Video Foreground Detection Based on Adaptive Mixture Gaussian Model for Video ... · 2015. 6....

Video Foreground Detection Based on Adaptive

Mixture Gaussian Model for Video Surveillance

Systems

Mohammad Ali Alavianmehr Deputy of Transportation and Traffic Shiraz Municipality, Shiraz Traffic Control Center, Iran

E-mail: [email protected]

Ashkan Tashk and Amir Sodagaran Dept. of Electronics and Electrical Engineering, Shiraz University of Technology, Iran

Deputy of Transportation and Traffic Shiraz Municipality, Shiraz Traffic Control Center, Iran

E-mail: [email protected]; [email protected]

Abstract—Modeling background and moving objects are

significant techniques for video surveillance and other video

processing applications. This paper presents a foreground

detection algorithm that is robust against illumination

changes and noise based on adaptive Gaussian mixture

model (GMM), and provides a novel and practical choice for

intelligent video surveillance systems using static cameras. In

the previous methods, the image of still objects (background

image) is not significant. On the contrary, this method is

based on forming a meticulous background image and

exploiting it for separating moving objects from their

background. The background image is specified either

manually, by taking an image without vehicles, or is detected

in real-time by forming a mathematical or exponential

average of successive images. In comparison to other mixture

models that are complex and computationally expensive, the

proposed method is fast and easy to implement. The

proposed scheme can offer low image degradation. The

simulation results demonstrate high degree of performance

for the proposed method. Index Terms—image processing, Background models, video

surveillance, foreground detection, Gaussian mixture model.

I. INTRODUCTION

The application of image processing and computer

vision techniques to the analysis of video sequences of

traffic flow suggests noticeable improvements over the

present methods of traffic data collection and road traffic

monitoring. Other methods such as the inductive loop, the

sonar and microwave detectors suffer from important

drawbacks in that they cost a lot to install and maintain and

they are not able to identify slow or stationary vehicles.

Video sensors offer a relatively low installation cost with

little traffic disruption during maintenance. Moreover,

they offer wide area monitoring allowing analysis of

traffic flows and turning movements (important to

junction design), speed measurement, multiple point

Manuscript received October 15, 2014; revised March 29, 2015.

vehicle counts, vehicle classification and highway state

assessment (e.g. congestion or incident detection) [1].

Image processing also provides extensive applications

in the related field of autonomous vehicle guidance,

mainly for recognizing the vehicle’s relative location in

the lane and for obstacle detection. The problem of

autonomous vehicle guidance solves several problems at

different abstraction levels. The vision system can assist in

the meticulous localization of the vehicle with regard to its

environment, which is made up of the appropriate lane and

obstacles or other moving vehicles. Both lane and obstacle

detection are based on estimation procedures for detecting

the borders of the lane and determining the path of the

vehicle. The estimation is often done by matching the

observations (images) to a presumed road and/or vehicle

model. Video systems for either traffic monitoring or

autonomous vehicle guidance normally involve two

important tasks of perception: (a) estimation of road

geometry and (b) vehicle and obstacle detection. Road

traffic monitoring is to recognize and analyze traffic

figures, including presence and numbers of vehicles, speed

distribution data, turning traffic flows at intersections,

queue-lengths, space and time occupancy rates, etc.

Therefore, for traffic monitoring it is vital to identify the

lane of the road and then sense and determine presence

and/or motion parameters of a vehicle. Also, in

autonomous vehicle guidance, the knowledge about road

geometry allows a vehicle to follow its route and the

detection of road obstacles becomes an essential and

serious task for avoiding other vehicles present on the

road.

In road traffic monitoring, the video acquisition

cameras are stationary. They are located on posts above

the ground to get optimal view of the road and the passing

vehicles. In automatic vehicle guidance, the cameras are

moving with the vehicle. In these applications it is

necessary to analyze the dynamic change of the

environment and its contents, as well as the dynamic

change of the camera itself. Accordingly, object detection

Journal of Traffic and Logistics Engineering Vol. 3, No. 1, June 2015

45©2015 Journal of Traffic and Logistics Engineeringdoi: 10.12720/jtle.3.1.45-51

mailto:[email protected]



from a stationary camera is easier because it involves

fewer estimation procedures.

