images This stepeffect could be avoided by exploiting edges in eight differentdirections This paper proposes a steganography techniqueusing a judicious combination of LSB substitution and eight-directional PVD The LSB substitution is performed only inone pixel out of nine pixels in a block so RS analysis cannotdetect it By using the edges in multiple directions the PDHanalysis cannot detect itThere exist two types of the proposedtechnique with regard to two different quantization tablesThe type 1 provides higher PSNR and the type 2 provideshigher hiding capacity This proposed technique performsbetter in terms of hiding capacity and PSNR over othercompeting PVD techniques
Conflicts of Interest
The author declares that there are no conflicts of interest
References
[1] D-CWu andW-H Tsai ldquoA steganographicmethod for imagesby pixel-value differencingrdquo Pattern Recognition Letters vol 24no 9-10 pp 1613ndash1626 2003
[2] K C Chang C P Chang P S Huang and T M Tu ldquoAnovel image steganographic method using tri-way pixel-valuedifferencingrdquo Journal of Multimedia vol 3 no 2 pp 37ndash442008
[3] Y-P Lee J-C Lee W-K Chen K-C Chang I-J Su and C-P Chang ldquoHigh-payload image hiding with quality recoveryusing tri-way pixel-value differencingrdquo Information Sciencesvol 191 pp 214ndash225 2012
[4] C-C Chang and H-W Tseng ldquoA steganographic method fordigital images using side matchrdquo Pattern Recognition Lettersvol 25 no 12 pp 1431ndash1437 2004
[5] C-H Yang C-Y Weng H-K Tso and S-J Wang ldquoA datahiding scheme using the varieties of pixel-value differencing inmultimedia imagesrdquo The Journal of Systems and Software vol84 no 4 pp 669ndash678 2011
[6] W Hong T-S Chen and C-W Luo ldquoData embedding usingpixel value differencing and diamond encoding with multiple-base notational systemrdquo The Journal of Systems and Softwarevol 85 no 5 pp 1166ndash1175 2012
[7] H C Wu N I Wu C S Tsai and M-S Hwang ldquoImagesteganographic scheme based on pixel-value differencing andLSB replacementmethodsrdquo IEE ProceedingsmdashVision Image andSignal Processing vol 152 no 5 pp 611ndash615 2005
[8] C-H Yang C-Y Weng S-J Wang and H-M Sun ldquoVariedPVD + LSB evading detection programs to spatial domain in
data embedding systemsrdquo The Journal of Systems and Softwarevol 83 no 10 pp 1635ndash1643 2010
[9] J Chen ldquoA PVD-based data hiding method with histogrampreserving using pixel pair matchingrdquo Signal Processing ImageCommunication vol 29 no 3 pp 375ndash384 2014
[10] MKhodaei andK Faez ldquoNew adaptive steganographicmethodusing least-significant-bit substitution and pixel-value differ-encingrdquo IET Image Processing vol 6 no 6 pp 677ndash686 2012
[11] G Swain ldquoA steganographic method combining LSB substitu-tion and PVD in a blockrdquo Procedia Computer Science vol 85pp 39ndash44 2016
[12] X Liao Q Wen and J Zhang ldquoA steganographic methodfor digital images with four-pixel differencing and modifiedLSB substitutionrdquo Journal of Visual Communication and ImageRepresentation vol 22 no 1 pp 1ndash8 2011
[13] G Swain ldquoDigital image steganography using nine-pixel dif-ferencing and modified LSB substitutionrdquo Indian Journal ofScience and Technology vol 7 no 9 pp 1444ndash1450 2014httpwwwindjstorgindexphpindjstarticle
[14] W Luo F Huang and J Huang ldquoA more secure steganographybased on adaptive pixel-value differencing schemerdquoMultimediaTools and Applications vol 52 no 2-3 pp 407ndash430 2011
[15] G Swain ldquoAdaptive pixel value differencing steganographyusing both vertical and horizontal edgesrdquoMultimedia Tools andApplications vol 75 no 21 pp 13541ndash13556 2016
[16] S Chakraborty A S Jalal and C Bhatnagar ldquoLSB based nonblind predictive edge adaptive image steganographyrdquoMultime-dia Tools and Applications vol 76 no 6 pp 7973ndash7987 2017
[17] C Balasubramanian S Selvakumar and S Geetha ldquoHighpayload image steganography with reduced distortion usingoctonary pixel pairing schemerdquoMultimedia Tools and Applica-tions vol 73 no 3 pp 2223ndash2245 2014
[18] A Pradhan K R Sekhar and G Swain ldquoDigital imagesteganography based on seven way pixel value differencingrdquoIndian Journal of Science and Technology vol 9 no 37 pp 1ndash112016
[19] K A Darabkh A K Al-Dhamari and I F Jafar ldquoA newsteganographic algorithm based on multi directional PVD andmodified LSBrdquo Information Technology and Control vol 46 no1 pp 16ndash36 2017
[20] A K Gulve and M S Joshi ldquoA high capacity secured imagesteganography method with five pixel pair differencing andLSB substitutionrdquo International Journal of Image Graphics andSignal Processing vol 7 no 5 pp 66ndash74 2015
[21] X Zhang and S Wang ldquoEfficient steganographic embeddingby exploiting modification directionrdquo IEEE CommunicationsLetters vol 10 no 11 pp 781ndash783 2006
[22] C Kim ldquoData hiding by an improved exploiting modificationdirectionrdquoMultimedia Tools and Applications vol 69 no 3 pp569ndash584 2014
[23] S Y Shen and L H Huang ldquoA data hiding scheme usingpixel value differencing and improving exploiting modificationdirectionsrdquo Computers amp Security vol 48 pp 131ndash141 2015
[24] T D Nguyen S Arch-Int and N Arch-Int ldquoAn adaptivemulti bit-plane image steganography using block data-hidingrdquoMultimediaTools andApplications vol 75 no 14 pp 8319ndash83452016
[25] A Soria-Lorente and S Berres ldquoA secure steganographicalgorithm based on frequency domain for the transmission ofhidden informationrdquo Security and Communication Networksvol 2017 Article ID 5397028 14 pages 2017
Advances in Multimedia 13
[26] A Pradhan K R Sekhar and G Swain ldquoAdaptive PVDsteganography using horizontal vertical and diagonal edges insix-pixel blocksrdquo Security and Communication Networks vol2017 Article ID 1924618 13 pages 2017
[27] A Pradhan K R ekhar andG Swain ldquoDigital image steganog-raphy using LSB substitution PVD and EMDrdquo MathematicalProblems in Engineering 2018
[28] G Swain ldquoHigh capacity image steganography using modifiedLSB substitution and PVD against pixel difference histogramanalysisrdquo Security and Communication Networks 2018
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
4 Advances in Multimedia
Secret data to be embedded 000 100 100 010 001 010 100 110 101 t = 3
192 190 200
192 190 188
192 190 200
Original block
192 190 200
192 184 188
192 190 200
[0 7] 3
[8 15] 3
[16 31] 3
[32 63] 3
[64 127] 4
[128 255] 4
Range table 1
196 188 202
193 184 186
196 190 205
Stego block
Value of Pc is 190
190 = (10111)2
3 bits of data =
(10111)2 = 184
t bit LSB substitution at Pc
After t bit LSB substitution at Pc
Pc is 184
d1 = |184 minus 192| = 8
d2 = |184 minus 190| = 6
d3 = |184 minus 200| = 16
d4 = |184 minus 192| = 8
d5 = |184 minus 188| = 4
d6 = |184 minus 192| = 8
d7 = |184 minus 190| = 6
d8 = |184 minus 200| = 16
t1 = 3
t2 = 3
t3 = 3
t4 = 3
t5 = 3
t6 = 3
t7 = 3
t8 = 3
L1 = 8
L2 = 0
L3 = 16
L4 = 8
L5 = 0
L6 = 8
L7 = 0
L8 = 16
S1 = 4
S2 = 4
S3 = 2
S4 = 1
S5 = 2
S6 = 4
S7 = 6
S8 = 5
d1 = 8 + 4 = 12
d2 = 0 + 4 = 4
d4 = 8 + 1 = 9
d5 = 0 + 2 = 2
d6 = 8 + 4 = 12
d7 = 0 + 6 = 6
d3 = 16 + 2 = 18
d8 = 16 + 5 = 21
P1 = 184 minus 12 = 172
P3 = 184 minus 18 = 166
P6 = 184 minus 12 = 172
P8 = 184 minus 21 = 163
P2 = 184 minus 4 = 180
P4 = 184 minus 9 = 175
P5 = 184 minus 2 = 182
P7 = 184 minus 6 = 178
P1 = 184 + 12 = 196
P3 = 184 + 18 = 202
P6 = 184 + 12 = 196
P8 = 184 + 21 = 205
P2 = 184 + 4 = 188
P4 = 184 + 9 = 193
P5 = 184 + 2 = 186
P7 = 184 + 6 = 190
P1 = 196 P
2 = 188 P
3 = 202
P5 = 186
P8 = 205P
7 = 190
P4 = 193
P6 = 196
Pc = 184
Figure 2 Example of embedding
Advances in Multimedia 5
196 188 202
193 184 186
196 190 205
Stego block
[0 7] 3
[8 15] 3
[16 31] 3
[32 63] 3
[64 127] 4
[128 255] 4
Range table 1
4 4 2 1 2 4 6 5
Extracted data 000 100 100 010 001 010 100 110 101
|184 minus 196| = 12
|184 minus 188| = 4
|184 minus 202| = 18
|184 minus 193| = 9
|184 minus 186| = 2
|184 minus 196| = 12
|184 minus 190| = 6
|184 minus 205| = 21Pc is 184
184 = (10111)2
t1 = 3
t2 = 3
t3 = 3
t4 = 3
t5 = 3
t6 = 3
t7 = 3
t8 = 3
L1 = 8
L2 = 0
L3 = 16
L4 = 8
L5 = 0
L6 = 8
L7 = 0
L8 = 16
S1 = 12 minus 8 = 4
S3 = 18 minus 6 = 2
S6 = 12 minus 8 = 4
S8 = 21 minus 16 = 5
S2 = 4 minus 0 = 4
S4 = 9 minus 8 = 1
S5 = 2 minus 0 = 2
S7 = 6 minus 0 = 6
Si rarr
d1 =
d2 =
d3 =
d4 =
d5 =
d6 =
d7 =
d8 =
Figure 3 Example of extraction
(a) Pot (b) House (c) Boat (d) Jet
(e) Peppers (f) Tiffany (g) Baboon (h) Lena
Figure 4 Original images
per byte (BPB) indicate the average hiding capacity per a byteof the original image Equation (13) is used to calculate BPB
BPB = hiding capacityimage size in bytes (13)
Distortion in stego image is measured in terms of PSNREquation (14) is used to calculate it
PSNR = 10 times log10
255 times 255MSE
(14)
where MSE denotes the mean square error and can becomputed using
MSE = 1m times n
msumi=1
nsumj=1(pij minus qij)
2
(15)
where pij is a pixel of cover image and qij is the correspondingstego pixel
Thequality index (Q) is another parameter tomeasure thestructural similarity between original image and stego image
6 Advances in Multimedia
PSNR = 4115(a)
PSNR = 3969(b)
PSNR = 3725(c)
PSNR = 4221(d)
PSNR = 3781(e)
PSNR = 4146(f)
PSNR = 3426(g)
PSNR = 4262(h)
Figure 5 Stego images of proposed type 1 (t=3 range table 1)
PSNR = 3711(a)
PSNR = 3839(b)
PSNR = 3408(c)
PSNR = 4038(d)
PSNR = 3469(e)
PSNR = 4053(f)
PSNR = 3209(g)
PSNR = 4048(h)
Figure 6 Stego images of proposed type 2 (t=4 range table 2)
Advances in Multimedia 7
Table 3 Results of Pradhan et alrsquos seven-way PVD technique [18]
Images 512times512times3 Seven-way PVD [18] Seven-way+ one-way PVD [18]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4237 1795551 09996 228 4240 1803635 09996 229House 3897 1972223 09986 250 3898 1977403 09986 251Boat 3789 1972086 09988 250 3791 1991005 09988 253Jet 4209 1906254 09989 242 4198 1909595 09989 242Peppers 4042 1778072 09993 226 4020 1806166 09992 229Tiffany 4123 1400756 09986 178 4130 1450799 09986 184Baboon 3379 2226806 09957 283 3377 2243218 09957 285Lena 4173 1896662 09993 241 4173 1901149 09993 241Average 3981 1868551 09986 237 3978 1885371 09985 239
Table 4 Results of Khodaei and Faezrsquos technique
Images 512times512times3 Khodaei and Faez type 1 [10] Khodaei and Faez type 2 [10]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4121 2366001 09995 300 3783 2387494 09990 303House 3916 2387183 09986 303 3875 2470824 09985 314Boat 4005 2391994 09993 304 3749 2504613 09987 318Jet 4089 2374048 09986 301 4064 2418419 09985 307Peppers 3932 2372858 09991 301 3791 2435223 09987 309Tiffany 3973 2372396 09980 301 4025 2416944 09982 307Baboon 3777 2443361 09982 310 3449 2662080 09963 338Lena 4278 2375248 09995 302 4125 2434603 09993 309Average 4011 2385386 09988 303 3857 2466275 09984 313
Equation (16) is used to compute itThemaximum value of Qcan be 1 if original image p and stego image q are the same
Q =4120590xyp q
(1205902x + 1205902y) [(p)2 + (q)2] (16)
where p stands for the mean pixel value of the original imageq stands for the mean pixel value of the stego image1205902x standsfor the standard deviation of pixel values of original image1205902y stands for the standard deviation of pixel values of stegoimage and 120590xy is the covariance between original image andstego image Equations (17) (18) (19) (20) and (21) are usedto calculate these values
p = 1m times n
msumi=1
nsumj=1pij (17)
q = 1m times n
msumi=1
nsumj=1qij (18)
1205902x =1
m times n minus 1msumi=1
nsumj=1(pij minus p)
2
(19)
1205902y =1
m times n minus 1msumi=1
nsumj=1(qij minus q)
2
(20)
120590xy =1
m times n minus 1msumi=1
nsumj=1(pij minus p) (qij minus q) (21)
Table 3 notes the results of seven-way PVD [18] whichis an extension of Wu and Tsairsquos [1] technique Table 4 notes
the results of Khodaei and Faezrsquos technique [10] and Table 5notes the results of the proposed technique The PSNR andcapacity of the proposed technique (both type 1 and type2) are better than seven-way PVD The bit rate and Q arealso better Although as compared to Khodaei and Faezrsquostechnique the PSNR is not improved but the hiding capacityis improved The major advantage over Khodaei and Faezrsquostechnique is that the proposed technique is resistant to PDHanalysis but Khodaei and Faezrsquos technique is detectable byPDH analysis This is made possible by exploiting the edgesin eight directions
Now let us observe the results of proposed technique andcompare type 1 and type 2 We can observe that the capacityis higher in type 2 as compared to type 1 This is because ofthe 4 LSBs substitution in type 2 at the central pixel ie the tvalue is 4 On the other hand the PSNR of type 1 is better thanthat of type 2
Table 6 represents the comparison of the proposedtechnique with the adaptive PVD technique in [26] andhybrid techniques in [27 28] The comparison parametersare average PSNR and average hiding capacity values Thetwo variants of adaptive PVD technique possess lesser hidingcapacity as compared to the proposed eight-directional PVDAs the hiding capacity of adaptive PVD [26] is very littleso PSNR is larger The hybrid technique proposed in [27] isa combination of PVD LSB and EMD The hiding capacityof proposed technique is greater than all the four variantsof the hybrid technique proposed in [27] The technique in[28] uses PVD with modified LSB (MLSB) substitution Theresults of this technique (type 1) are as good as the results of
8 Advances in Multimedia
Table 5 Results of proposed eight-directional PVD technique
Images 512times512times3 Proposed type 1 (t=3 Range Table 1) Proposed type 2 (t=4 Range Table 2)PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4115 2354240 09996 299 3711 2475977 09988 315House 3969 2358575 09988 300 3839 2625804 09984 334Boat 3725 2364685 09987 301 3408 2659795 09973 338Jet 4221 2356828 09989 300 4038 2538801 09985 323Peppers 3781 2356645 09988 300 3469 2544392 09975 324Tiffany 4146 2353505 09987 299 4053 2511139 09984 319Baboon 3426 2392573 09961 304 3209 2939376 09937 374Lena 4262 2353892 09995 299 4048 2533551 09992 322Average 3955 2361368 09986 300 3722 2603604 09977 331
Table 6 Comparison of the proposed technique with techniques in [26ndash28]
Technique PSNR Capacity Amount of hidden data in stego image to compute PSNR2times3 adaptive PVD [26] 5093 1445645 One lakh and forty thousand (140000) bits3times2 adaptive PVD [26] 5093 14356233 PVD +3 LSB +EMD type 1 [27] 4107 1686041
Seven lakhs (700000) bits3 PVD +3 LSB +EMD type 2 [27] 3895 17988347 PVD +3 LSB +EMD type 1 [27] 4028 17108887 PVD +3 LSB +EMD type 2 [27] 3726 1823282Proposed eight-directional PVD type 1 3955 2361368 Seven lakhs (700000) bitsProposed eight-directional PVD type 2 3722 2603604Five-directional PVD+MLSB type 1 [28] 3762 2399853
Eight lakhs and forty thousand (840000) bitsFive-directional PVD+MLSB type 2 [28] 3623 2514038Eight-directional PVD+MLSB type 1 [28] 3749 2394086Eight-directional PVD+MLSB type 2 [28] 3560 2603604
the proposed technique (type 1) Similarly the results of thistechnique (type 2) are as good as the results of the proposedtechnique (type 2)
5 Security Analysis
The LSB substitution techniques suffer from RS analysis andPVD techniques suffer from PDH analysis The proposedtechnique uses the concepts like modified LSB substitutionandPVD so itmust be analyzed by both RS analysis andPDHanalysis
