ifsr_ImageFusion
-
Upload
matlab5903 -
Category
Documents
-
view
220 -
download
0
Transcript of ifsr_ImageFusion
-
8/3/2019 ifsr_ImageFusion
1/17
Image Fusion by Image
Blending
Arthur Goshtasby
ri ht t te ni ersit
Image Registration and Fusion Systems
-
8/3/2019 ifsr_ImageFusion
2/17
more images of a scene into a single highly.
Fusion methods: ow- eve p xe - ase
Mid-level (region-based)
High-level (description-based)
A. Goshtasby 2
-
8/3/2019 ifsr_ImageFusion
3/17
of different images of a scene into a single.
Image 1 Image 2 Combined image
A. Goshtasby 3
-
8/3/2019 ifsr_ImageFusion
4/17
Im f i n n im l n inproblem
A. Goshtasby 4
-
8/3/2019 ifsr_ImageFusion
5/17
.domain, find the image
.
2. Cut out theImage domain
area out o
the mostinformative
.
A. Goshtasby 5
Selected local areas in images
-
8/3/2019 ifsr_ImageFusion
6/17
.image.
A. Goshtasby 6
-
8/3/2019 ifsr_ImageFusion
7/17
-
Highest contrast
0
)log(i
iiG ppE Gray-scale
bgrC EEEE Color
A. Goshtasby 7
-
8/3/2019 ifsr_ImageFusion
8/17
i, i
the center of the ith block, use
N
j j
ii
yxG
yxyxW
1 ),(
,),(
as the blending function, where
22 .
2exp),( 2
iii yxG
.
A. Goshtasby 8
-
8/3/2019 ifsr_ImageFusion
9/17
=
1) the image domain is subdivided intoN
,
2) for blocki, the most informative image is
i.k , ,
the fused image is computed from:
N
kii yxIyxWyxO . ),(),(),(
A. Goshtasby 9
-
8/3/2019 ifsr_ImageFusion
10/17
,
block size is chosen based on image size.
The blending process does not create
input.
e met o s versat e, t can e use to usevarious types of images.
A. Goshtasby 10
-
8/3/2019 ifsr_ImageFusion
11/17
Application 1:
Fusion of multi-exposure imagesOptimizing criterion: Entropy
A. Goshtasby 11
Images courtesy of Paul Debevec, USC
-
8/3/2019 ifsr_ImageFusion
12/17
Optimizing criterion: Entropy
A. Goshtasby 12
r g na mages cour esy o ree ayar, o um a n vers y
-
8/3/2019 ifsr_ImageFusion
13/17
Optimizing criterion: Entropy
Original images courtesy of Max Lyons
A. Goshtasby 13
-
8/3/2019 ifsr_ImageFusion
14/17
Application 2:
Fusion of multi-focus images
Original
mages
Optimizing criterion:
Fused
image
Contrast
A. Goshtasby 14
r g na mages cour esy o mage
Fusion Systems Research
-
8/3/2019 ifsr_ImageFusion
15/17
Originalimages
Fused
ima e
A. Goshtasby 15
-
8/3/2019 ifsr_ImageFusion
16/17
this method is guaranteed to produce an image withthe highest information content at the desired
resolution.
It can be adapted to image resolution by makingblock size a function of image resolution.
By changing the optimization criterion, the method
can be used to fuse various types of images.
A. Goshtasby 16
-
8/3/2019 ifsr_ImageFusion
17/17
1. T. Porter and T. Duff, Compositing digital images,ACM SIGGRAPH Computer
Graphics, vol. 18, no. 3, 1984, pp. 253-259.
2. P. Willis, Generalized compositing, ACM SIGGRAPH, 2007, pp. 129-135.
3. A. Goshtasby, Fusion of multi-exposure images,Image and Vision Computing,
vo . , , pp. - .
4. A. Goshtasby, Fusion of multifocus images to maximize image information,
Defense and Security SYmposium, 17-21 April 2006, Orlando, Florida, 2006.. n orma on us on , . .
Nikolov (Eds.), vol. 8, 2007.
6. R. S. Blum and Z. Liu,Multi-sensor Image Fusion and its Applications, CRC
Press 2005.
7. F. Sroubek and J. Flusser, Registration and fusion of blurred images,Intl Conf.
Image Analysis and Recognition, 2004, 124-129.
8. C. Genderen and J. L. van Pohl, Multisensor image fusion in remote sensing:
Concepts, methods and applications,Intl J. Remote Sensing, vol. 19, no. 5, 1998,
pp. 823-854.
A. Goshtasby17