Initial approaches in this field include spatial, temporal

and spatio-temporal analysis of video sequences. More

advanced and effective approaches consider object

modeling and tracking using state-space estimation

procedures for matching the model to the observations and

for estimating the next state of the object. The most

common techniques, i.e. analysis of the optical flow field

and processing of stereo images, involve processing two or

more images. With optical-flow-field analysis, multiple

images are recognized at different times [2];

Optical-flow-based techniques detect obstacles indirectly

by analyzing the velocity field. Stereo image techniques

determine the correspondences between pixels in the

different images.

Object detection approaches have been classified

according to the method used to isolate the object from the

background on a single frame or a sequence of frames.

Approaches in this class are based on the edge-features

of objects. They can be applied to single images to detect

the edge structure of even still vehicles. Morphological

edge-detection schemes have been extensively applied,

since they exhibit superior performance. In traffic scenes,

the results of an edge detector generally highlight vehicles

as complex groups of edges, while road areas yield

relatively low edge content. Hence the existence of

vehicles may be detected by the edge complexity within

the road area, which can be quantified through analysis of

the histogram. Alternatively, the edges can be grouped

together to form the vehicle’s boundary. Towards this

direction, the algorithm must determine relevant features

(often line segments) and define a grouping strategy that

allows the identification of feature sets, each of which may

correspond to an object of interest (e.g. potential vehicle or

road obstacle). Vertical edges are more probable to form

dominant line segments corresponding to the vertical

boundaries of the profile of a road obstacle. Moreover, a

dominant line segment of a vehicle must have other line

segments in its neighborhood that are detected in nearly

perpendicular directions. Consequently, the detection of

vehicles and/or obstacles can be simply comprised of

finding the rectangles that enclose the dominant line

segments and their neighbors in the image plane. To

improve the shape of object regions Ref. [3], [4] uses the

Hought transform to extract consistent contour lines and

morphological operations to restore small breaks on the

detected contours. Symmetry offers an additional useful

feature for relating these line segments, since vehicle rears

are generally contour and region-symmetric about a

vertical central line.

The detection is then achieved by means of subtracting

the reference image from the current image. Thresholding

is done in order to obtain presence/absence information of

an object in motion. The shadows cast by buildings and

clouds or changes in lighting conditions can affect the

background significantly. With these changing

environmental conditions, it is essential to update the

background frame regularly. There is numerous

background updating techniques. The most common ones

are averaging and selective updating. In averaging, the

background is formed gradually by taking the average of

the previous background with the current frame.

While in exponential updating the background is built

by forming a weighted average between the previous

background and the current frame. In selective updating,

the background is replaced by the current frame only at

motionless regions; where the difference between the

current and the preceding frames is smaller than a

threshold. Selective updating can be done in a more robust

averaging form; that is the still regions of the background

are replaced by the average of the current frame and the

previous background.

This is the most direct method for making motionless

objects vanish and preserving only the traces of moving

objects between two successive frames. The immediate

consequence is that still or slow-moving objects are not

detected. The inter-frame difference succeeds in detecting

motion when temporal changes are obvious. Nevertheless,

it fails when the objects in motion are not sufficiently

textured and preserve uniform regions with the

background. To cope with this problem, the inter-frame

difference is described using a statistical framework often

utilizing spatial Markov random fields [5]. The

inter-frame difference is modeled trough a two-component

mixture density. These components are zero mean

corresponding to the static (background) and changing

(moving object) parts of the image.

Inter-frame differencing provides a crude but simple

tool for estimating moving regions. This process can be

complemented with background frame differencing in

order to improve the estimation accuracy [6]. With color

segmentation or accurate motion estimation we can further

refine the resulting mask of moving regions by means of

optical flow estimation and optimization of the displaced

frame difference to refine the segmentation of moving

objects.