The PDH analysis is performed by calculating the differ-ence value between every pair of pixels A pair comprises twoconsecutive pixels This difference values will be from -255 to+255 including 0 The frequency of each of these differencevalues is counted A graph is plotted with the pixel differencevalue on X-axis and frequency on Y-axis The curve obtainedis called the PDHThePDHof original imagewill be a smoothcurve If the PDH of stego image is also a smooth curve thensteganography is not detected If the PDH of stego imageshows step effects then steganography is identified
Figure 7 is the PDH analysis of type 1 and Figure 8 isthe PDH analysis of type 2 of the proposed technique Theanalysis is carried over all the eight test images For eachimage the PDH of the original image is represented by solidline and the PDH of the stego image is represented by dottedlineThe solid line curves will be obviously free of step effectsas they are of the original images The dotted line curves in
all the sixteen cases do not show any step effects This justifiesthat the proposed technique is resistant to PDH analysis
The RS analysis is based on statistical measures It is donein the following manner Define a function F
1 2nlarrrarr2n+1
It defines two transformations (i) from value 2n to value2n + 1 and (ii) from value 2n + 1 to value 2n Similarlydefine another function F
minus1 2n larrrarr2n -1 It defines other
two transformations (i) from value 2n to value 2n ndash 1 and(ii) from value 2n - 1 to value 2n The image say M isdivided into a number of equal size blocks Suppose such ablock is G whose pixels are X
1X2X3 Xn Then use the
function f(X1X2X3 Xn)=sumnminus1i=1 |Xi+1minusXi| tomeasure the
smoothness of G Then apply F1to all the blocks of M and
define the two parameters Rm and Sm as in (22) and (23)Similarly apply F
minus1to all the blocks of M and define the two
parameters Rminusm and S
minusm as in (24) and (25)
Rm
= No of blocks satisfying the condition f (F1(G)) gt f (G)
Total number of blocks
(22)
Sm
= No of blocks satisfying the condition f (F1(G)) lt f (G)
Total number of blocks
(23)
Rminusm
= No of blocks satisfying the condition f (Fminus1(G)) gt f (G)
Total number of blocks
(24)
Advances in Multimedia 9
0
1
2
Occ
urre
nces
3 times104
Lena coverLena stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3 times104
Baboon coverBaboon stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
Tiffany coverTiffany stego
times104
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Pot coverPot stego
times104
Figure 7 PDH analysis for type 1 of the proposed technique
10 Advances in Multimedia
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Lena coverLena stego
times104
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Baboon coverBaboon stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Tiffany coverTiffany stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Pot coverPot stego
times104
Figure 8 PDH analysis for type 2 of the proposed technique
Advances in Multimedia 11
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(a) Type 1 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(b) Type 1 Baboon
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(c) Type 2 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(d) Type 2 Baboon
Figure 9 RS analysis for type 1 and type 2 of the proposed technique
Sminusm
= No of blocks satisfying the condition f (Fminus1(G)) lt f (G)
Total number of blocks
(25)
RS analysis is performed by using these four parametersIf the condition R
119898asymp Rminus119898gt Sm asymp S
minusm is true then RSanalysis fails to detect the steganography technique But if the
condition Rminus119898
- Sminusm gt Rm - Sm is true then the RS analysis
succeeds in detecting the steganography techniqueFigure 9 represents RS analysis for the proposed tech-
nique Figures 9(a) and 9(b) stand for RS analysis over Lenaand Baboon images respectively of type 1 Similarly Figures9(c) and 9(d) stand for RS analysis over Lena and Baboonimages respectively of type 2 In all the four cases the
12 Advances in Multimedia
condition R119898asymp Rminus119898gt Sm asymp Sminusm is true so we can conclude
that RS analysis could not detect the proposed techniqueLena image has more smooth regions and Baboon imagehas more edge regions all the remaining six images are inbetween these two in terms of smoothness So it is sufficientto draw a conclusion based on these two images
6 Conclusion
The traditional PVD steganography techniques follow a staticrange table Due to this some undesired steps are introducedin pixel difference histograms of the stego images This stepeffect could be avoided by exploiting edges in eight differentdirections This paper proposes a steganography techniqueusing a judicious combination of LSB substitution and eight-directional PVD The LSB substitution is performed only inone pixel out of nine pixels in a block so RS analysis cannotdetect it By using the edges in multiple directions the PDHanalysis cannot detect itThere exist two types of the proposedtechnique with regard to two different quantization tablesThe type 1 provides higher PSNR and the type 2 provideshigher hiding capacity This proposed technique performsbetter in terms of hiding capacity and PSNR over othercompeting PVD techniques
Conflicts of Interest
The author declares that there are no conflicts of interest
References
[1] D-CWu andW-H Tsai ldquoA steganographicmethod for imagesby pixel-value differencingrdquo Pattern Recognition Letters vol 24no 9-10 pp 1613ndash1626 2003
[2] K C Chang C P Chang P S Huang and T M Tu ldquoAnovel image steganographic method using tri-way pixel-valuedifferencingrdquo Journal of Multimedia vol 3 no 2 pp 37ndash442008
[3] Y-P Lee J-C Lee W-K Chen K-C Chang I-J Su and C-P Chang ldquoHigh-payload image hiding with quality recoveryusing tri-way pixel-value differencingrdquo Information Sciencesvol 191 pp 214ndash225 2012
[4] C-C Chang and H-W Tseng ldquoA steganographic method fordigital images using side matchrdquo Pattern Recognition Lettersvol 25 no 12 pp 1431ndash1437 2004
[5] C-H Yang C-Y Weng H-K Tso and S-J Wang ldquoA datahiding scheme using the varieties of pixel-value differencing inmultimedia imagesrdquo The Journal of Systems and Software vol84 no 4 pp 669ndash678 2011
[6] W Hong T-S Chen and C-W Luo ldquoData embedding usingpixel value differencing and diamond encoding with multiple-base notational systemrdquo The Journal of Systems and Softwarevol 85 no 5 pp 1166ndash1175 2012
[7] H C Wu N I Wu C S Tsai and M-S Hwang ldquoImagesteganographic scheme based on pixel-value differencing andLSB replacementmethodsrdquo IEE ProceedingsmdashVision Image andSignal Processing vol 152 no 5 pp 611ndash615 2005
[8] C-H Yang C-Y Weng S-J Wang and H-M Sun ldquoVariedPVD + LSB evading detection programs to spatial domain in
data embedding systemsrdquo The Journal of Systems and Softwarevol 83 no 10 pp 1635ndash1643 2010
[9] J Chen ldquoA PVD-based data hiding method with histogrampreserving using pixel pair matchingrdquo Signal Processing ImageCommunication vol 29 no 3 pp 375ndash384 2014
[10] MKhodaei andK Faez ldquoNew adaptive steganographicmethodusing least-significant-bit substitution and pixel-value differ-encingrdquo IET Image Processing vol 6 no 6 pp 677ndash686 2012
[11] G Swain ldquoA steganographic method combining LSB substitu-tion and PVD in a blockrdquo Procedia Computer Science vol 85pp 39ndash44 2016
[12] X Liao Q Wen and J Zhang ldquoA steganographic methodfor digital images with four-pixel differencing and modifiedLSB substitutionrdquo Journal of Visual Communication and ImageRepresentation vol 22 no 1 pp 1ndash8 2011
[13] G Swain ldquoDigital image steganography using nine-pixel dif-ferencing and modified LSB substitutionrdquo Indian Journal ofScience and Technology vol 7 no 9 pp 1444ndash1450 2014httpwwwindjstorgindexphpindjstarticle
[14] W Luo F Huang and J Huang ldquoA more secure steganographybased on adaptive pixel-value differencing schemerdquoMultimediaTools and Applications vol 52 no 2-3 pp 407ndash430 2011
[15] G Swain ldquoAdaptive pixel value differencing steganographyusing both vertical and horizontal edgesrdquoMultimedia Tools andApplications vol 75 no 21 pp 13541ndash13556 2016
[16] S Chakraborty A S Jalal and C Bhatnagar ldquoLSB based nonblind predictive edge adaptive image steganographyrdquoMultime-dia Tools and Applications vol 76 no 6 pp 7973ndash7987 2017
[17] C Balasubramanian S Selvakumar and S Geetha ldquoHighpayload image steganography with reduced distortion usingoctonary pixel pairing schemerdquoMultimedia Tools and Applica-tions vol 73 no 3 pp 2223ndash2245 2014
[18] A Pradhan K R Sekhar and G Swain ldquoDigital imagesteganography based on seven way pixel value differencingrdquoIndian Journal of Science and Technology vol 9 no 37 pp 1ndash112016
[19] K A Darabkh A K Al-Dhamari and I F Jafar ldquoA newsteganographic algorithm based on multi directional PVD andmodified LSBrdquo Information Technology and Control vol 46 no1 pp 16ndash36 2017
[20] A K Gulve and M S Joshi ldquoA high capacity secured imagesteganography method with five pixel pair differencing andLSB substitutionrdquo International Journal of Image Graphics andSignal Processing vol 7 no 5 pp 66ndash74 2015
[21] X Zhang and S Wang ldquoEfficient steganographic embeddingby exploiting modification directionrdquo IEEE CommunicationsLetters vol 10 no 11 pp 781ndash783 2006
[22] C Kim ldquoData hiding by an improved exploiting modificationdirectionrdquoMultimedia Tools and Applications vol 69 no 3 pp569ndash584 2014
[23] S Y Shen and L H Huang ldquoA data hiding scheme usingpixel value differencing and improving exploiting modificationdirectionsrdquo Computers amp Security vol 48 pp 131ndash141 2015
[24] T D Nguyen S Arch-Int and N Arch-Int ldquoAn adaptivemulti bit-plane image steganography using block data-hidingrdquoMultimediaTools andApplications vol 75 no 14 pp 8319ndash83452016
[25] A Soria-Lorente and S Berres ldquoA secure steganographicalgorithm based on frequency domain for the transmission ofhidden informationrdquo Security and Communication Networksvol 2017 Article ID 5397028 14 pages 2017
Advances in Multimedia 13
[26] A Pradhan K R Sekhar and G Swain ldquoAdaptive PVDsteganography using horizontal vertical and diagonal edges insix-pixel blocksrdquo Security and Communication Networks vol2017 Article ID 1924618 13 pages 2017
[27] A Pradhan K R ekhar andG Swain ldquoDigital image steganog-raphy using LSB substitution PVD and EMDrdquo MathematicalProblems in Engineering 2018
[28] G Swain ldquoHigh capacity image steganography using modifiedLSB substitution and PVD against pixel difference histogramanalysisrdquo Security and Communication Networks 2018
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Advances in Multimedia 5
196 188 202
193 184 186
196 190 205
Stego block
[0 7] 3
[8 15] 3
[16 31] 3
[32 63] 3
[64 127] 4
[128 255] 4
Range table 1
4 4 2 1 2 4 6 5
Extracted data 000 100 100 010 001 010 100 110 101
|184 minus 196| = 12
|184 minus 188| = 4
|184 minus 202| = 18
|184 minus 193| = 9
|184 minus 186| = 2
|184 minus 196| = 12
|184 minus 190| = 6
|184 minus 205| = 21Pc is 184
184 = (10111)2
t1 = 3
t2 = 3
t3 = 3
t4 = 3
t5 = 3
t6 = 3
t7 = 3
t8 = 3
L1 = 8
L2 = 0
L3 = 16
L4 = 8
L5 = 0
L6 = 8
L7 = 0
L8 = 16
S1 = 12 minus 8 = 4
S3 = 18 minus 6 = 2
S6 = 12 minus 8 = 4
S8 = 21 minus 16 = 5
S2 = 4 minus 0 = 4
S4 = 9 minus 8 = 1
S5 = 2 minus 0 = 2
S7 = 6 minus 0 = 6
Si rarr
d1 =
d2 =
d3 =
d4 =
d5 =
d6 =
d7 =
d8 =
Figure 3 Example of extraction
(a) Pot (b) House (c) Boat (d) Jet
(e) Peppers (f) Tiffany (g) Baboon (h) Lena
Figure 4 Original images
per byte (BPB) indicate the average hiding capacity per a byteof the original image Equation (13) is used to calculate BPB
BPB = hiding capacityimage size in bytes (13)
Distortion in stego image is measured in terms of PSNREquation (14) is used to calculate it
PSNR = 10 times log10
255 times 255MSE
(14)
where MSE denotes the mean square error and can becomputed using
MSE = 1m times n
msumi=1
nsumj=1(pij minus qij)
2
(15)
where pij is a pixel of cover image and qij is the correspondingstego pixel
Thequality index (Q) is another parameter tomeasure thestructural similarity between original image and stego image
6 Advances in Multimedia
PSNR = 4115(a)
PSNR = 3969(b)
PSNR = 3725(c)
PSNR = 4221(d)
PSNR = 3781(e)
PSNR = 4146(f)
PSNR = 3426(g)
PSNR = 4262(h)
Figure 5 Stego images of proposed type 1 (t=3 range table 1)
PSNR = 3711(a)
PSNR = 3839(b)
PSNR = 3408(c)
PSNR = 4038(d)
PSNR = 3469(e)
PSNR = 4053(f)
PSNR = 3209(g)
PSNR = 4048(h)
Figure 6 Stego images of proposed type 2 (t=4 range table 2)
Advances in Multimedia 7
Table 3 Results of Pradhan et alrsquos seven-way PVD technique [18]
Images 512times512times3 Seven-way PVD [18] Seven-way+ one-way PVD [18]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4237 1795551 09996 228 4240 1803635 09996 229House 3897 1972223 09986 250 3898 1977403 09986 251Boat 3789 1972086 09988 250 3791 1991005 09988 253Jet 4209 1906254 09989 242 4198 1909595 09989 242Peppers 4042 1778072 09993 226 4020 1806166 09992 229Tiffany 4123 1400756 09986 178 4130 1450799 09986 184Baboon 3379 2226806 09957 283 3377 2243218 09957 285Lena 4173 1896662 09993 241 4173 1901149 09993 241Average 3981 1868551 09986 237 3978 1885371 09985 239
Table 4 Results of Khodaei and Faezrsquos technique
Images 512times512times3 Khodaei and Faez type 1 [10] Khodaei and Faez type 2 [10]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4121 2366001 09995 300 3783 2387494 09990 303House 3916 2387183 09986 303 3875 2470824 09985 314Boat 4005 2391994 09993 304 3749 2504613 09987 318Jet 4089 2374048 09986 301 4064 2418419 09985 307Peppers 3932 2372858 09991 301 3791 2435223 09987 309Tiffany 3973 2372396 09980 301 4025 2416944 09982 307Baboon 3777 2443361 09982 310 3449 2662080 09963 338Lena 4278 2375248 09995 302 4125 2434603 09993 309Average 4011 2385386 09988 303 3857 2466275 09984 313
Equation (16) is used to compute itThemaximum value of Qcan be 1 if original image p and stego image q are the same
Q =4120590xyp q
(1205902x + 1205902y) [(p)2 + (q)2] (16)
where p stands for the mean pixel value of the original imageq stands for the mean pixel value of the stego image1205902x standsfor the standard deviation of pixel values of original image1205902y stands for the standard deviation of pixel values of stegoimage and 120590xy is the covariance between original image andstego image Equations (17) (18) (19) (20) and (21) are usedto calculate these values
p = 1m times n
msumi=1
nsumj=1pij (17)
q = 1m times n
msumi=1
nsumj=1qij (18)
1205902x =1
m times n minus 1msumi=1
nsumj=1(pij minus p)
2
(19)
1205902y =1
m times n minus 1msumi=1
nsumj=1(qij minus q)
2
(20)
120590xy =1
m times n minus 1msumi=1
nsumj=1(pij minus p) (qij minus q) (21)
Table 3 notes the results of seven-way PVD [18] whichis an extension of Wu and Tsairsquos [1] technique Table 4 notes
the results of Khodaei and Faezrsquos technique [10] and Table 5notes the results of the proposed technique The PSNR andcapacity of the proposed technique (both type 1 and type2) are better than seven-way PVD The bit rate and Q arealso better Although as compared to Khodaei and Faezrsquostechnique the PSNR is not improved but the hiding capacityis improved The major advantage over Khodaei and Faezrsquostechnique is that the proposed technique is resistant to PDHanalysis but Khodaei and Faezrsquos technique is detectable byPDH analysis This is made possible by exploiting the edgesin eight directions
