Background Subtraction Based on Phase and Distance Transform Under Sudden Illumination Change
1. Background Subtraction Based on Phase and Distance
Transform Under Sudden Illumination Change
Authors: Gengjian Xue, Jun Sun, Li Song
Reporter: Li Song
Shanghai Jiao Tong University
2012-09-27 1
2. Outline
Introduction
Background subtraction based on phase
feature and distance transform
Experimental results
Conclusion
2012-09-27 2
3. Introduction (1)
Moving object detection is an active
research subject in computer vision.
Foreground detection under sudden
illumination change is still a challenging
problem.
2012-09-27 3
4. Introduction (2)
Parametric density estimation technique
Gaussian Mixture Models (GMM), C.Stauffer and
Grimson, CVPR, 1999.
Nonparametric density estimation technique
Kernel Density Estimation (KDE), A. Elgammal,
etc., Pro.IEEE, 2002.
Region-based technique
Local Binary Pattern based (LBP), M.Heikkila and
M. Pietikainen, TPAMI, 2006
2012-09-27 4
5. Introduction (3)
Other methods
Wallflower system, K.Toyama, etc., ICCV, 1999
Pixel order information based, Binglong Xie, etc.,
Image Vision Computing, 2004
Pilet’s method, ECCV 2008.
---a statistical model combing two texture-based
features and the ratios of the current image to the
background image in each color channel
2012-09-27 5
6. Motivation
Phase is insensitive to illumination
change.
The basic framework of GMM method.
Blob aggregation using distance
transform.
2012-09-27 6
7. Diagram
Background subtraction
based on phase feature
and distance transform
2012-09-27 7
8. Phase feature extraction (1)
The Gabor wavelet coefficient:
G , ( z ) A , z ( z ) exp(i ,v ( z ))
A ,v ( z ) :amplitude item
, ( z ) [0,2 ) :phase item
Each pixel has max max wavelet coefficients
2012-09-27 8
9. Phase feature extraction (2)
Phase extraction principle:
The larger amplitude of the coefficient is, the
more efficient the coefficient represents the pixel
information and the more discriminative of the
corresponding phase is.
Multiple phase extraction reasons:
rotation property
range of single phase is small
2012-09-27 9
10. Phase feature extraction (3)
The phase feature is defined:
L
p( z ) i ( z )
i 1
p(z ) : the phase feature
i (z ) : the Gabor phase corresponding to the ith largest Gabor
amplitude of the pixel
L : the number of phase to be extracted
N N image subdivision
2012-09-27 10
11. Background modeling
K mixture of Gaussian distributions
In the background modeling stage, each pixel is modeled
independently by a mixture of Gaussian distributions
where each Gaussian represents the phase distribution.
kth Gaussian distribution represented by
k , and k , total weights i 1
K
2
k
i 1
2012-09-27 11
12. Model updating (1)
Set L 3, p( z ) [0,6 )
Singular regions exists, matching condition
is in different expressions
6 k X t 2.5 k if X t and 6 k
6 k X t 2.5 k if k and 6 X t
X 2.5 others
t k k
2012-09-27 12
13. Model updating (2)
Condition satisfied
1. if k and 6 X t
k k ( X t k 6 )
2
k (1 ) k2 ( X t k 6 ) 2
2. if X t and 6 k
k k ( X t k 6 )
2
k (1 ) k2 ( X t k 6 ) 2
3. Others
k k ( X t k )
2
k (1 ) k2 ( X t k ) 2
2012-09-27 13
14. Model updating (3)
Weights update k (1 )k M k
k is beyond [0,6 )
k k 6 if k 6
k k 6 if k 0
Condition not satisfied
New Gaussian replace X t , init , low init
2012-09-27 14
15. Foreground detection
Sorted in
Background distributions:
b
B arg min( k T )
k 1
2012-09-27 15
17. Distance transform (2)
Euclidean distance transform of pixel a(i, j )
d (i, j ) min (i x) 2 ( j y) 2
( x , y )W
2012-09-27 17
18. Blob aggregation
Example results before and after blob aggregation
2012-09-27 18
19. Experiments (1)
Several sequences for testing.
GMM and LBP methods are employed for
comparison.
Visual and numerical evaluation are used
Precision and Recall as the metrics.
TP TP
Precision 100% Recall 100%
TP FP TP FN
2012-09-27 19
20. Experiments (2)
Parameter selection
Gabor filter: max 4 max 6
Phase extraction: N 4, L 3
Gaussian model: K 3, Tb 0.7
Segment value for DT: Td 1.2
2012-09-27 20