In [7], a new mixture model for image segmentation is

presented. They propose a new way to incorporate spatial

information between neighboring pixels into the Gaussian

mixture model based on Markov random field (MRF). In

mixture models based on MRF, the M-step of the

expectation-maximization (EM) algorithm cannot be

directly applied to the prior distribution for maximization

of the log likelihood with respect to the corresponding

parameters.

Adaptive Gaussian mixture background model is

excellent background model cause of its good analytic

form and high operation efficiency. It is more appropriate

than single Gaussian models for high speed moving

objects detection under outdoor environment where image

background and illumination intensity changes slowly.

However, low convergence speed is its main

disadvantages especially as the illumination intensity

suddenly changes. It cannot adapt to these kinds of rapid

changes and often take changing background pixels as

moving objects, which makes the image foreground

information confused and moving objects lost. Recently,

although many modified Gaussian models (Chen, &He,


46©2015 Journal of Traffic and Logistics Engineering

2007; Ma&Zhu, 2007; Chen& Zhou, 2007) [8]-[10] have

been developed this problem has not been solved

effectively. In their models, the moving objects shadows

are occasionally detected in HIS (Hue), saturation

intensity space or HSV (hue, saturation, value) space,

which cause the objects detection and shadow removal

need to be processed in different spaces and joined

together by complicated space transform. Yinghong Li in

[11] proposed a new method of detecting moving vehicles

based on edged Gaussian mixture model. Firstly, mixture

Gaussian model is founded for image edge processing

according to the features of image edge information which

is not sensitive to sudden change of illumination intensity.

And some parameters such as the mean vectors and

variance vectors of the pixels are obtained from the model

for moving vehicle shadows detection. Moreover, through

statistical approach moving vehicle shadows are

eliminated based on the differences of brightness

distortion between vehicle shadows and moving vehicles

relative to background.

In [12], presented an enhanced GMM method for the

task of background subtraction. To that purpose, they use

an adaptive variance estimator for the initialization of new

modes. The value computed by the variance estimator is

further employed to trigger a splitting rule, which is

applied to over dominating nodes in order to avoid

under-fitting the underlying background distribution.

Dibyendu [13] proposed a Gaussian mixture model with

advanced distance measure based on support weights and

histogram of gradients for background suppression. The

method also employs variable number of clusters for

generalization. The main advantages of the method are

implicit use of pixel relationships through distance

measure with least modification to the conventional GMM

and effective background noise removal through the use of

background layer concept with no post processing

involved.

Tom S.F. Haines [14] proposed a new method based on

Dirichlet process GMM, which are used to compute

per-pixel background distributions. Using a

non-parametric Bayesian method allows per-pixel mode

counts to be automatically inferred, avoiding over-/under-

fitting.

The main contributions of this paper are (i) to couple

real time systems in order to achieve good detection results

while being able to correctly classify pixels from

foreground images and new static objects, therefore,

notably improving the results presented, (ii) to further

enhance the system proposed in [12] by means of a more

robust estimator versus noise for the variance initialization

value, (iii) learning detected background according to 3

previous frames for more accuracy which means in this

paper as adaptively. The remainder of this paper is

ordered as follows. In section II, the proposed method is

described completely and background images detection

from foreground images described. Section III assigns to

the experimental results and simulations. Eventually, in

section IV, this paper is concluded.

II. PROPOSED METHOD

In this section, we explain the proposed method in detail.

Four phases exist in our algorithm in order to estimate the

traffic status and to compute the velocity. (Fig. 1).