Now let us observe the results of proposed technique andcompare type 1 and type 2 We can observe that the capacityis higher in type 2 as compared to type 1 This is because ofthe 4 LSBs substitution in type 2 at the central pixel ie the tvalue is 4 On the other hand the PSNR of type 1 is better thanthat of type 2
Table 6 represents the comparison of the proposedtechnique with the adaptive PVD technique in [26] andhybrid techniques in [27 28] The comparison parametersare average PSNR and average hiding capacity values Thetwo variants of adaptive PVD technique possess lesser hidingcapacity as compared to the proposed eight-directional PVDAs the hiding capacity of adaptive PVD [26] is very littleso PSNR is larger The hybrid technique proposed in [27] isa combination of PVD LSB and EMD The hiding capacityof proposed technique is greater than all the four variantsof the hybrid technique proposed in [27] The technique in[28] uses PVD with modified LSB (MLSB) substitution Theresults of this technique (type 1) are as good as the results of
8 Advances in Multimedia
Table 5 Results of proposed eight-directional PVD technique
Images 512times512times3 Proposed type 1 (t=3 Range Table 1) Proposed type 2 (t=4 Range Table 2)PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4115 2354240 09996 299 3711 2475977 09988 315House 3969 2358575 09988 300 3839 2625804 09984 334Boat 3725 2364685 09987 301 3408 2659795 09973 338Jet 4221 2356828 09989 300 4038 2538801 09985 323Peppers 3781 2356645 09988 300 3469 2544392 09975 324Tiffany 4146 2353505 09987 299 4053 2511139 09984 319Baboon 3426 2392573 09961 304 3209 2939376 09937 374Lena 4262 2353892 09995 299 4048 2533551 09992 322Average 3955 2361368 09986 300 3722 2603604 09977 331
Table 6 Comparison of the proposed technique with techniques in [26ndash28]
Technique PSNR Capacity Amount of hidden data in stego image to compute PSNR2times3 adaptive PVD [26] 5093 1445645 One lakh and forty thousand (140000) bits3times2 adaptive PVD [26] 5093 14356233 PVD +3 LSB +EMD type 1 [27] 4107 1686041
Seven lakhs (700000) bits3 PVD +3 LSB +EMD type 2 [27] 3895 17988347 PVD +3 LSB +EMD type 1 [27] 4028 17108887 PVD +3 LSB +EMD type 2 [27] 3726 1823282Proposed eight-directional PVD type 1 3955 2361368 Seven lakhs (700000) bitsProposed eight-directional PVD type 2 3722 2603604Five-directional PVD+MLSB type 1 [28] 3762 2399853
Eight lakhs and forty thousand (840000) bitsFive-directional PVD+MLSB type 2 [28] 3623 2514038Eight-directional PVD+MLSB type 1 [28] 3749 2394086Eight-directional PVD+MLSB type 2 [28] 3560 2603604
the proposed technique (type 1) Similarly the results of thistechnique (type 2) are as good as the results of the proposedtechnique (type 2)
5 Security Analysis
The LSB substitution techniques suffer from RS analysis andPVD techniques suffer from PDH analysis The proposedtechnique uses the concepts like modified LSB substitutionandPVD so itmust be analyzed by both RS analysis andPDHanalysis
The PDH analysis is performed by calculating the differ-ence value between every pair of pixels A pair comprises twoconsecutive pixels This difference values will be from -255 to+255 including 0 The frequency of each of these differencevalues is counted A graph is plotted with the pixel differencevalue on X-axis and frequency on Y-axis The curve obtainedis called the PDHThePDHof original imagewill be a smoothcurve If the PDH of stego image is also a smooth curve thensteganography is not detected If the PDH of stego imageshows step effects then steganography is identified
Figure 7 is the PDH analysis of type 1 and Figure 8 isthe PDH analysis of type 2 of the proposed technique Theanalysis is carried over all the eight test images For eachimage the PDH of the original image is represented by solidline and the PDH of the stego image is represented by dottedlineThe solid line curves will be obviously free of step effectsas they are of the original images The dotted line curves in
all the sixteen cases do not show any step effects This justifiesthat the proposed technique is resistant to PDH analysis
The RS analysis is based on statistical measures It is donein the following manner Define a function F
1 2nlarrrarr2n+1
It defines two transformations (i) from value 2n to value2n + 1 and (ii) from value 2n + 1 to value 2n Similarlydefine another function F
minus1 2n larrrarr2n -1 It defines other
two transformations (i) from value 2n to value 2n ndash 1 and(ii) from value 2n - 1 to value 2n The image say M isdivided into a number of equal size blocks Suppose such ablock is G whose pixels are X
1X2X3 Xn Then use the
function f(X1X2X3 Xn)=sumnminus1i=1 |Xi+1minusXi| tomeasure the
smoothness of G Then apply F1to all the blocks of M and
define the two parameters Rm and Sm as in (22) and (23)Similarly apply F
minus1to all the blocks of M and define the two
parameters Rminusm and S
minusm as in (24) and (25)
Rm
= No of blocks satisfying the condition f (F1(G)) gt f (G)
Total number of blocks
(22)
Sm
= No of blocks satisfying the condition f (F1(G)) lt f (G)
Total number of blocks
(23)
Rminusm
= No of blocks satisfying the condition f (Fminus1(G)) gt f (G)
Total number of blocks
(24)
Advances in Multimedia 9
0
1
2
Occ
urre
nces
3 times104
Lena coverLena stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3 times104
Baboon coverBaboon stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
Tiffany coverTiffany stego
times104
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Pot coverPot stego
times104
Figure 7 PDH analysis for type 1 of the proposed technique
10 Advances in Multimedia
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Lena coverLena stego
times104
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Baboon coverBaboon stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Tiffany coverTiffany stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Pot coverPot stego
times104
Figure 8 PDH analysis for type 2 of the proposed technique
Advances in Multimedia 11
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(a) Type 1 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(b) Type 1 Baboon
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(c) Type 2 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(d) Type 2 Baboon
Figure 9 RS analysis for type 1 and type 2 of the proposed technique
Sminusm
= No of blocks satisfying the condition f (Fminus1(G)) lt f (G)
Total number of blocks
(25)
RS analysis is performed by using these four parametersIf the condition R
119898asymp Rminus119898gt Sm asymp S
minusm is true then RSanalysis fails to detect the steganography technique But if the
condition Rminus119898
- Sminusm gt Rm - Sm is true then the RS analysis
succeeds in detecting the steganography techniqueFigure 9 represents RS analysis for the proposed tech-
nique Figures 9(a) and 9(b) stand for RS analysis over Lenaand Baboon images respectively of type 1 Similarly Figures9(c) and 9(d) stand for RS analysis over Lena and Baboonimages respectively of type 2 In all the four cases the
12 Advances in Multimedia
condition R119898asymp Rminus119898gt Sm asymp Sminusm is true so we can conclude
that RS analysis could not detect the proposed techniqueLena image has more smooth regions and Baboon imagehas more edge regions all the remaining six images are inbetween these two in terms of smoothness So it is sufficientto draw a conclusion based on these two images
6 Conclusion
The traditional PVD steganography techniques follow a staticrange table Due to this some undesired steps are introducedin pixel difference histograms of the stego images This stepeffect could be avoided by exploiting edges in eight differentdirections This paper proposes a steganography techniqueusing a judicious combination of LSB substitution and eight-directional PVD The LSB substitution is performed only inone pixel out of nine pixels in a block so RS analysis cannotdetect it By using the edges in multiple directions the PDHanalysis cannot detect itThere exist two types of the proposedtechnique with regard to two different quantization tablesThe type 1 provides higher PSNR and the type 2 provideshigher hiding capacity This proposed technique performsbetter in terms of hiding capacity and PSNR over othercompeting PVD techniques
Conflicts of Interest
The author declares that there are no conflicts of interest
References
[1] D-CWu andW-H Tsai ldquoA steganographicmethod for imagesby pixel-value differencingrdquo Pattern Recognition Letters vol 24no 9-10 pp 1613ndash1626 2003
[2] K C Chang C P Chang P S Huang and T M Tu ldquoAnovel image steganographic method using tri-way pixel-valuedifferencingrdquo Journal of Multimedia vol 3 no 2 pp 37ndash442008
[3] Y-P Lee J-C Lee W-K Chen K-C Chang I-J Su and C-P Chang ldquoHigh-payload image hiding with quality recoveryusing tri-way pixel-value differencingrdquo Information Sciencesvol 191 pp 214ndash225 2012
[4] C-C Chang and H-W Tseng ldquoA steganographic method fordigital images using side matchrdquo Pattern Recognition Lettersvol 25 no 12 pp 1431ndash1437 2004
[5] C-H Yang C-Y Weng H-K Tso and S-J Wang ldquoA datahiding scheme using the varieties of pixel-value differencing inmultimedia imagesrdquo The Journal of Systems and Software vol84 no 4 pp 669ndash678 2011
[6] W Hong T-S Chen and C-W Luo ldquoData embedding usingpixel value differencing and diamond encoding with multiple-base notational systemrdquo The Journal of Systems and Softwarevol 85 no 5 pp 1166ndash1175 2012
[7] H C Wu N I Wu C S Tsai and M-S Hwang ldquoImagesteganographic scheme based on pixel-value differencing andLSB replacementmethodsrdquo IEE ProceedingsmdashVision Image andSignal Processing vol 152 no 5 pp 611ndash615 2005
[8] C-H Yang C-Y Weng S-J Wang and H-M Sun ldquoVariedPVD + LSB evading detection programs to spatial domain in
data embedding systemsrdquo The Journal of Systems and Softwarevol 83 no 10 pp 1635ndash1643 2010
[9] J Chen ldquoA PVD-based data hiding method with histogrampreserving using pixel pair matchingrdquo Signal Processing ImageCommunication vol 29 no 3 pp 375ndash384 2014
[10] MKhodaei andK Faez ldquoNew adaptive steganographicmethodusing least-significant-bit substitution and pixel-value differ-encingrdquo IET Image Processing vol 6 no 6 pp 677ndash686 2012
[11] G Swain ldquoA steganographic method combining LSB substitu-tion and PVD in a blockrdquo Procedia Computer Science vol 85pp 39ndash44 2016
[12] X Liao Q Wen and J Zhang ldquoA steganographic methodfor digital images with four-pixel differencing and modifiedLSB substitutionrdquo Journal of Visual Communication and ImageRepresentation vol 22 no 1 pp 1ndash8 2011
[13] G Swain ldquoDigital image steganography using nine-pixel dif-ferencing and modified LSB substitutionrdquo Indian Journal ofScience and Technology vol 7 no 9 pp 1444ndash1450 2014httpwwwindjstorgindexphpindjstarticle
[14] W Luo F Huang and J Huang ldquoA more secure steganographybased on adaptive pixel-value differencing schemerdquoMultimediaTools and Applications vol 52 no 2-3 pp 407ndash430 2011
[15] G Swain ldquoAdaptive pixel value differencing steganographyusing both vertical and horizontal edgesrdquoMultimedia Tools andApplications vol 75 no 21 pp 13541ndash13556 2016
[16] S Chakraborty A S Jalal and C Bhatnagar ldquoLSB based nonblind predictive edge adaptive image steganographyrdquoMultime-dia Tools and Applications vol 76 no 6 pp 7973ndash7987 2017
[17] C Balasubramanian S Selvakumar and S Geetha ldquoHighpayload image steganography with reduced distortion usingoctonary pixel pairing schemerdquoMultimedia Tools and Applica-tions vol 73 no 3 pp 2223ndash2245 2014
[18] A Pradhan K R Sekhar and G Swain ldquoDigital imagesteganography based on seven way pixel value differencingrdquoIndian Journal of Science and Technology vol 9 no 37 pp 1ndash112016
[19] K A Darabkh A K Al-Dhamari and I F Jafar ldquoA newsteganographic algorithm based on multi directional PVD andmodified LSBrdquo Information Technology and Control vol 46 no1 pp 16ndash36 2017
[20] A K Gulve and M S Joshi ldquoA high capacity secured imagesteganography method with five pixel pair differencing andLSB substitutionrdquo International Journal of Image Graphics andSignal Processing vol 7 no 5 pp 66ndash74 2015
[21] X Zhang and S Wang ldquoEfficient steganographic embeddingby exploiting modification directionrdquo IEEE CommunicationsLetters vol 10 no 11 pp 781ndash783 2006
[22] C Kim ldquoData hiding by an improved exploiting modificationdirectionrdquoMultimedia Tools and Applications vol 69 no 3 pp569ndash584 2014
[23] S Y Shen and L H Huang ldquoA data hiding scheme usingpixel value differencing and improving exploiting modificationdirectionsrdquo Computers amp Security vol 48 pp 131ndash141 2015
[24] T D Nguyen S Arch-Int and N Arch-Int ldquoAn adaptivemulti bit-plane image steganography using block data-hidingrdquoMultimediaTools andApplications vol 75 no 14 pp 8319ndash83452016
[25] A Soria-Lorente and S Berres ldquoA secure steganographicalgorithm based on frequency domain for the transmission ofhidden informationrdquo Security and Communication Networksvol 2017 Article ID 5397028 14 pages 2017
Advances in Multimedia 13
[26] A Pradhan K R Sekhar and G Swain ldquoAdaptive PVDsteganography using horizontal vertical and diagonal edges insix-pixel blocksrdquo Security and Communication Networks vol2017 Article ID 1924618 13 pages 2017
[27] A Pradhan K R ekhar andG Swain ldquoDigital image steganog-raphy using LSB substitution PVD and EMDrdquo MathematicalProblems in Engineering 2018
[28] G Swain ldquoHigh capacity image steganography using modifiedLSB substitution and PVD against pixel difference histogramanalysisrdquo Security and Communication Networks 2018
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
6 Advances in Multimedia
PSNR = 4115(a)
PSNR = 3969(b)
PSNR = 3725(c)
PSNR = 4221(d)
PSNR = 3781(e)
PSNR = 4146(f)
PSNR = 3426(g)
PSNR = 4262(h)
Figure 5 Stego images of proposed type 1 (t=3 range table 1)
PSNR = 3711(a)
PSNR = 3839(b)
PSNR = 3408(c)
PSNR = 4038(d)
PSNR = 3469(e)
PSNR = 4053(f)
PSNR = 3209(g)
PSNR = 4048(h)
Figure 6 Stego images of proposed type 2 (t=4 range table 2)
Advances in Multimedia 7
Table 3 Results of Pradhan et alrsquos seven-way PVD technique [18]
Images 512times512times3 Seven-way PVD [18] Seven-way+ one-way PVD [18]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4237 1795551 09996 228 4240 1803635 09996 229House 3897 1972223 09986 250 3898 1977403 09986 251Boat 3789 1972086 09988 250 3791 1991005 09988 253Jet 4209 1906254 09989 242 4198 1909595 09989 242Peppers 4042 1778072 09993 226 4020 1806166 09992 229Tiffany 4123 1400756 09986 178 4130 1450799 09986 184Baboon 3379 2226806 09957 283 3377 2243218 09957 285Lena 4173 1896662 09993 241 4173 1901149 09993 241Average 3981 1868551 09986 237 3978 1885371 09985 239
Table 4 Results of Khodaei and Faezrsquos technique
Images 512times512times3 Khodaei and Faez type 1 [10] Khodaei and Faez type 2 [10]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4121 2366001 09995 300 3783 2387494 09990 303House 3916 2387183 09986 303 3875 2470824 09985 314Boat 4005 2391994 09993 304 3749 2504613 09987 318Jet 4089 2374048 09986 301 4064 2418419 09985 307Peppers 3932 2372858 09991 301 3791 2435223 09987 309Tiffany 3973 2372396 09980 301 4025 2416944 09982 307Baboon 3777 2443361 09982 310 3449 2662080 09963 338Lena 4278 2375248 09995 302 4125 2434603 09993 309Average 4011 2385386 09988 303 3857 2466275 09984 313
