Foreground segmentation under sudden illumination changes by feature fusion

Gong Weiguo; Liu Runyao; Zhang Rui

To address the challenging problem of robust background subtraction under sudden illumination changes, a novel foreground segmentation method based on feature fusion was proposed. The method consisted of three stages. First, a scene model through integrating the global illumination function into the framework of Gaussian mixture models was built. Second, three kinds of illumination invariant features, i. e. zero mean normalized cross -correlation (ZNCC), textures, and contours, were extracted from the current frame image. Third, the illumination invariant features were combined for foreground segmentation in two steps. Specifically, the ZNCC and textures were combined in the first step, and the contour was integrated in the second step. The experimental results show that the proposed method can effectively improve the accuracy and robustness of foreground segmentation.

1. Key Lab of Optoelectronic Technology & Systems of Education Ministry,Chongqing University,Chongqing 400044,China

Received Date: 2014-04-10

Rev Recd Date: 2014-05-13

Publish Date: 2014-12-25

Key words:

Abstract: To address the challenging problem of robust background subtraction under sudden illumination changes, a novel foreground segmentation method based on feature fusion was proposed. The method consisted of three stages. First, a scene model through integrating the global illumination function into the framework of Gaussian mixture models was built. Second, three kinds of illumination invariant features, i. e. zero mean normalized cross -correlation (ZNCC), textures, and contours, were extracted from the current frame image. Third, the illumination invariant features were combined for foreground segmentation in two steps. Specifically, the ZNCC and textures were combined in the first step, and the contour was integrated in the second step. The experimental results show that the proposed method can effectively improve the accuracy and robustness of foreground segmentation.

HTML

Reference (15)

[1]	Toyama K, Krumm J, B rumitt B, et al. Wall flower: principles and practice of background maintenance [C]//ICCV, 1999, 1: 255-261.
[2]
[3]
[4]	Li Y, Xu R, Nie G, et al. Moving object detection method for change in scene light [J]. Computer Technology and Development, 2011, 21(2): 140-146. (in Chinese)
[5]
[6]	Xue G, Sun J, Song L. Background subtraction based on phase and distance transform under sudden illumination change[C]//ICIP 2010, 2010: 3465-3468.
[7]	Strecha Pilet C, Fua P. Making background subtraction robust to sudden illumination changes [C]//Proc European Conf on Computer Vision, 2008: 567-580.
[8]
[9]	Osters L P J V, Caifeng Shan, Tommaso Gritti. Background subtraction under sudden illumination changes [C]//Advanced Video and Signal Based Surveillance (AVSS), 2010: 384-391.
[10]
[11]
[12]	Luc Vosters, Caifeng Shan, Tommaso Gritti. Real-time robust background subtraction under rapidly changing illumination conditions [J]. Image and Vision Computing, 2012, 30(12): 1004-1015.
[13]
[14]	Yuan C, Wang C, Zhang X, et al. Video segmentation of illuminance abrupt variation based on MOGs and gradient information [J]. Journal of Image and Graphics, 2007, 11 (12): 2068-2072. (in Chinese)
[15]	Xu Li, Yan Qiong, Xia Yang, et al. Structure extraction from texture via relative total variation [J]. ACM Transactions on Graphics, 2012, 31(6): 139.

Foreground segmentation under sudden illumination changes by feature fusion

Abstract

References

Proportional views

通讯作者: 陈斌, bchen63@163.com

Article Metrics

Related

Proportional views