Receiving the images through video surveillance camera

in first phase, we get use of GMM for each frame to

achieve a precise background image. This process will be

repeated as long as we seize an accurate background

images. This phase is called training phase. In the second

phase, the received images will be analyzed a long with

the trained images to extract the vehicles (moving objects)

based on this analysis. As the mention above, we may

extract vehicles more accurately as long as, we have a

more precise trained background images. In third phase, a

green block will surround each vehicles to enable the

researches count them. Either inaccurate training of the

background images or the shadow of moving vehicles

might cause problems in detecting vehicles in motion in

the second phase. To solve these problems besides

merging the blocks which overlap the other ones, the

researchers also make use of shadow removing algorithm

to compute the volume and density of traffic in final phase.

Figure. 1. Algorithm of proposed method

The background learning module is our

multi-dimensional Gaussian kernel density transform

employed for spatio-temporal smoothing. Other stages in

this flow chart are either explained later in this paper or are

self-evident. Note that ‘dilation’ and 'erosion’ are standard

morphological operations applied to a binary image,

where contiguous areas of either ones or zeros are grown

(dilation) or shrunk (erosion) by the radius of some

so-called structuring element. Morphological opening is





an often-used methodology in image processing. Opening

of a binary image is erosion followed by a dilation using

the same structuring element for both operations.

Basic background subtraction (BS) techniques detect

foreground objects as the difference between two

consecutive video frames, operate at pixel level, and are

applicable to still backgrounds. Although the generic BS

method is easy to understand and implement, the

drawbacks of the frame difference BS are that it does not

provide a mechanism for selecting the parameters, such as

the detection threshold, and it is unable to deal with

multimodal distributions. One of the significant

techniques able to cope with multimodal background

distributions and to update the detection threshold

employs Gaussian mixture models (GMMs).

A Gaussian Mixture Model (GMM) is a parametric

probability density function represented as a weighted sum

of Gaussian component densities. GMMs are commonly

used as a parametric model of the probability distribution

of continuous measurements or as features in a biometric

system, such as color based tracking of an object in video.

In many computer related vision technology, it is vital to

determine moving objects from a sequence of videos

frames. In order to obtain this, background subtraction is

applied which mainly recognizes moving objects from

each portion of video frames. In video surveillance, target

recognitions and banks background subtraction or

segmentation technique is widely used. By using the

Gaussian Mixture Model background model, frame pixels

are removed from the required video to obtain the desired

results. The application of background subtraction

involves different factors which contain developing an

algorithm which is able to detect the required object

robustly, it should also be able to react to various changes

like illumination, starting and stopping of moving objects.

A. Background Subtraction

Background Subtraction is based on four important

steps which are stated below.

1) Preprocessing

Temporal or spatial smoothing is employed in the early

preprocessing stage to remove device noise which can be a

factor under different light intensity. Smoothing technique

also includes omitting various elements like environment

such as rain and snow (Fig. 2). In real-time systems, frame

size and frame rate are commonly adopted to decrease the

data processing rate. Another important factor in

preprocessing technique is the data format used by the

background subtraction model. Most algorithms can

handle luminance intensity which is one scalar value per

each pixel.

2) Background modeling

This step makes use of the new video frame in order to

compute and update the background model. The main

purpose of developing a background model is that it should

be robust against environmental changes in the background,

but sensitive enough to determine all moving objects of

interest. (Fig. 2).

3) Foreground detection

In this step, it recognizes the pixels in the frame.

Foreground detection compares the video frame with the

background model, and identify candidate foreground

pixels from the frame. To check whether the pixel is

significantly different from the corresponding background

estimate is a widely-used approach for foreground

detection.

4) Data validation

Finally, this step removed any pixels which are not

relevant to the image. It involves the process of improving

the foreground mask based on the information obtained

from the outside background model.

B. Algorithm of Gaussian Mixture Model

To give a better understanding of the algorithm used for

background subtraction the following steps were adopted

to achieve the desired results: 1. Firstly, we compare each

input pixels to the mean 'μ' of the associated components.

If the value of a pixel is close enough to a chosen

component's mean, then that component is counted as the

matched component. To be a matched component, the

difference between the pixels and mean must be less than

compared to the component's standard deviation. 2.