Equation (16) is used to compute itThemaximum value of Qcan be 1 if original image p and stego image q are the same
Q =4120590xyp q
(1205902x + 1205902y) [(p)2 + (q)2] (16)
where p stands for the mean pixel value of the original imageq stands for the mean pixel value of the stego image1205902x standsfor the standard deviation of pixel values of original image1205902y stands for the standard deviation of pixel values of stegoimage and 120590xy is the covariance between original image andstego image Equations (17) (18) (19) (20) and (21) are usedto calculate these values
p = 1m times n
msumi=1
nsumj=1pij (17)
q = 1m times n
msumi=1
nsumj=1qij (18)
1205902x =1
m times n minus 1msumi=1
nsumj=1(pij minus p)
2
(19)
1205902y =1
m times n minus 1msumi=1
nsumj=1(qij minus q)
2
(20)
120590xy =1
m times n minus 1msumi=1
nsumj=1(pij minus p) (qij minus q) (21)
Table 3 notes the results of seven-way PVD [18] whichis an extension of Wu and Tsairsquos [1] technique Table 4 notes
the results of Khodaei and Faezrsquos technique [10] and Table 5notes the results of the proposed technique The PSNR andcapacity of the proposed technique (both type 1 and type2) are better than seven-way PVD The bit rate and Q arealso better Although as compared to Khodaei and Faezrsquostechnique the PSNR is not improved but the hiding capacityis improved The major advantage over Khodaei and Faezrsquostechnique is that the proposed technique is resistant to PDHanalysis but Khodaei and Faezrsquos technique is detectable byPDH analysis This is made possible by exploiting the edgesin eight directions
Now let us observe the results of proposed technique andcompare type 1 and type 2 We can observe that the capacityis higher in type 2 as compared to type 1 This is because ofthe 4 LSBs substitution in type 2 at the central pixel ie the tvalue is 4 On the other hand the PSNR of type 1 is better thanthat of type 2
Table 6 represents the comparison of the proposedtechnique with the adaptive PVD technique in [26] andhybrid techniques in [27 28] The comparison parametersare average PSNR and average hiding capacity values Thetwo variants of adaptive PVD technique possess lesser hidingcapacity as compared to the proposed eight-directional PVDAs the hiding capacity of adaptive PVD [26] is very littleso PSNR is larger The hybrid technique proposed in [27] isa combination of PVD LSB and EMD The hiding capacityof proposed technique is greater than all the four variantsof the hybrid technique proposed in [27] The technique in[28] uses PVD with modified LSB (MLSB) substitution Theresults of this technique (type 1) are as good as the results of
8 Advances in Multimedia
Table 5 Results of proposed eight-directional PVD technique
Images 512times512times3 Proposed type 1 (t=3 Range Table 1) Proposed type 2 (t=4 Range Table 2)PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4115 2354240 09996 299 3711 2475977 09988 315House 3969 2358575 09988 300 3839 2625804 09984 334Boat 3725 2364685 09987 301 3408 2659795 09973 338Jet 4221 2356828 09989 300 4038 2538801 09985 323Peppers 3781 2356645 09988 300 3469 2544392 09975 324Tiffany 4146 2353505 09987 299 4053 2511139 09984 319Baboon 3426 2392573 09961 304 3209 2939376 09937 374Lena 4262 2353892 09995 299 4048 2533551 09992 322Average 3955 2361368 09986 300 3722 2603604 09977 331
Table 6 Comparison of the proposed technique with techniques in [26ndash28]
Technique PSNR Capacity Amount of hidden data in stego image to compute PSNR2times3 adaptive PVD [26] 5093 1445645 One lakh and forty thousand (140000) bits3times2 adaptive PVD [26] 5093 14356233 PVD +3 LSB +EMD type 1 [27] 4107 1686041
Seven lakhs (700000) bits3 PVD +3 LSB +EMD type 2 [27] 3895 17988347 PVD +3 LSB +EMD type 1 [27] 4028 17108887 PVD +3 LSB +EMD type 2 [27] 3726 1823282Proposed eight-directional PVD type 1 3955 2361368 Seven lakhs (700000) bitsProposed eight-directional PVD type 2 3722 2603604Five-directional PVD+MLSB type 1 [28] 3762 2399853
Eight lakhs and forty thousand (840000) bitsFive-directional PVD+MLSB type 2 [28] 3623 2514038Eight-directional PVD+MLSB type 1 [28] 3749 2394086Eight-directional PVD+MLSB type 2 [28] 3560 2603604
the proposed technique (type 1) Similarly the results of thistechnique (type 2) are as good as the results of the proposedtechnique (type 2)
5 Security Analysis
The LSB substitution techniques suffer from RS analysis andPVD techniques suffer from PDH analysis The proposedtechnique uses the concepts like modified LSB substitutionandPVD so itmust be analyzed by both RS analysis andPDHanalysis
The PDH analysis is performed by calculating the differ-ence value between every pair of pixels A pair comprises twoconsecutive pixels This difference values will be from -255 to+255 including 0 The frequency of each of these differencevalues is counted A graph is plotted with the pixel differencevalue on X-axis and frequency on Y-axis The curve obtainedis called the PDHThePDHof original imagewill be a smoothcurve If the PDH of stego image is also a smooth curve thensteganography is not detected If the PDH of stego imageshows step effects then steganography is identified
Figure 7 is the PDH analysis of type 1 and Figure 8 isthe PDH analysis of type 2 of the proposed technique Theanalysis is carried over all the eight test images For eachimage the PDH of the original image is represented by solidline and the PDH of the stego image is represented by dottedlineThe solid line curves will be obviously free of step effectsas they are of the original images The dotted line curves in
all the sixteen cases do not show any step effects This justifiesthat the proposed technique is resistant to PDH analysis
The RS analysis is based on statistical measures It is donein the following manner Define a function F
1 2nlarrrarr2n+1
It defines two transformations (i) from value 2n to value2n + 1 and (ii) from value 2n + 1 to value 2n Similarlydefine another function F
minus1 2n larrrarr2n -1 It defines other
two transformations (i) from value 2n to value 2n ndash 1 and(ii) from value 2n - 1 to value 2n The image say M isdivided into a number of equal size blocks Suppose such ablock is G whose pixels are X
1X2X3 Xn Then use the
function f(X1X2X3 Xn)=sumnminus1i=1 |Xi+1minusXi| tomeasure the
smoothness of G Then apply F1to all the blocks of M and
define the two parameters Rm and Sm as in (22) and (23)Similarly apply F
minus1to all the blocks of M and define the two
parameters Rminusm and S
minusm as in (24) and (25)
Rm
= No of blocks satisfying the condition f (F1(G)) gt f (G)
Total number of blocks
(22)
Sm
= No of blocks satisfying the condition f (F1(G)) lt f (G)
Total number of blocks
(23)
Rminusm
= No of blocks satisfying the condition f (Fminus1(G)) gt f (G)
Total number of blocks
(24)
Advances in Multimedia 9
0
1
2
Occ
urre
nces
3 times104
Lena coverLena stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3 times104
Baboon coverBaboon stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
Tiffany coverTiffany stego
times104
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Pot coverPot stego
times104
Figure 7 PDH analysis for type 1 of the proposed technique
10 Advances in Multimedia
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Lena coverLena stego
times104
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Baboon coverBaboon stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Tiffany coverTiffany stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Pot coverPot stego
times104
Figure 8 PDH analysis for type 2 of the proposed technique
Advances in Multimedia 11
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(a) Type 1 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(b) Type 1 Baboon
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(c) Type 2 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(d) Type 2 Baboon
Figure 9 RS analysis for type 1 and type 2 of the proposed technique
Sminusm
= No of blocks satisfying the condition f (Fminus1(G)) lt f (G)
Total number of blocks
(25)
RS analysis is performed by using these four parametersIf the condition R
119898asymp Rminus119898gt Sm asymp S
minusm is true then RSanalysis fails to detect the steganography technique But if the
condition Rminus119898
- Sminusm gt Rm - Sm is true then the RS analysis
succeeds in detecting the steganography techniqueFigure 9 represents RS analysis for the proposed tech-
nique Figures 9(a) and 9(b) stand for RS analysis over Lenaand Baboon images respectively of type 1 Similarly Figures9(c) and 9(d) stand for RS analysis over Lena and Baboonimages respectively of type 2 In all the four cases the
12 Advances in Multimedia
condition R119898asymp Rminus119898gt Sm asymp Sminusm is true so we can conclude
that RS analysis could not detect the proposed techniqueLena image has more smooth regions and Baboon imagehas more edge regions all the remaining six images are inbetween these two in terms of smoothness So it is sufficientto draw a conclusion based on these two images
6 Conclusion
The traditional PVD steganography techniques follow a staticrange table Due to this some undesired steps are introducedin pixel difference histograms of the stego images This stepeffect could be avoided by exploiting edges in eight differentdirections This paper proposes a steganography techniqueusing a judicious combination of LSB substitution and eight-directional PVD The LSB substitution is performed only inone pixel out of nine pixels in a block so RS analysis cannotdetect it By using the edges in multiple directions the PDHanalysis cannot detect itThere exist two types of the proposedtechnique with regard to two different quantization tablesThe type 1 provides higher PSNR and the type 2 provideshigher hiding capacity This proposed technique performsbetter in terms of hiding capacity and PSNR over othercompeting PVD techniques
Conflicts of Interest
The author declares that there are no conflicts of interest
References
[1] D-CWu andW-H Tsai ldquoA steganographicmethod for imagesby pixel-value differencingrdquo Pattern Recognition Letters vol 24no 9-10 pp 1613ndash1626 2003
[2] K C Chang C P Chang P S Huang and T M Tu ldquoAnovel image steganographic method using tri-way pixel-valuedifferencingrdquo Journal of Multimedia vol 3 no 2 pp 37ndash442008
[3] Y-P Lee J-C Lee W-K Chen K-C Chang I-J Su and C-P Chang ldquoHigh-payload image hiding with quality recoveryusing tri-way pixel-value differencingrdquo Information Sciencesvol 191 pp 214ndash225 2012
[4] C-C Chang and H-W Tseng ldquoA steganographic method fordigital images using side matchrdquo Pattern Recognition Lettersvol 25 no 12 pp 1431ndash1437 2004
[5] C-H Yang C-Y Weng H-K Tso and S-J Wang ldquoA datahiding scheme using the varieties of pixel-value differencing inmultimedia imagesrdquo The Journal of Systems and Software vol84 no 4 pp 669ndash678 2011
[6] W Hong T-S Chen and C-W Luo ldquoData embedding usingpixel value differencing and diamond encoding with multiple-base notational systemrdquo The Journal of Systems and Softwarevol 85 no 5 pp 1166ndash1175 2012
[7] H C Wu N I Wu C S Tsai and M-S Hwang ldquoImagesteganographic scheme based on pixel-value differencing andLSB replacementmethodsrdquo IEE ProceedingsmdashVision Image andSignal Processing vol 152 no 5 pp 611ndash615 2005
[8] C-H Yang C-Y Weng S-J Wang and H-M Sun ldquoVariedPVD + LSB evading detection programs to spatial domain in
data embedding systemsrdquo The Journal of Systems and Softwarevol 83 no 10 pp 1635ndash1643 2010
[9] J Chen ldquoA PVD-based data hiding method with histogrampreserving using pixel pair matchingrdquo Signal Processing ImageCommunication vol 29 no 3 pp 375ndash384 2014
[10] MKhodaei andK Faez ldquoNew adaptive steganographicmethodusing least-significant-bit substitution and pixel-value differ-encingrdquo IET Image Processing vol 6 no 6 pp 677ndash686 2012
[11] G Swain ldquoA steganographic method combining LSB substitu-tion and PVD in a blockrdquo Procedia Computer Science vol 85pp 39ndash44 2016
[12] X Liao Q Wen and J Zhang ldquoA steganographic methodfor digital images with four-pixel differencing and modifiedLSB substitutionrdquo Journal of Visual Communication and ImageRepresentation vol 22 no 1 pp 1ndash8 2011
[13] G Swain ldquoDigital image steganography using nine-pixel dif-ferencing and modified LSB substitutionrdquo Indian Journal ofScience and Technology vol 7 no 9 pp 1444ndash1450 2014httpwwwindjstorgindexphpindjstarticle
[14] W Luo F Huang and J Huang ldquoA more secure steganographybased on adaptive pixel-value differencing schemerdquoMultimediaTools and Applications vol 52 no 2-3 pp 407ndash430 2011
[15] G Swain ldquoAdaptive pixel value differencing steganographyusing both vertical and horizontal edgesrdquoMultimedia Tools andApplications vol 75 no 21 pp 13541ndash13556 2016
[16] S Chakraborty A S Jalal and C Bhatnagar ldquoLSB based nonblind predictive edge adaptive image steganographyrdquoMultime-dia Tools and Applications vol 76 no 6 pp 7973ndash7987 2017
[17] C Balasubramanian S Selvakumar and S Geetha ldquoHighpayload image steganography with reduced distortion usingoctonary pixel pairing schemerdquoMultimedia Tools and Applica-tions vol 73 no 3 pp 2223ndash2245 2014
[18] A Pradhan K R Sekhar and G Swain ldquoDigital imagesteganography based on seven way pixel value differencingrdquoIndian Journal of Science and Technology vol 9 no 37 pp 1ndash112016
[19] K A Darabkh A K Al-Dhamari and I F Jafar ldquoA newsteganographic algorithm based on multi directional PVD andmodified LSBrdquo Information Technology and Control vol 46 no1 pp 16ndash36 2017
[20] A K Gulve and M S Joshi ldquoA high capacity secured imagesteganography method with five pixel pair differencing andLSB substitutionrdquo International Journal of Image Graphics andSignal Processing vol 7 no 5 pp 66ndash74 2015
[21] X Zhang and S Wang ldquoEfficient steganographic embeddingby exploiting modification directionrdquo IEEE CommunicationsLetters vol 10 no 11 pp 781ndash783 2006
[22] C Kim ldquoData hiding by an improved exploiting modificationdirectionrdquoMultimedia Tools and Applications vol 69 no 3 pp569ndash584 2014
[23] S Y Shen and L H Huang ldquoA data hiding scheme usingpixel value differencing and improving exploiting modificationdirectionsrdquo Computers amp Security vol 48 pp 131ndash141 2015
[24] T D Nguyen S Arch-Int and N Arch-Int ldquoAn adaptivemulti bit-plane image steganography using block data-hidingrdquoMultimediaTools andApplications vol 75 no 14 pp 8319ndash83452016
[25] A Soria-Lorente and S Berres ldquoA secure steganographicalgorithm based on frequency domain for the transmission ofhidden informationrdquo Security and Communication Networksvol 2017 Article ID 5397028 14 pages 2017
Advances in Multimedia 13
[26] A Pradhan K R Sekhar and G Swain ldquoAdaptive PVDsteganography using horizontal vertical and diagonal edges insix-pixel blocksrdquo Security and Communication Networks vol2017 Article ID 1924618 13 pages 2017
[27] A Pradhan K R ekhar andG Swain ldquoDigital image steganog-raphy using LSB substitution PVD and EMDrdquo MathematicalProblems in Engineering 2018
[28] G Swain ldquoHigh capacity image steganography using modifiedLSB substitution and PVD against pixel difference histogramanalysisrdquo Security and Communication Networks 2018
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Advances in Multimedia 7
Table 3 Results of Pradhan et alrsquos seven-way PVD technique [18]
