Fast Background Subtraction and Shadow Elimination Using Improved Gaussian Mixture Model

Background subtraction is widely used to detect moving object from static cameras. It is usually regard as one of the most important step in applications such as traffic monitoring, human motion capture and recognition, video surveillance, etc. In order to get a good performance of the whole system, the background subtraction method could not be so time and space consuming, and the accuracy is also required. Gaussian mixture model is a robust background subtraction method and is widely used ever since it is proposed. Some of the shortcomings of this model such as slow updating rate, slow initialization procedure and time and space consuming can be seen in some literatures and the corresponding resolution methods are also proposed. In this paper, an improved Gaussian mixture model is proposed to save time and space. New shadow detection and noise removing method are also proposed. the accuracy is also required.

[1]  Berna Erol,et al.  A Bayesian framework for Gaussian mixture background modeling , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[2]  D. Koller,et al.  Towards robust automatic traffic scene analysis in real-time , 1994, Proceedings of 1994 33rd IEEE Conference on Decision and Control.

[3]  Alessandro Leone,et al.  A shadow elimination approach in video-surveillance context , 2006, Pattern Recognit. Lett..

[4]  Mohan M. Trivedi,et al.  Detecting Moving Shadows: Algorithms and Evaluation , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  W. Eric L. Grimson,et al.  Adaptive background mixture models for real-time tracking , 1999, Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149).

[6]  Alex Pentland,et al.  Pfinder: Real-Time Tracking of the Human Body , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Viktor Öwall,et al.  Background Segmentation Beyond RGB , 2006, ACCV.

[8]  Zize Liang,et al.  An adaptive mixture Gaussian background model with online background reconstruction and adjustable foreground mergence time for motion segmentation , 2005, 2005 IEEE International Conference on Industrial Technology.

[9]  Kazuhiko Sumi,et al.  Background subtraction based on cooccurrence of image variations , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[10]  Nicolas Martel-Brisson,et al.  Moving cast shadow detection from a Gaussian mixture shadow model , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[11]  P. KaewTrakulPong,et al.  An Improved Adaptive Background Mixture Model for Real-time Tracking with Shadow Detection , 2002 .

[12]  Rita Cucchiara,et al.  Detecting objects, shadows and ghosts in video streams by exploiting color and motion information , 2001, Proceedings 11th International Conference on Image Analysis and Processing.

[13]  Rita Cucchiara,et al.  Detecting Moving Objects, Ghosts, and Shadows in Video Streams , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  Sergio A. Velastin,et al.  Automatic congestion detection system for underground platforms , 2001, Proceedings of 2001 International Symposium on Intelligent Multimedia, Video and Speech Processing. ISIMP 2001 (IEEE Cat. No.01EX489).

[15]  Larry S. Davis,et al.  Non-parametric Model for Background Subtraction , 2000, ECCV.