Secondly, update the Gaussian weight, mean and standard

deviation (variance) to reflect the new obtained pixel value.

Most background models lack three main points: 1.

ignoring any correlation between neighboring pixels 2. The

rate of adaption may not match the moving speed of the

foreground object. 3. Non-stationary pixels, from moving

leavers or shadow cast by objects in motion are at times

mistaken for true foreground objects.

In relation to non-matched components the weights 'w'

reduces whereas the mean and standard deviation stay the

same. It is dependent upon the learning component 'p' in

relation to how fast they are change. 3. Thirdly, here we

identify which components are parts of the background

model. To do this a threshold value applied to the

component weights 'w'. 4. Fourthly, in the final step we

determine the foreground pixels. Here the pixels that are

determined as foreground do not correspond with any

components identified to be the background.

(a)

(b)

Video Frame



(c)

Figure. 2. a, Original frame b, Background frame, c. Detected edge of

Background frame.

C. Formula of Gaussian Mixture Model

A Gaussian mixture model can be formulated in general

as follows:

𝑃(𝑋𝑡) = ∑ 𝑤𝑖,𝑡𝜂(𝑋𝑡; 𝜇𝑖,𝑡 , ∑ 𝑞𝑖,𝑡 )𝐾𝑖=1 (1)

where, obviously,

∑ 𝑤𝑖,𝑡 = 1𝐾𝑖=1 (2)

The mean of such a mixture equals

𝜇𝑡 = ∑ 𝑤𝑖,𝑡𝜇𝑖,𝑡𝐾𝑖=1 (3)

That is, the weighted sum of the means of the component densities.

where be the variable which represents the current pixel in frame, K is the number of distributions, and t represents time (i.e., the frame index), w is an estimate of the weight

of the ith Gaussian in the mixture at time t, is the mean value of the ith Gaussian in the mixture at time t, is the

covariance matrix of the ith Gaussian in the mixture at time

t.

III. EXPERIMENTAL RESULT

In this section, we present a quantitive evaluation of our

system. First, we evaluate the segmentation system, which includes background subtraction and shadow removal.

We present experiments that demonstrate how our

system detects the objects. Fig. 3 shows the results of Gaussian mixture model (GMM) in our method. The classical GMM relies on the statistical information of each

pixel, it fails to correctly detect moving objects in these

two datasets, however, and moving objects are correctly

detected by our method.

In order to quantitatively compare the proposed algorithms with their counterparts in the spatial domain,

we selected four typical frames and manually label all the

pixels of moving objects as the ground truth.

The sequential numbers of these frames are 505, 841,

1357, and 2269, respectively. Two metrics, false negative rate (FNR) and false positive rate (FPR), were used to

quantify the performance of each algorithm. They are

defined as

𝐹𝑁𝑅

=𝑡ℎ𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑓𝑜𝑟𝑒𝑔𝑟𝑜𝑢𝑛𝑑 𝑝𝑖𝑥𝑒𝑙𝑠 𝑤𝑟𝑜𝑛𝑔𝑙𝑦 𝑐𝑙𝑎𝑠𝑠𝑖𝑓𝑖𝑒𝑑

𝑡ℎ𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑓𝑜𝑟𝑒𝑔𝑟𝑜𝑢𝑛𝑑 𝑝𝑖𝑥𝑒𝑙𝑠 𝑖𝑛 𝑡ℎ𝑒 𝑔𝑟𝑜𝑢𝑛𝑑 𝑡𝑟𝑢𝑡ℎ

(4)

𝐹𝑃𝑅

=the number of background pixels wrongly classified

𝑡ℎ𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑏𝑎𝑐𝑘𝑔𝑟𝑜𝑢𝑛𝑑 𝑝𝑖𝑥𝑒𝑙𝑠 𝑖𝑛 𝑡ℎ𝑒 𝑔𝑟𝑜𝑢𝑛𝑑 𝑡𝑟𝑢𝑡ℎ

(5)

In Table I, the experimental results show our algorithms

commonly have a lower FPR than their counterparts in

[12]. The background subtraction technique assumes each

pixel is independent of each other, so it classifies each

pixel independently. That makes the segmentation

procedure is very sensitive to illumination changes and

noises. From Fig. 4, the related FNR and FPR are very

different from that for the Dibyendu's algorithms. The

maximum FNR (about 10%) in our method occurs near the

15th frame. Because of the learning of background image

is not accomplished accuracy.