Images 512times512times3 Seven-way PVD [18] Seven-way+ one-way PVD [18]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4237 1795551 09996 228 4240 1803635 09996 229House 3897 1972223 09986 250 3898 1977403 09986 251Boat 3789 1972086 09988 250 3791 1991005 09988 253Jet 4209 1906254 09989 242 4198 1909595 09989 242Peppers 4042 1778072 09993 226 4020 1806166 09992 229Tiffany 4123 1400756 09986 178 4130 1450799 09986 184Baboon 3379 2226806 09957 283 3377 2243218 09957 285Lena 4173 1896662 09993 241 4173 1901149 09993 241Average 3981 1868551 09986 237 3978 1885371 09985 239
Table 4 Results of Khodaei and Faezrsquos technique
Images 512times512times3 Khodaei and Faez type 1 [10] Khodaei and Faez type 2 [10]PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4121 2366001 09995 300 3783 2387494 09990 303House 3916 2387183 09986 303 3875 2470824 09985 314Boat 4005 2391994 09993 304 3749 2504613 09987 318Jet 4089 2374048 09986 301 4064 2418419 09985 307Peppers 3932 2372858 09991 301 3791 2435223 09987 309Tiffany 3973 2372396 09980 301 4025 2416944 09982 307Baboon 3777 2443361 09982 310 3449 2662080 09963 338Lena 4278 2375248 09995 302 4125 2434603 09993 309Average 4011 2385386 09988 303 3857 2466275 09984 313
Equation (16) is used to compute itThemaximum value of Qcan be 1 if original image p and stego image q are the same
Q =4120590xyp q
(1205902x + 1205902y) [(p)2 + (q)2] (16)
where p stands for the mean pixel value of the original imageq stands for the mean pixel value of the stego image1205902x standsfor the standard deviation of pixel values of original image1205902y stands for the standard deviation of pixel values of stegoimage and 120590xy is the covariance between original image andstego image Equations (17) (18) (19) (20) and (21) are usedto calculate these values
p = 1m times n
msumi=1
nsumj=1pij (17)
q = 1m times n
msumi=1
nsumj=1qij (18)
1205902x =1
m times n minus 1msumi=1
nsumj=1(pij minus p)
2
(19)
1205902y =1
m times n minus 1msumi=1
nsumj=1(qij minus q)
2
(20)
120590xy =1
m times n minus 1msumi=1
nsumj=1(pij minus p) (qij minus q) (21)
Table 3 notes the results of seven-way PVD [18] whichis an extension of Wu and Tsairsquos [1] technique Table 4 notes
the results of Khodaei and Faezrsquos technique [10] and Table 5notes the results of the proposed technique The PSNR andcapacity of the proposed technique (both type 1 and type2) are better than seven-way PVD The bit rate and Q arealso better Although as compared to Khodaei and Faezrsquostechnique the PSNR is not improved but the hiding capacityis improved The major advantage over Khodaei and Faezrsquostechnique is that the proposed technique is resistant to PDHanalysis but Khodaei and Faezrsquos technique is detectable byPDH analysis This is made possible by exploiting the edgesin eight directions
Now let us observe the results of proposed technique andcompare type 1 and type 2 We can observe that the capacityis higher in type 2 as compared to type 1 This is because ofthe 4 LSBs substitution in type 2 at the central pixel ie the tvalue is 4 On the other hand the PSNR of type 1 is better thanthat of type 2
Table 6 represents the comparison of the proposedtechnique with the adaptive PVD technique in [26] andhybrid techniques in [27 28] The comparison parametersare average PSNR and average hiding capacity values Thetwo variants of adaptive PVD technique possess lesser hidingcapacity as compared to the proposed eight-directional PVDAs the hiding capacity of adaptive PVD [26] is very littleso PSNR is larger The hybrid technique proposed in [27] isa combination of PVD LSB and EMD The hiding capacityof proposed technique is greater than all the four variantsof the hybrid technique proposed in [27] The technique in[28] uses PVD with modified LSB (MLSB) substitution Theresults of this technique (type 1) are as good as the results of
8 Advances in Multimedia
Table 5 Results of proposed eight-directional PVD technique
Images 512times512times3 Proposed type 1 (t=3 Range Table 1) Proposed type 2 (t=4 Range Table 2)PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4115 2354240 09996 299 3711 2475977 09988 315House 3969 2358575 09988 300 3839 2625804 09984 334Boat 3725 2364685 09987 301 3408 2659795 09973 338Jet 4221 2356828 09989 300 4038 2538801 09985 323Peppers 3781 2356645 09988 300 3469 2544392 09975 324Tiffany 4146 2353505 09987 299 4053 2511139 09984 319Baboon 3426 2392573 09961 304 3209 2939376 09937 374Lena 4262 2353892 09995 299 4048 2533551 09992 322Average 3955 2361368 09986 300 3722 2603604 09977 331
Table 6 Comparison of the proposed technique with techniques in [26ndash28]
Technique PSNR Capacity Amount of hidden data in stego image to compute PSNR2times3 adaptive PVD [26] 5093 1445645 One lakh and forty thousand (140000) bits3times2 adaptive PVD [26] 5093 14356233 PVD +3 LSB +EMD type 1 [27] 4107 1686041
Seven lakhs (700000) bits3 PVD +3 LSB +EMD type 2 [27] 3895 17988347 PVD +3 LSB +EMD type 1 [27] 4028 17108887 PVD +3 LSB +EMD type 2 [27] 3726 1823282Proposed eight-directional PVD type 1 3955 2361368 Seven lakhs (700000) bitsProposed eight-directional PVD type 2 3722 2603604Five-directional PVD+MLSB type 1 [28] 3762 2399853
Eight lakhs and forty thousand (840000) bitsFive-directional PVD+MLSB type 2 [28] 3623 2514038Eight-directional PVD+MLSB type 1 [28] 3749 2394086Eight-directional PVD+MLSB type 2 [28] 3560 2603604
the proposed technique (type 1) Similarly the results of thistechnique (type 2) are as good as the results of the proposedtechnique (type 2)
5 Security Analysis
The LSB substitution techniques suffer from RS analysis andPVD techniques suffer from PDH analysis The proposedtechnique uses the concepts like modified LSB substitutionandPVD so itmust be analyzed by both RS analysis andPDHanalysis
The PDH analysis is performed by calculating the differ-ence value between every pair of pixels A pair comprises twoconsecutive pixels This difference values will be from -255 to+255 including 0 The frequency of each of these differencevalues is counted A graph is plotted with the pixel differencevalue on X-axis and frequency on Y-axis The curve obtainedis called the PDHThePDHof original imagewill be a smoothcurve If the PDH of stego image is also a smooth curve thensteganography is not detected If the PDH of stego imageshows step effects then steganography is identified
Figure 7 is the PDH analysis of type 1 and Figure 8 isthe PDH analysis of type 2 of the proposed technique Theanalysis is carried over all the eight test images For eachimage the PDH of the original image is represented by solidline and the PDH of the stego image is represented by dottedlineThe solid line curves will be obviously free of step effectsas they are of the original images The dotted line curves in
all the sixteen cases do not show any step effects This justifiesthat the proposed technique is resistant to PDH analysis
The RS analysis is based on statistical measures It is donein the following manner Define a function F
1 2nlarrrarr2n+1
It defines two transformations (i) from value 2n to value2n + 1 and (ii) from value 2n + 1 to value 2n Similarlydefine another function F
minus1 2n larrrarr2n -1 It defines other
two transformations (i) from value 2n to value 2n ndash 1 and(ii) from value 2n - 1 to value 2n The image say M isdivided into a number of equal size blocks Suppose such ablock is G whose pixels are X
1X2X3 Xn Then use the
function f(X1X2X3 Xn)=sumnminus1i=1 |Xi+1minusXi| tomeasure the
smoothness of G Then apply F1to all the blocks of M and
define the two parameters Rm and Sm as in (22) and (23)Similarly apply F
minus1to all the blocks of M and define the two
parameters Rminusm and S
minusm as in (24) and (25)
Rm
= No of blocks satisfying the condition f (F1(G)) gt f (G)
Total number of blocks
(22)
Sm
= No of blocks satisfying the condition f (F1(G)) lt f (G)
Total number of blocks
(23)
Rminusm
= No of blocks satisfying the condition f (Fminus1(G)) gt f (G)
Total number of blocks
(24)
Advances in Multimedia 9
0
1
2
Occ
urre
nces
3 times104
Lena coverLena stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3 times104
Baboon coverBaboon stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
Tiffany coverTiffany stego
times104
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Pot coverPot stego
times104
Figure 7 PDH analysis for type 1 of the proposed technique
10 Advances in Multimedia
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Lena coverLena stego
times104
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Baboon coverBaboon stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Tiffany coverTiffany stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Pot coverPot stego
times104
Figure 8 PDH analysis for type 2 of the proposed technique
Advances in Multimedia 11
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(a) Type 1 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(b) Type 1 Baboon
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(c) Type 2 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(d) Type 2 Baboon
Figure 9 RS analysis for type 1 and type 2 of the proposed technique
Sminusm
= No of blocks satisfying the condition f (Fminus1(G)) lt f (G)
Total number of blocks
(25)
RS analysis is performed by using these four parametersIf the condition R
119898asymp Rminus119898gt Sm asymp S
minusm is true then RSanalysis fails to detect the steganography technique But if the
condition Rminus119898
- Sminusm gt Rm - Sm is true then the RS analysis
succeeds in detecting the steganography techniqueFigure 9 represents RS analysis for the proposed tech-
nique Figures 9(a) and 9(b) stand for RS analysis over Lenaand Baboon images respectively of type 1 Similarly Figures9(c) and 9(d) stand for RS analysis over Lena and Baboonimages respectively of type 2 In all the four cases the
12 Advances in Multimedia
condition R119898asymp Rminus119898gt Sm asymp Sminusm is true so we can conclude
that RS analysis could not detect the proposed techniqueLena image has more smooth regions and Baboon imagehas more edge regions all the remaining six images are inbetween these two in terms of smoothness So it is sufficientto draw a conclusion based on these two images
6 Conclusion
The traditional PVD steganography techniques follow a staticrange table Due to this some undesired steps are introducedin pixel difference histograms of the stego images This stepeffect could be avoided by exploiting edges in eight differentdirections This paper proposes a steganography techniqueusing a judicious combination of LSB substitution and eight-directional PVD The LSB substitution is performed only inone pixel out of nine pixels in a block so RS analysis cannotdetect it By using the edges in multiple directions the PDHanalysis cannot detect itThere exist two types of the proposedtechnique with regard to two different quantization tablesThe type 1 provides higher PSNR and the type 2 provideshigher hiding capacity This proposed technique performsbetter in terms of hiding capacity and PSNR over othercompeting PVD techniques
Conflicts of Interest
The author declares that there are no conflicts of interest
References
[1] D-CWu andW-H Tsai ldquoA steganographicmethod for imagesby pixel-value differencingrdquo Pattern Recognition Letters vol 24no 9-10 pp 1613ndash1626 2003
[2] K C Chang C P Chang P S Huang and T M Tu ldquoAnovel image steganographic method using tri-way pixel-valuedifferencingrdquo Journal of Multimedia vol 3 no 2 pp 37ndash442008
[3] Y-P Lee J-C Lee W-K Chen K-C Chang I-J Su and C-P Chang ldquoHigh-payload image hiding with quality recoveryusing tri-way pixel-value differencingrdquo Information Sciencesvol 191 pp 214ndash225 2012
[4] C-C Chang and H-W Tseng ldquoA steganographic method fordigital images using side matchrdquo Pattern Recognition Lettersvol 25 no 12 pp 1431ndash1437 2004
[5] C-H Yang C-Y Weng H-K Tso and S-J Wang ldquoA datahiding scheme using the varieties of pixel-value differencing inmultimedia imagesrdquo The Journal of Systems and Software vol84 no 4 pp 669ndash678 2011
[6] W Hong T-S Chen and C-W Luo ldquoData embedding usingpixel value differencing and diamond encoding with multiple-base notational systemrdquo The Journal of Systems and Softwarevol 85 no 5 pp 1166ndash1175 2012
[7] H C Wu N I Wu C S Tsai and M-S Hwang ldquoImagesteganographic scheme based on pixel-value differencing andLSB replacementmethodsrdquo IEE ProceedingsmdashVision Image andSignal Processing vol 152 no 5 pp 611ndash615 2005
[8] C-H Yang C-Y Weng S-J Wang and H-M Sun ldquoVariedPVD + LSB evading detection programs to spatial domain in
data embedding systemsrdquo The Journal of Systems and Softwarevol 83 no 10 pp 1635ndash1643 2010
[9] J Chen ldquoA PVD-based data hiding method with histogrampreserving using pixel pair matchingrdquo Signal Processing ImageCommunication vol 29 no 3 pp 375ndash384 2014
[10] MKhodaei andK Faez ldquoNew adaptive steganographicmethodusing least-significant-bit substitution and pixel-value differ-encingrdquo IET Image Processing vol 6 no 6 pp 677ndash686 2012
[11] G Swain ldquoA steganographic method combining LSB substitu-tion and PVD in a blockrdquo Procedia Computer Science vol 85pp 39ndash44 2016
[12] X Liao Q Wen and J Zhang ldquoA steganographic methodfor digital images with four-pixel differencing and modifiedLSB substitutionrdquo Journal of Visual Communication and ImageRepresentation vol 22 no 1 pp 1ndash8 2011
[13] G Swain ldquoDigital image steganography using nine-pixel dif-ferencing and modified LSB substitutionrdquo Indian Journal ofScience and Technology vol 7 no 9 pp 1444ndash1450 2014httpwwwindjstorgindexphpindjstarticle
[14] W Luo F Huang and J Huang ldquoA more secure steganographybased on adaptive pixel-value differencing schemerdquoMultimediaTools and Applications vol 52 no 2-3 pp 407ndash430 2011
[15] G Swain ldquoAdaptive pixel value differencing steganographyusing both vertical and horizontal edgesrdquoMultimedia Tools andApplications vol 75 no 21 pp 13541ndash13556 2016
[16] S Chakraborty A S Jalal and C Bhatnagar ldquoLSB based nonblind predictive edge adaptive image steganographyrdquoMultime-dia Tools and Applications vol 76 no 6 pp 7973ndash7987 2017
[17] C Balasubramanian S Selvakumar and S Geetha ldquoHighpayload image steganography with reduced distortion usingoctonary pixel pairing schemerdquoMultimedia Tools and Applica-tions vol 73 no 3 pp 2223ndash2245 2014
[18] A Pradhan K R Sekhar and G Swain ldquoDigital imagesteganography based on seven way pixel value differencingrdquoIndian Journal of Science and Technology vol 9 no 37 pp 1ndash112016
[19] K A Darabkh A K Al-Dhamari and I F Jafar ldquoA newsteganographic algorithm based on multi directional PVD andmodified LSBrdquo Information Technology and Control vol 46 no1 pp 16ndash36 2017
[20] A K Gulve and M S Joshi ldquoA high capacity secured imagesteganography method with five pixel pair differencing andLSB substitutionrdquo International Journal of Image Graphics andSignal Processing vol 7 no 5 pp 66ndash74 2015
[21] X Zhang and S Wang ldquoEfficient steganographic embeddingby exploiting modification directionrdquo IEEE CommunicationsLetters vol 10 no 11 pp 781ndash783 2006
[22] C Kim ldquoData hiding by an improved exploiting modificationdirectionrdquoMultimedia Tools and Applications vol 69 no 3 pp569ndash584 2014
[23] S Y Shen and L H Huang ldquoA data hiding scheme usingpixel value differencing and improving exploiting modificationdirectionsrdquo Computers amp Security vol 48 pp 131ndash141 2015
[24] T D Nguyen S Arch-Int and N Arch-Int ldquoAn adaptivemulti bit-plane image steganography using block data-hidingrdquoMultimediaTools andApplications vol 75 no 14 pp 8319ndash83452016
[25] A Soria-Lorente and S Berres ldquoA secure steganographicalgorithm based on frequency domain for the transmission ofhidden informationrdquo Security and Communication Networksvol 2017 Article ID 5397028 14 pages 2017
Advances in Multimedia 13
[26] A Pradhan K R Sekhar and G Swain ldquoAdaptive PVDsteganography using horizontal vertical and diagonal edges insix-pixel blocksrdquo Security and Communication Networks vol2017 Article ID 1924618 13 pages 2017
[27] A Pradhan K R ekhar andG Swain ldquoDigital image steganog-raphy using LSB substitution PVD and EMDrdquo MathematicalProblems in Engineering 2018
[28] G Swain ldquoHigh capacity image steganography using modifiedLSB substitution and PVD against pixel difference histogramanalysisrdquo Security and Communication Networks 2018
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
8 Advances in Multimedia
Table 5 Results of proposed eight-directional PVD technique
Images 512times512times3 Proposed type 1 (t=3 Range Table 1) Proposed type 2 (t=4 Range Table 2)PSNR Capacity Q BPB PSNR Capacity Q BPB
Pot 4115 2354240 09996 299 3711 2475977 09988 315House 3969 2358575 09988 300 3839 2625804 09984 334Boat 3725 2364685 09987 301 3408 2659795 09973 338Jet 4221 2356828 09989 300 4038 2538801 09985 323Peppers 3781 2356645 09988 300 3469 2544392 09975 324Tiffany 4146 2353505 09987 299 4053 2511139 09984 319Baboon 3426 2392573 09961 304 3209 2939376 09937 374Lena 4262 2353892 09995 299 4048 2533551 09992 322Average 3955 2361368 09986 300 3722 2603604 09977 331
Table 6 Comparison of the proposed technique with techniques in [26ndash28]
Technique PSNR Capacity Amount of hidden data in stego image to compute PSNR2times3 adaptive PVD [26] 5093 1445645 One lakh and forty thousand (140000) bits3times2 adaptive PVD [26] 5093 14356233 PVD +3 LSB +EMD type 1 [27] 4107 1686041
Seven lakhs (700000) bits3 PVD +3 LSB +EMD type 2 [27] 3895 17988347 PVD +3 LSB +EMD type 1 [27] 4028 17108887 PVD +3 LSB +EMD type 2 [27] 3726 1823282Proposed eight-directional PVD type 1 3955 2361368 Seven lakhs (700000) bitsProposed eight-directional PVD type 2 3722 2603604Five-directional PVD+MLSB type 1 [28] 3762 2399853
Eight lakhs and forty thousand (840000) bitsFive-directional PVD+MLSB type 2 [28] 3623 2514038Eight-directional PVD+MLSB type 1 [28] 3749 2394086Eight-directional PVD+MLSB type 2 [28] 3560 2603604
the proposed technique (type 1) Similarly the results of thistechnique (type 2) are as good as the results of the proposedtechnique (type 2)
5 Security Analysis
The LSB substitution techniques suffer from RS analysis andPVD techniques suffer from PDH analysis The proposedtechnique uses the concepts like modified LSB substitutionandPVD so itmust be analyzed by both RS analysis andPDHanalysis
The PDH analysis is performed by calculating the differ-ence value between every pair of pixels A pair comprises twoconsecutive pixels This difference values will be from -255 to+255 including 0 The frequency of each of these differencevalues is counted A graph is plotted with the pixel differencevalue on X-axis and frequency on Y-axis The curve obtainedis called the PDHThePDHof original imagewill be a smoothcurve If the PDH of stego image is also a smooth curve thensteganography is not detected If the PDH of stego imageshows step effects then steganography is identified
Figure 7 is the PDH analysis of type 1 and Figure 8 isthe PDH analysis of type 2 of the proposed technique Theanalysis is carried over all the eight test images For eachimage the PDH of the original image is represented by solidline and the PDH of the stego image is represented by dottedlineThe solid line curves will be obviously free of step effectsas they are of the original images The dotted line curves in
all the sixteen cases do not show any step effects This justifiesthat the proposed technique is resistant to PDH analysis
The RS analysis is based on statistical measures It is donein the following manner Define a function F
1 2nlarrrarr2n+1
It defines two transformations (i) from value 2n to value2n + 1 and (ii) from value 2n + 1 to value 2n Similarlydefine another function F
minus1 2n larrrarr2n -1 It defines other
two transformations (i) from value 2n to value 2n ndash 1 and(ii) from value 2n - 1 to value 2n The image say M isdivided into a number of equal size blocks Suppose such ablock is G whose pixels are X
1X2X3 Xn Then use the
function f(X1X2X3 Xn)=sumnminus1i=1 |Xi+1minusXi| tomeasure the
smoothness of G Then apply F1to all the blocks of M and
define the two parameters Rm and Sm as in (22) and (23)Similarly apply F
minus1to all the blocks of M and define the two
parameters Rminusm and S
minusm as in (24) and (25)
Rm
= No of blocks satisfying the condition f (F1(G)) gt f (G)
Total number of blocks
(22)
Sm
= No of blocks satisfying the condition f (F1(G)) lt f (G)
Total number of blocks
(23)
Rminusm
= No of blocks satisfying the condition f (Fminus1(G)) gt f (G)
Total number of blocks
(24)
Advances in Multimedia 9
0
1
2
Occ
urre
nces
3 times104
Lena coverLena stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3 times104
Baboon coverBaboon stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
Tiffany coverTiffany stego
times104
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Pot coverPot stego
times104
Figure 7 PDH analysis for type 1 of the proposed technique
10 Advances in Multimedia
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Lena coverLena stego
times104
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Baboon coverBaboon stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Tiffany coverTiffany stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Pot coverPot stego
times104
Figure 8 PDH analysis for type 2 of the proposed technique
Advances in Multimedia 11
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(a) Type 1 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(b) Type 1 Baboon
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(c) Type 2 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(d) Type 2 Baboon
Figure 9 RS analysis for type 1 and type 2 of the proposed technique
Sminusm
= No of blocks satisfying the condition f (Fminus1(G)) lt f (G)
Total number of blocks
(25)
RS analysis is performed by using these four parametersIf the condition R
119898asymp Rminus119898gt Sm asymp S
minusm is true then RSanalysis fails to detect the steganography technique But if the
condition Rminus119898
- Sminusm gt Rm - Sm is true then the RS analysis
succeeds in detecting the steganography techniqueFigure 9 represents RS analysis for the proposed tech-
nique Figures 9(a) and 9(b) stand for RS analysis over Lenaand Baboon images respectively of type 1 Similarly Figures9(c) and 9(d) stand for RS analysis over Lena and Baboonimages respectively of type 2 In all the four cases the
12 Advances in Multimedia
condition R119898asymp Rminus119898gt Sm asymp Sminusm is true so we can conclude
that RS analysis could not detect the proposed techniqueLena image has more smooth regions and Baboon imagehas more edge regions all the remaining six images are inbetween these two in terms of smoothness So it is sufficientto draw a conclusion based on these two images
6 Conclusion
The traditional PVD steganography techniques follow a staticrange table Due to this some undesired steps are introducedin pixel difference histograms of the stego images This stepeffect could be avoided by exploiting edges in eight differentdirections This paper proposes a steganography techniqueusing a judicious combination of LSB substitution and eight-directional PVD The LSB substitution is performed only inone pixel out of nine pixels in a block so RS analysis cannotdetect it By using the edges in multiple directions the PDHanalysis cannot detect itThere exist two types of the proposedtechnique with regard to two different quantization tablesThe type 1 provides higher PSNR and the type 2 provideshigher hiding capacity This proposed technique performsbetter in terms of hiding capacity and PSNR over othercompeting PVD techniques
Conflicts of Interest
The author declares that there are no conflicts of interest
References
[1] D-CWu andW-H Tsai ldquoA steganographicmethod for imagesby pixel-value differencingrdquo Pattern Recognition Letters vol 24no 9-10 pp 1613ndash1626 2003
[2] K C Chang C P Chang P S Huang and T M Tu ldquoAnovel image steganographic method using tri-way pixel-valuedifferencingrdquo Journal of Multimedia vol 3 no 2 pp 37ndash442008
[3] Y-P Lee J-C Lee W-K Chen K-C Chang I-J Su and C-P Chang ldquoHigh-payload image hiding with quality recoveryusing tri-way pixel-value differencingrdquo Information Sciencesvol 191 pp 214ndash225 2012
[4] C-C Chang and H-W Tseng ldquoA steganographic method fordigital images using side matchrdquo Pattern Recognition Lettersvol 25 no 12 pp 1431ndash1437 2004
[5] C-H Yang C-Y Weng H-K Tso and S-J Wang ldquoA datahiding scheme using the varieties of pixel-value differencing inmultimedia imagesrdquo The Journal of Systems and Software vol84 no 4 pp 669ndash678 2011
[6] W Hong T-S Chen and C-W Luo ldquoData embedding usingpixel value differencing and diamond encoding with multiple-base notational systemrdquo The Journal of Systems and Softwarevol 85 no 5 pp 1166ndash1175 2012
[7] H C Wu N I Wu C S Tsai and M-S Hwang ldquoImagesteganographic scheme based on pixel-value differencing andLSB replacementmethodsrdquo IEE ProceedingsmdashVision Image andSignal Processing vol 152 no 5 pp 611ndash615 2005
[8] C-H Yang C-Y Weng S-J Wang and H-M Sun ldquoVariedPVD + LSB evading detection programs to spatial domain in
data embedding systemsrdquo The Journal of Systems and Softwarevol 83 no 10 pp 1635ndash1643 2010
[9] J Chen ldquoA PVD-based data hiding method with histogrampreserving using pixel pair matchingrdquo Signal Processing ImageCommunication vol 29 no 3 pp 375ndash384 2014
[10] MKhodaei andK Faez ldquoNew adaptive steganographicmethodusing least-significant-bit substitution and pixel-value differ-encingrdquo IET Image Processing vol 6 no 6 pp 677ndash686 2012
[11] G Swain ldquoA steganographic method combining LSB substitu-tion and PVD in a blockrdquo Procedia Computer Science vol 85pp 39ndash44 2016
[12] X Liao Q Wen and J Zhang ldquoA steganographic methodfor digital images with four-pixel differencing and modifiedLSB substitutionrdquo Journal of Visual Communication and ImageRepresentation vol 22 no 1 pp 1ndash8 2011
[13] G Swain ldquoDigital image steganography using nine-pixel dif-ferencing and modified LSB substitutionrdquo Indian Journal ofScience and Technology vol 7 no 9 pp 1444ndash1450 2014httpwwwindjstorgindexphpindjstarticle
[14] W Luo F Huang and J Huang ldquoA more secure steganographybased on adaptive pixel-value differencing schemerdquoMultimediaTools and Applications vol 52 no 2-3 pp 407ndash430 2011
[15] G Swain ldquoAdaptive pixel value differencing steganographyusing both vertical and horizontal edgesrdquoMultimedia Tools andApplications vol 75 no 21 pp 13541ndash13556 2016
[16] S Chakraborty A S Jalal and C Bhatnagar ldquoLSB based nonblind predictive edge adaptive image steganographyrdquoMultime-dia Tools and Applications vol 76 no 6 pp 7973ndash7987 2017
[17] C Balasubramanian S Selvakumar and S Geetha ldquoHighpayload image steganography with reduced distortion usingoctonary pixel pairing schemerdquoMultimedia Tools and Applica-tions vol 73 no 3 pp 2223ndash2245 2014
[18] A Pradhan K R Sekhar and G Swain ldquoDigital imagesteganography based on seven way pixel value differencingrdquoIndian Journal of Science and Technology vol 9 no 37 pp 1ndash112016
[19] K A Darabkh A K Al-Dhamari and I F Jafar ldquoA newsteganographic algorithm based on multi directional PVD andmodified LSBrdquo Information Technology and Control vol 46 no1 pp 16ndash36 2017
[20] A K Gulve and M S Joshi ldquoA high capacity secured imagesteganography method with five pixel pair differencing andLSB substitutionrdquo International Journal of Image Graphics andSignal Processing vol 7 no 5 pp 66ndash74 2015
[21] X Zhang and S Wang ldquoEfficient steganographic embeddingby exploiting modification directionrdquo IEEE CommunicationsLetters vol 10 no 11 pp 781ndash783 2006
[22] C Kim ldquoData hiding by an improved exploiting modificationdirectionrdquoMultimedia Tools and Applications vol 69 no 3 pp569ndash584 2014
[23] S Y Shen and L H Huang ldquoA data hiding scheme usingpixel value differencing and improving exploiting modificationdirectionsrdquo Computers amp Security vol 48 pp 131ndash141 2015
[24] T D Nguyen S Arch-Int and N Arch-Int ldquoAn adaptivemulti bit-plane image steganography using block data-hidingrdquoMultimediaTools andApplications vol 75 no 14 pp 8319ndash83452016
[25] A Soria-Lorente and S Berres ldquoA secure steganographicalgorithm based on frequency domain for the transmission ofhidden informationrdquo Security and Communication Networksvol 2017 Article ID 5397028 14 pages 2017
Advances in Multimedia 13
[26] A Pradhan K R Sekhar and G Swain ldquoAdaptive PVDsteganography using horizontal vertical and diagonal edges insix-pixel blocksrdquo Security and Communication Networks vol2017 Article ID 1924618 13 pages 2017
[27] A Pradhan K R ekhar andG Swain ldquoDigital image steganog-raphy using LSB substitution PVD and EMDrdquo MathematicalProblems in Engineering 2018
[28] G Swain ldquoHigh capacity image steganography using modifiedLSB substitution and PVD against pixel difference histogramanalysisrdquo Security and Communication Networks 2018
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Advances in Multimedia 9
0
1
2
Occ
urre
nces
3 times104
Lena coverLena stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3 times104
Baboon coverBaboon stego
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
Tiffany coverTiffany stego
times104
minus20 0 20 40minus40
Pixel difference
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
1
2
Occ
urre
nces
3
0minus20minus40 4020
Pixel difference0
Pot coverPot stego
times104
Figure 7 PDH analysis for type 1 of the proposed technique
10 Advances in Multimedia
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Lena coverLena stego
times104
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Baboon coverBaboon stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Tiffany coverTiffany stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Pot coverPot stego
times104
Figure 8 PDH analysis for type 2 of the proposed technique
Advances in Multimedia 11
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(a) Type 1 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(b) Type 1 Baboon
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(c) Type 2 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(d) Type 2 Baboon
Figure 9 RS analysis for type 1 and type 2 of the proposed technique
Sminusm
= No of blocks satisfying the condition f (Fminus1(G)) lt f (G)
Total number of blocks
(25)
RS analysis is performed by using these four parametersIf the condition R
119898asymp Rminus119898gt Sm asymp S
minusm is true then RSanalysis fails to detect the steganography technique But if the
condition Rminus119898
- Sminusm gt Rm - Sm is true then the RS analysis