TABLE I: COMPARISON OF PERFORMANCES OF THE PROPOSED

ALGORITHM WITH THEIR COUTERPARTS IN THE SPATIAL DOMAIN

Proposed Method RUBÉN 's

Method [12]

FNR% FPR% FNR% FPR%

Chamran video 3.86 2.52 4.98 4.69

Zand Video 2.64 1.35 3.89 3.85

Namazi Video 1.25 1.85 2.35 5.86

Neyaesh Video 2.63 1.15 3.26 3.93

Pole Bagh Safa video

1.56 0.95 2.66 2.65

Gomrok Video 0.91 0.63 2.36 1.98

Figure. 3. Original and foreground images in our method based on GMM.

Figure 4. Comparison between FPR and FNR (%) versus frame number

for test videos with [13].

Since the method is based on pixels, it results in a higher

probability to introduce foreground pixels than its

counterpart in the spatial domain. For instance, for a block

containing only one foreground pixel, the pixel will not

generally give rise to large change for the average of the

block compared with the case of no foreground pixel in it,

so the block is likely to be selected to represent the

background. Thus, the pixel-based median performs better

in the case. But the values of FNR for most frames are

below 5%.

IV. CONCLUSION

In this paper propose an approach to solve the crucial

problems related to Intelligent Transportation System (ITS)

for estimating important traffic parameters from video

sequences, that is detection algorithm as long with step

merging and estimate the moment velocity of vehicles, for

highway traffic surveillance. Our approach is low cost and

high accuracy, so the integration of this algorithm in vision

systems can enhance the performance of traffic parameters

estimation and vehicle detection. The proposed approach

succeeds to harness the advantages of both cues by

improving the detection rate without deteriorating the

precision. The results demonstrate its ability to improve

significantly the accuracy of the foreground segmentation

compared to other existing approaches in the literature.

The obtained superior results are significant since many

applications can be carried out after performing reliable

foreground segmentation.

ACKNOWLEDGMENTS

The authors would like to thank the anonymous

reviewers for their insightful comments that significantly

improved the quality of this paper. The authors would also

like to thank Mr. A. Zahmatkesh and Mrs. Sanaz

Akbarzadeh, Vice Chancellor for Transportation of Shiraz

Municipality, for their help on data collection and

experimental evaluations.

REFERENCES

[1] B. D. Stewart, I. Reading, M. S. Thomson, T. D. Binnie, K. W.

Dickinson, and C. L. Wan, “Adaptive lane finding in road traffic

image analysis,” in Proc. Seventh International Conference on Road Traffic Monitoring and Control, London, 1994.

[2] W. Enkelmann, “Obstacle detection by evaluation of optical flow

field from image sequences,” in Proc. European Conference on Computer Vision, Antibes, France, 1990, pp. 134–138.

[3] X. Li, Z. Q. Liu, and K. M. Leung, “Detection of vehicles from

traffic scenes using fuzzy integrals,” Pattern Recognition, vol. 35, pp. 967–980, 2002.

[4] H. Moon, R. Chellapa, and A. Rosenfeld, “Performance analysis of

a simple vehicle detection algorithm,” Image and Vision Computing, vol. 20, pp. 1–13, 2003.

[5] T. Aach and A. Kaup, “Bayesian algorithms for adaptive change

detection in image sequences using Markov random fields,” Signal

Processing: Image Communication, vol. 7, pp. 147–160, 1995. [6] J. B. Kim, H. S. Park, M. H. Park, and H. J. Kim, “A real-time

region-based motion segmentation using adaptive thresholding and

K-means clustering,” in, AI 2001, M. Brooks, D. Corbett, and M. Stumptner, Eds., Springer, Berlin, 2001, pp. 213–224.