succeeds in detecting the steganography techniqueFigure 9 represents RS analysis for the proposed tech-
nique Figures 9(a) and 9(b) stand for RS analysis over Lenaand Baboon images respectively of type 1 Similarly Figures9(c) and 9(d) stand for RS analysis over Lena and Baboonimages respectively of type 2 In all the four cases the
12 Advances in Multimedia
condition R119898asymp Rminus119898gt Sm asymp Sminusm is true so we can conclude
that RS analysis could not detect the proposed techniqueLena image has more smooth regions and Baboon imagehas more edge regions all the remaining six images are inbetween these two in terms of smoothness So it is sufficientto draw a conclusion based on these two images
6 Conclusion
The traditional PVD steganography techniques follow a staticrange table Due to this some undesired steps are introducedin pixel difference histograms of the stego images This stepeffect could be avoided by exploiting edges in eight differentdirections This paper proposes a steganography techniqueusing a judicious combination of LSB substitution and eight-directional PVD The LSB substitution is performed only inone pixel out of nine pixels in a block so RS analysis cannotdetect it By using the edges in multiple directions the PDHanalysis cannot detect itThere exist two types of the proposedtechnique with regard to two different quantization tablesThe type 1 provides higher PSNR and the type 2 provideshigher hiding capacity This proposed technique performsbetter in terms of hiding capacity and PSNR over othercompeting PVD techniques
Conflicts of Interest
The author declares that there are no conflicts of interest
References
[1] D-CWu andW-H Tsai ldquoA steganographicmethod for imagesby pixel-value differencingrdquo Pattern Recognition Letters vol 24no 9-10 pp 1613ndash1626 2003
[2] K C Chang C P Chang P S Huang and T M Tu ldquoAnovel image steganographic method using tri-way pixel-valuedifferencingrdquo Journal of Multimedia vol 3 no 2 pp 37ndash442008
[3] Y-P Lee J-C Lee W-K Chen K-C Chang I-J Su and C-P Chang ldquoHigh-payload image hiding with quality recoveryusing tri-way pixel-value differencingrdquo Information Sciencesvol 191 pp 214ndash225 2012
[4] C-C Chang and H-W Tseng ldquoA steganographic method fordigital images using side matchrdquo Pattern Recognition Lettersvol 25 no 12 pp 1431ndash1437 2004
[5] C-H Yang C-Y Weng H-K Tso and S-J Wang ldquoA datahiding scheme using the varieties of pixel-value differencing inmultimedia imagesrdquo The Journal of Systems and Software vol84 no 4 pp 669ndash678 2011
[6] W Hong T-S Chen and C-W Luo ldquoData embedding usingpixel value differencing and diamond encoding with multiple-base notational systemrdquo The Journal of Systems and Softwarevol 85 no 5 pp 1166ndash1175 2012
[7] H C Wu N I Wu C S Tsai and M-S Hwang ldquoImagesteganographic scheme based on pixel-value differencing andLSB replacementmethodsrdquo IEE ProceedingsmdashVision Image andSignal Processing vol 152 no 5 pp 611ndash615 2005
[8] C-H Yang C-Y Weng S-J Wang and H-M Sun ldquoVariedPVD + LSB evading detection programs to spatial domain in
data embedding systemsrdquo The Journal of Systems and Softwarevol 83 no 10 pp 1635ndash1643 2010
[9] J Chen ldquoA PVD-based data hiding method with histogrampreserving using pixel pair matchingrdquo Signal Processing ImageCommunication vol 29 no 3 pp 375ndash384 2014
[10] MKhodaei andK Faez ldquoNew adaptive steganographicmethodusing least-significant-bit substitution and pixel-value differ-encingrdquo IET Image Processing vol 6 no 6 pp 677ndash686 2012
[11] G Swain ldquoA steganographic method combining LSB substitu-tion and PVD in a blockrdquo Procedia Computer Science vol 85pp 39ndash44 2016
[12] X Liao Q Wen and J Zhang ldquoA steganographic methodfor digital images with four-pixel differencing and modifiedLSB substitutionrdquo Journal of Visual Communication and ImageRepresentation vol 22 no 1 pp 1ndash8 2011
[13] G Swain ldquoDigital image steganography using nine-pixel dif-ferencing and modified LSB substitutionrdquo Indian Journal ofScience and Technology vol 7 no 9 pp 1444ndash1450 2014httpwwwindjstorgindexphpindjstarticle
[14] W Luo F Huang and J Huang ldquoA more secure steganographybased on adaptive pixel-value differencing schemerdquoMultimediaTools and Applications vol 52 no 2-3 pp 407ndash430 2011
[15] G Swain ldquoAdaptive pixel value differencing steganographyusing both vertical and horizontal edgesrdquoMultimedia Tools andApplications vol 75 no 21 pp 13541ndash13556 2016
[16] S Chakraborty A S Jalal and C Bhatnagar ldquoLSB based nonblind predictive edge adaptive image steganographyrdquoMultime-dia Tools and Applications vol 76 no 6 pp 7973ndash7987 2017
[17] C Balasubramanian S Selvakumar and S Geetha ldquoHighpayload image steganography with reduced distortion usingoctonary pixel pairing schemerdquoMultimedia Tools and Applica-tions vol 73 no 3 pp 2223ndash2245 2014
[18] A Pradhan K R Sekhar and G Swain ldquoDigital imagesteganography based on seven way pixel value differencingrdquoIndian Journal of Science and Technology vol 9 no 37 pp 1ndash112016
[19] K A Darabkh A K Al-Dhamari and I F Jafar ldquoA newsteganographic algorithm based on multi directional PVD andmodified LSBrdquo Information Technology and Control vol 46 no1 pp 16ndash36 2017
[20] A K Gulve and M S Joshi ldquoA high capacity secured imagesteganography method with five pixel pair differencing andLSB substitutionrdquo International Journal of Image Graphics andSignal Processing vol 7 no 5 pp 66ndash74 2015
[21] X Zhang and S Wang ldquoEfficient steganographic embeddingby exploiting modification directionrdquo IEEE CommunicationsLetters vol 10 no 11 pp 781ndash783 2006
[22] C Kim ldquoData hiding by an improved exploiting modificationdirectionrdquoMultimedia Tools and Applications vol 69 no 3 pp569ndash584 2014
[23] S Y Shen and L H Huang ldquoA data hiding scheme usingpixel value differencing and improving exploiting modificationdirectionsrdquo Computers amp Security vol 48 pp 131ndash141 2015
[24] T D Nguyen S Arch-Int and N Arch-Int ldquoAn adaptivemulti bit-plane image steganography using block data-hidingrdquoMultimediaTools andApplications vol 75 no 14 pp 8319ndash83452016
[25] A Soria-Lorente and S Berres ldquoA secure steganographicalgorithm based on frequency domain for the transmission ofhidden informationrdquo Security and Communication Networksvol 2017 Article ID 5397028 14 pages 2017
Advances in Multimedia 13
[26] A Pradhan K R Sekhar and G Swain ldquoAdaptive PVDsteganography using horizontal vertical and diagonal edges insix-pixel blocksrdquo Security and Communication Networks vol2017 Article ID 1924618 13 pages 2017
[27] A Pradhan K R ekhar andG Swain ldquoDigital image steganog-raphy using LSB substitution PVD and EMDrdquo MathematicalProblems in Engineering 2018
[28] G Swain ldquoHigh capacity image steganography using modifiedLSB substitution and PVD against pixel difference histogramanalysisrdquo Security and Communication Networks 2018
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
10 Advances in Multimedia
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Lena coverLena stego
times104
0
1
2
Occ
urre
nces
3
Pixel differenceminus20minus40 40200
Baboon coverBaboon stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Tiffany coverTiffany stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
peppers coverpeppers stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Jet coverJet stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Boat coverBoat stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
House coverHouse stego
times104
0
1
2
Occ
urre
nces
3
minus20minus40 4020Pixel difference
0
Pot coverPot stego
times104
Figure 8 PDH analysis for type 2 of the proposed technique
Advances in Multimedia 11
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(a) Type 1 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(b) Type 1 Baboon
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(c) Type 2 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(d) Type 2 Baboon
Figure 9 RS analysis for type 1 and type 2 of the proposed technique
Sminusm
= No of blocks satisfying the condition f (Fminus1(G)) lt f (G)
Total number of blocks
(25)
RS analysis is performed by using these four parametersIf the condition R
119898asymp Rminus119898gt Sm asymp S
minusm is true then RSanalysis fails to detect the steganography technique But if the
condition Rminus119898
- Sminusm gt Rm - Sm is true then the RS analysis
succeeds in detecting the steganography techniqueFigure 9 represents RS analysis for the proposed tech-
nique Figures 9(a) and 9(b) stand for RS analysis over Lenaand Baboon images respectively of type 1 Similarly Figures9(c) and 9(d) stand for RS analysis over Lena and Baboonimages respectively of type 2 In all the four cases the
12 Advances in Multimedia
condition R119898asymp Rminus119898gt Sm asymp Sminusm is true so we can conclude
that RS analysis could not detect the proposed techniqueLena image has more smooth regions and Baboon imagehas more edge regions all the remaining six images are inbetween these two in terms of smoothness So it is sufficientto draw a conclusion based on these two images
6 Conclusion
The traditional PVD steganography techniques follow a staticrange table Due to this some undesired steps are introducedin pixel difference histograms of the stego images This stepeffect could be avoided by exploiting edges in eight differentdirections This paper proposes a steganography techniqueusing a judicious combination of LSB substitution and eight-directional PVD The LSB substitution is performed only inone pixel out of nine pixels in a block so RS analysis cannotdetect it By using the edges in multiple directions the PDHanalysis cannot detect itThere exist two types of the proposedtechnique with regard to two different quantization tablesThe type 1 provides higher PSNR and the type 2 provideshigher hiding capacity This proposed technique performsbetter in terms of hiding capacity and PSNR over othercompeting PVD techniques
Conflicts of Interest
The author declares that there are no conflicts of interest
References
[1] D-CWu andW-H Tsai ldquoA steganographicmethod for imagesby pixel-value differencingrdquo Pattern Recognition Letters vol 24no 9-10 pp 1613ndash1626 2003
[2] K C Chang C P Chang P S Huang and T M Tu ldquoAnovel image steganographic method using tri-way pixel-valuedifferencingrdquo Journal of Multimedia vol 3 no 2 pp 37ndash442008
[3] Y-P Lee J-C Lee W-K Chen K-C Chang I-J Su and C-P Chang ldquoHigh-payload image hiding with quality recoveryusing tri-way pixel-value differencingrdquo Information Sciencesvol 191 pp 214ndash225 2012
[4] C-C Chang and H-W Tseng ldquoA steganographic method fordigital images using side matchrdquo Pattern Recognition Lettersvol 25 no 12 pp 1431ndash1437 2004
[5] C-H Yang C-Y Weng H-K Tso and S-J Wang ldquoA datahiding scheme using the varieties of pixel-value differencing inmultimedia imagesrdquo The Journal of Systems and Software vol84 no 4 pp 669ndash678 2011
[6] W Hong T-S Chen and C-W Luo ldquoData embedding usingpixel value differencing and diamond encoding with multiple-base notational systemrdquo The Journal of Systems and Softwarevol 85 no 5 pp 1166ndash1175 2012
[7] H C Wu N I Wu C S Tsai and M-S Hwang ldquoImagesteganographic scheme based on pixel-value differencing andLSB replacementmethodsrdquo IEE ProceedingsmdashVision Image andSignal Processing vol 152 no 5 pp 611ndash615 2005
[8] C-H Yang C-Y Weng S-J Wang and H-M Sun ldquoVariedPVD + LSB evading detection programs to spatial domain in
data embedding systemsrdquo The Journal of Systems and Softwarevol 83 no 10 pp 1635ndash1643 2010
[9] J Chen ldquoA PVD-based data hiding method with histogrampreserving using pixel pair matchingrdquo Signal Processing ImageCommunication vol 29 no 3 pp 375ndash384 2014
[10] MKhodaei andK Faez ldquoNew adaptive steganographicmethodusing least-significant-bit substitution and pixel-value differ-encingrdquo IET Image Processing vol 6 no 6 pp 677ndash686 2012
[11] G Swain ldquoA steganographic method combining LSB substitu-tion and PVD in a blockrdquo Procedia Computer Science vol 85pp 39ndash44 2016
[12] X Liao Q Wen and J Zhang ldquoA steganographic methodfor digital images with four-pixel differencing and modifiedLSB substitutionrdquo Journal of Visual Communication and ImageRepresentation vol 22 no 1 pp 1ndash8 2011
[13] G Swain ldquoDigital image steganography using nine-pixel dif-ferencing and modified LSB substitutionrdquo Indian Journal ofScience and Technology vol 7 no 9 pp 1444ndash1450 2014httpwwwindjstorgindexphpindjstarticle
[14] W Luo F Huang and J Huang ldquoA more secure steganographybased on adaptive pixel-value differencing schemerdquoMultimediaTools and Applications vol 52 no 2-3 pp 407ndash430 2011
[15] G Swain ldquoAdaptive pixel value differencing steganographyusing both vertical and horizontal edgesrdquoMultimedia Tools andApplications vol 75 no 21 pp 13541ndash13556 2016
[16] S Chakraborty A S Jalal and C Bhatnagar ldquoLSB based nonblind predictive edge adaptive image steganographyrdquoMultime-dia Tools and Applications vol 76 no 6 pp 7973ndash7987 2017
[17] C Balasubramanian S Selvakumar and S Geetha ldquoHighpayload image steganography with reduced distortion usingoctonary pixel pairing schemerdquoMultimedia Tools and Applica-tions vol 73 no 3 pp 2223ndash2245 2014
[18] A Pradhan K R Sekhar and G Swain ldquoDigital imagesteganography based on seven way pixel value differencingrdquoIndian Journal of Science and Technology vol 9 no 37 pp 1ndash112016
[19] K A Darabkh A K Al-Dhamari and I F Jafar ldquoA newsteganographic algorithm based on multi directional PVD andmodified LSBrdquo Information Technology and Control vol 46 no1 pp 16ndash36 2017
[20] A K Gulve and M S Joshi ldquoA high capacity secured imagesteganography method with five pixel pair differencing andLSB substitutionrdquo International Journal of Image Graphics andSignal Processing vol 7 no 5 pp 66ndash74 2015
[21] X Zhang and S Wang ldquoEfficient steganographic embeddingby exploiting modification directionrdquo IEEE CommunicationsLetters vol 10 no 11 pp 781ndash783 2006
[22] C Kim ldquoData hiding by an improved exploiting modificationdirectionrdquoMultimedia Tools and Applications vol 69 no 3 pp569ndash584 2014
[23] S Y Shen and L H Huang ldquoA data hiding scheme usingpixel value differencing and improving exploiting modificationdirectionsrdquo Computers amp Security vol 48 pp 131ndash141 2015
[24] T D Nguyen S Arch-Int and N Arch-Int ldquoAn adaptivemulti bit-plane image steganography using block data-hidingrdquoMultimediaTools andApplications vol 75 no 14 pp 8319ndash83452016
[25] A Soria-Lorente and S Berres ldquoA secure steganographicalgorithm based on frequency domain for the transmission ofhidden informationrdquo Security and Communication Networksvol 2017 Article ID 5397028 14 pages 2017
Advances in Multimedia 13
[26] A Pradhan K R Sekhar and G Swain ldquoAdaptive PVDsteganography using horizontal vertical and diagonal edges insix-pixel blocksrdquo Security and Communication Networks vol2017 Article ID 1924618 13 pages 2017
[27] A Pradhan K R ekhar andG Swain ldquoDigital image steganog-raphy using LSB substitution PVD and EMDrdquo MathematicalProblems in Engineering 2018
[28] G Swain ldquoHigh capacity image steganography using modifiedLSB substitution and PVD against pixel difference histogramanalysisrdquo Security and Communication Networks 2018
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Advances in Multimedia 11
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(a) Type 1 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
0 40 60 80 10020Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(b) Type 1 Baboon
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(c) Type 2 Lena
20
25
30
35
40
45
50
Perc
enta
ge o
f the
regu
lar a
nd si
ngul
ar p
ixel
gro
ups
20 40 60 80 1000Percentage of hiding capacity
Rm
Rminusm
Sminusm
Sm
(d) Type 2 Baboon
Figure 9 RS analysis for type 1 and type 2 of the proposed technique
Sminusm
= No of blocks satisfying the condition f (Fminus1(G)) lt f (G)
Total number of blocks
(25)
RS analysis is performed by using these four parametersIf the condition R
119898asymp Rminus119898gt Sm asymp S
minusm is true then RSanalysis fails to detect the steganography technique But if the
condition Rminus119898
- Sminusm gt Rm - Sm is true then the RS analysis
succeeds in detecting the steganography techniqueFigure 9 represents RS analysis for the proposed tech-
nique Figures 9(a) and 9(b) stand for RS analysis over Lenaand Baboon images respectively of type 1 Similarly Figures9(c) and 9(d) stand for RS analysis over Lena and Baboonimages respectively of type 2 In all the four cases the
12 Advances in Multimedia
condition R119898asymp Rminus119898gt Sm asymp Sminusm is true so we can conclude
that RS analysis could not detect the proposed techniqueLena image has more smooth regions and Baboon imagehas more edge regions all the remaining six images are inbetween these two in terms of smoothness So it is sufficientto draw a conclusion based on these two images
6 Conclusion
The traditional PVD steganography techniques follow a staticrange table Due to this some undesired steps are introducedin pixel difference histograms of the stego images This stepeffect could be avoided by exploiting edges in eight differentdirections This paper proposes a steganography techniqueusing a judicious combination of LSB substitution and eight-directional PVD The LSB substitution is performed only inone pixel out of nine pixels in a block so RS analysis cannotdetect it By using the edges in multiple directions the PDHanalysis cannot detect itThere exist two types of the proposedtechnique with regard to two different quantization tablesThe type 1 provides higher PSNR and the type 2 provideshigher hiding capacity This proposed technique performsbetter in terms of hiding capacity and PSNR over othercompeting PVD techniques
Conflicts of Interest
The author declares that there are no conflicts of interest
References
[1] D-CWu andW-H Tsai ldquoA steganographicmethod for imagesby pixel-value differencingrdquo Pattern Recognition Letters vol 24no 9-10 pp 1613ndash1626 2003
[2] K C Chang C P Chang P S Huang and T M Tu ldquoAnovel image steganographic method using tri-way pixel-valuedifferencingrdquo Journal of Multimedia vol 3 no 2 pp 37ndash442008
[3] Y-P Lee J-C Lee W-K Chen K-C Chang I-J Su and C-P Chang ldquoHigh-payload image hiding with quality recoveryusing tri-way pixel-value differencingrdquo Information Sciencesvol 191 pp 214ndash225 2012
[4] C-C Chang and H-W Tseng ldquoA steganographic method fordigital images using side matchrdquo Pattern Recognition Lettersvol 25 no 12 pp 1431ndash1437 2004
[5] C-H Yang C-Y Weng H-K Tso and S-J Wang ldquoA datahiding scheme using the varieties of pixel-value differencing inmultimedia imagesrdquo The Journal of Systems and Software vol84 no 4 pp 669ndash678 2011
[6] W Hong T-S Chen and C-W Luo ldquoData embedding usingpixel value differencing and diamond encoding with multiple-base notational systemrdquo The Journal of Systems and Softwarevol 85 no 5 pp 1166ndash1175 2012
[7] H C Wu N I Wu C S Tsai and M-S Hwang ldquoImagesteganographic scheme based on pixel-value differencing andLSB replacementmethodsrdquo IEE ProceedingsmdashVision Image andSignal Processing vol 152 no 5 pp 611ndash615 2005
[8] C-H Yang C-Y Weng S-J Wang and H-M Sun ldquoVariedPVD + LSB evading detection programs to spatial domain in
data embedding systemsrdquo The Journal of Systems and Softwarevol 83 no 10 pp 1635ndash1643 2010
[9] J Chen ldquoA PVD-based data hiding method with histogrampreserving using pixel pair matchingrdquo Signal Processing ImageCommunication vol 29 no 3 pp 375ndash384 2014
[10] MKhodaei andK Faez ldquoNew adaptive steganographicmethodusing least-significant-bit substitution and pixel-value differ-encingrdquo IET Image Processing vol 6 no 6 pp 677ndash686 2012
[11] G Swain ldquoA steganographic method combining LSB substitu-tion and PVD in a blockrdquo Procedia Computer Science vol 85pp 39ndash44 2016
[12] X Liao Q Wen and J Zhang ldquoA steganographic methodfor digital images with four-pixel differencing and modifiedLSB substitutionrdquo Journal of Visual Communication and ImageRepresentation vol 22 no 1 pp 1ndash8 2011
[13] G Swain ldquoDigital image steganography using nine-pixel dif-ferencing and modified LSB substitutionrdquo Indian Journal ofScience and Technology vol 7 no 9 pp 1444ndash1450 2014httpwwwindjstorgindexphpindjstarticle
[14] W Luo F Huang and J Huang ldquoA more secure steganographybased on adaptive pixel-value differencing schemerdquoMultimediaTools and Applications vol 52 no 2-3 pp 407ndash430 2011
[15] G Swain ldquoAdaptive pixel value differencing steganographyusing both vertical and horizontal edgesrdquoMultimedia Tools andApplications vol 75 no 21 pp 13541ndash13556 2016
[16] S Chakraborty A S Jalal and C Bhatnagar ldquoLSB based nonblind predictive edge adaptive image steganographyrdquoMultime-dia Tools and Applications vol 76 no 6 pp 7973ndash7987 2017
[17] C Balasubramanian S Selvakumar and S Geetha ldquoHighpayload image steganography with reduced distortion usingoctonary pixel pairing schemerdquoMultimedia Tools and Applica-tions vol 73 no 3 pp 2223ndash2245 2014
[18] A Pradhan K R Sekhar and G Swain ldquoDigital imagesteganography based on seven way pixel value differencingrdquoIndian Journal of Science and Technology vol 9 no 37 pp 1ndash112016
[19] K A Darabkh A K Al-Dhamari and I F Jafar ldquoA newsteganographic algorithm based on multi directional PVD andmodified LSBrdquo Information Technology and Control vol 46 no1 pp 16ndash36 2017
[20] A K Gulve and M S Joshi ldquoA high capacity secured imagesteganography method with five pixel pair differencing andLSB substitutionrdquo International Journal of Image Graphics andSignal Processing vol 7 no 5 pp 66ndash74 2015
[21] X Zhang and S Wang ldquoEfficient steganographic embeddingby exploiting modification directionrdquo IEEE CommunicationsLetters vol 10 no 11 pp 781ndash783 2006
[22] C Kim ldquoData hiding by an improved exploiting modificationdirectionrdquoMultimedia Tools and Applications vol 69 no 3 pp569ndash584 2014
[23] S Y Shen and L H Huang ldquoA data hiding scheme usingpixel value differencing and improving exploiting modificationdirectionsrdquo Computers amp Security vol 48 pp 131ndash141 2015
[24] T D Nguyen S Arch-Int and N Arch-Int ldquoAn adaptivemulti bit-plane image steganography using block data-hidingrdquoMultimediaTools andApplications vol 75 no 14 pp 8319ndash83452016
[25] A Soria-Lorente and S Berres ldquoA secure steganographicalgorithm based on frequency domain for the transmission ofhidden informationrdquo Security and Communication Networksvol 2017 Article ID 5397028 14 pages 2017
Advances in Multimedia 13
[26] A Pradhan K R Sekhar and G Swain ldquoAdaptive PVDsteganography using horizontal vertical and diagonal edges insix-pixel blocksrdquo Security and Communication Networks vol2017 Article ID 1924618 13 pages 2017
[27] A Pradhan K R ekhar andG Swain ldquoDigital image steganog-raphy using LSB substitution PVD and EMDrdquo MathematicalProblems in Engineering 2018
[28] G Swain ldquoHigh capacity image steganography using modifiedLSB substitution and PVD against pixel difference histogramanalysisrdquo Security and Communication Networks 2018
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
12 Advances in Multimedia
condition R119898asymp Rminus119898gt Sm asymp Sminusm is true so we can conclude
that RS analysis could not detect the proposed techniqueLena image has more smooth regions and Baboon imagehas more edge regions all the remaining six images are inbetween these two in terms of smoothness So it is sufficientto draw a conclusion based on these two images
6 Conclusion
The traditional PVD steganography techniques follow a staticrange table Due to this some undesired steps are introducedin pixel difference histograms of the stego images This stepeffect could be avoided by exploiting edges in eight differentdirections This paper proposes a steganography techniqueusing a judicious combination of LSB substitution and eight-directional PVD The LSB substitution is performed only inone pixel out of nine pixels in a block so RS analysis cannotdetect it By using the edges in multiple directions the PDHanalysis cannot detect itThere exist two types of the proposedtechnique with regard to two different quantization tablesThe type 1 provides higher PSNR and the type 2 provideshigher hiding capacity This proposed technique performsbetter in terms of hiding capacity and PSNR over othercompeting PVD techniques
Conflicts of Interest
The author declares that there are no conflicts of interest
References
[1] D-CWu andW-H Tsai ldquoA steganographicmethod for imagesby pixel-value differencingrdquo Pattern Recognition Letters vol 24no 9-10 pp 1613ndash1626 2003
[2] K C Chang C P Chang P S Huang and T M Tu ldquoAnovel image steganographic method using tri-way pixel-valuedifferencingrdquo Journal of Multimedia vol 3 no 2 pp 37ndash442008
[3] Y-P Lee J-C Lee W-K Chen K-C Chang I-J Su and C-P Chang ldquoHigh-payload image hiding with quality recoveryusing tri-way pixel-value differencingrdquo Information Sciencesvol 191 pp 214ndash225 2012
[4] C-C Chang and H-W Tseng ldquoA steganographic method fordigital images using side matchrdquo Pattern Recognition Lettersvol 25 no 12 pp 1431ndash1437 2004
[5] C-H Yang C-Y Weng H-K Tso and S-J Wang ldquoA datahiding scheme using the varieties of pixel-value differencing inmultimedia imagesrdquo The Journal of Systems and Software vol84 no 4 pp 669ndash678 2011
[6] W Hong T-S Chen and C-W Luo ldquoData embedding usingpixel value differencing and diamond encoding with multiple-base notational systemrdquo The Journal of Systems and Softwarevol 85 no 5 pp 1166ndash1175 2012
[7] H C Wu N I Wu C S Tsai and M-S Hwang ldquoImagesteganographic scheme based on pixel-value differencing andLSB replacementmethodsrdquo IEE ProceedingsmdashVision Image andSignal Processing vol 152 no 5 pp 611ndash615 2005
[8] C-H Yang C-Y Weng S-J Wang and H-M Sun ldquoVariedPVD + LSB evading detection programs to spatial domain in
data embedding systemsrdquo The Journal of Systems and Softwarevol 83 no 10 pp 1635ndash1643 2010
[9] J Chen ldquoA PVD-based data hiding method with histogrampreserving using pixel pair matchingrdquo Signal Processing ImageCommunication vol 29 no 3 pp 375ndash384 2014
[10] MKhodaei andK Faez ldquoNew adaptive steganographicmethodusing least-significant-bit substitution and pixel-value differ-encingrdquo IET Image Processing vol 6 no 6 pp 677ndash686 2012
[11] G Swain ldquoA steganographic method combining LSB substitu-tion and PVD in a blockrdquo Procedia Computer Science vol 85pp 39ndash44 2016
[12] X Liao Q Wen and J Zhang ldquoA steganographic methodfor digital images with four-pixel differencing and modifiedLSB substitutionrdquo Journal of Visual Communication and ImageRepresentation vol 22 no 1 pp 1ndash8 2011
[13] G Swain ldquoDigital image steganography using nine-pixel dif-ferencing and modified LSB substitutionrdquo Indian Journal ofScience and Technology vol 7 no 9 pp 1444ndash1450 2014httpwwwindjstorgindexphpindjstarticle
[14] W Luo F Huang and J Huang ldquoA more secure steganographybased on adaptive pixel-value differencing schemerdquoMultimediaTools and Applications vol 52 no 2-3 pp 407ndash430 2011
[15] G Swain ldquoAdaptive pixel value differencing steganographyusing both vertical and horizontal edgesrdquoMultimedia Tools andApplications vol 75 no 21 pp 13541ndash13556 2016
[16] S Chakraborty A S Jalal and C Bhatnagar ldquoLSB based nonblind predictive edge adaptive image steganographyrdquoMultime-dia Tools and Applications vol 76 no 6 pp 7973ndash7987 2017
[17] C Balasubramanian S Selvakumar and S Geetha ldquoHighpayload image steganography with reduced distortion usingoctonary pixel pairing schemerdquoMultimedia Tools and Applica-tions vol 73 no 3 pp 2223ndash2245 2014
[18] A Pradhan K R Sekhar and G Swain ldquoDigital imagesteganography based on seven way pixel value differencingrdquoIndian Journal of Science and Technology vol 9 no 37 pp 1ndash112016
[19] K A Darabkh A K Al-Dhamari and I F Jafar ldquoA newsteganographic algorithm based on multi directional PVD andmodified LSBrdquo Information Technology and Control vol 46 no1 pp 16ndash36 2017
[20] A K Gulve and M S Joshi ldquoA high capacity secured imagesteganography method with five pixel pair differencing andLSB substitutionrdquo International Journal of Image Graphics andSignal Processing vol 7 no 5 pp 66ndash74 2015
[21] X Zhang and S Wang ldquoEfficient steganographic embeddingby exploiting modification directionrdquo IEEE CommunicationsLetters vol 10 no 11 pp 781ndash783 2006
[22] C Kim ldquoData hiding by an improved exploiting modificationdirectionrdquoMultimedia Tools and Applications vol 69 no 3 pp569ndash584 2014
[23] S Y Shen and L H Huang ldquoA data hiding scheme usingpixel value differencing and improving exploiting modificationdirectionsrdquo Computers amp Security vol 48 pp 131ndash141 2015
[24] T D Nguyen S Arch-Int and N Arch-Int ldquoAn adaptivemulti bit-plane image steganography using block data-hidingrdquoMultimediaTools andApplications vol 75 no 14 pp 8319ndash83452016
[25] A Soria-Lorente and S Berres ldquoA secure steganographicalgorithm based on frequency domain for the transmission ofhidden informationrdquo Security and Communication Networksvol 2017 Article ID 5397028 14 pages 2017
Advances in Multimedia 13
[26] A Pradhan K R Sekhar and G Swain ldquoAdaptive PVDsteganography using horizontal vertical and diagonal edges insix-pixel blocksrdquo Security and Communication Networks vol2017 Article ID 1924618 13 pages 2017
[27] A Pradhan K R ekhar andG Swain ldquoDigital image steganog-raphy using LSB substitution PVD and EMDrdquo MathematicalProblems in Engineering 2018
[28] G Swain ldquoHigh capacity image steganography using modifiedLSB substitution and PVD against pixel difference histogramanalysisrdquo Security and Communication Networks 2018
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Advances in Multimedia 13
[26] A Pradhan K R Sekhar and G Swain ldquoAdaptive PVDsteganography using horizontal vertical and diagonal edges insix-pixel blocksrdquo Security and Communication Networks vol2017 Article ID 1924618 13 pages 2017
[27] A Pradhan K R ekhar andG Swain ldquoDigital image steganog-raphy using LSB substitution PVD and EMDrdquo MathematicalProblems in Engineering 2018
[28] G Swain ldquoHigh capacity image steganography using modifiedLSB substitution and PVD against pixel difference histogramanalysisrdquo Security and Communication Networks 2018
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