[7] T. M. Nguyen and Q. M. J. Wu, “Fast and robust spatially

constrained gaussian mixture model for image segmentation,” IEEE Transactions on Circuits and Systems for Video Technology,

vol. 23, no. 4, 2013.

[8] Y. Ma, W. Zhu et al. “Improved moving objects detection method based on Gaussian mixture mode,” Computer Applications, vol. 27,

no. 10, pp. 2544-2548, 2007.

[9] Z. Chen, X. Chen, and H. He, “Moving object detection based on improved mixture Gaussian models,” Journal of Image and

Graphics, vol. 12, no. 9, pp. 1586-1589, 2007. [10] Z. Chen, R. Zhou, et al. “Simulation of an improved Gaussian

mixture model for background subtraction,” Computer Simulation,

vol. 24, no. 11, pp. 190-192, 2007. [11] Y. H. Li, Z. X. Li, H. F. Tian, and Y. Q. Wang, “Vehicle detecting

and shadow removing based on edged mixture Gaussian model,” in

Proc. Preprints of the 18th IFAC World Congress, Milano, Italy, 2011.

[12] R. H. Evangelio, M. Pätzold, I. Keller, and T. Sikora, “Adaptively

splitted GMM with feedback improvement for the taskof background subtraction,” IEEE Transactions on Information

Forensics and Security, vol. 9, no. 5, 2014.

[13] D. Mukherjee, Q. M. J. Wu, and T. M. Nguyen, “Gaussian mixture model with advanced distance measure based on support weights

and histogram of gradients for background suppression,” IEEE

Transactions on Industrial Information, vol. 10, no. 2, 2014. [14] T. S. F. Haines and T. Xiang, “Background subtraction with

dirichlet process mixture models,” IEEE Transactions on Pattern

Analysis and Machine Intelligence, vol. 36, no. 4, 2014.

M. A. Alavianmehr was born in Shiraz, Iran,

1985. He received the B.S. Degree in Electronics Engineering from the Bushehr

University in 2010 and the M.Sc. Degree from

the Sistan and Bluchestan (Zahedan) university in 2012. His areas of interests are

Image, Speech and video processing. He has

14 publications in both International Journals and International conference papers at various

International and National Conferences.

Recently, he is PhD candidate in Bushehr University and investigating Image and Video Processing for Traffic Application. He is now with

Vice Chancellor for Transportation of Shiraz Municipality, Shiraz

Traffic Control Center.

0 50 100 1500

1

2

3

4

5

6

7

8

9

10

Frame Number(%)

FN

R(%

)

[13]

Proposed Method

0 50 100 1500

1

2

3

4

5

6

7

8

9

Frame Number

FP

R(%

)

[13]

Proposed Method



A. Tashk received the B.Sc. Degree in Communication Engineering from the Shiraz

University of Technology, and M.Sc. Degree

in Communication Engineering from Shiraz University of Technology in 2009 and 2014

respectively. He is PhD student in Shiraz

University of Technology now. His current research interests include image and video

processing. He has published more than 20

papers in international journals and conferences.

A. Sodagaran received the M.S. degree in IT engineering from the Shiraz University, and

the M.Sc. degree in Information Technology

from university of British, Nottingham University. He then joined the Vice Chancellor

for Transportation of Shiraz Municipality,

Shiraz Traffic Control Center, Iran. His current research interests include visual surveillance,

image and video analysis, intelligent

transportation systems. He has published more than 5 papers in international journals and conferences.



Video Foreground Detection Based on Adaptive Mixture Gaussian Model for Video ... · 2015. 6....

Documents

Transcript of Video Foreground Detection Based on Adaptive Mixture Gaussian Model for Video ... · 2015. 6....