A New Approach of Dynamic Background Modeling for Surveillance Information

This paper presents a new approach of best background modeling for surveillance information. The approach makes orthogonal non-separable wavelet transformation of information frames used for background modeling, extracts the approximate information to reconstruct information frames, filters out the disturbance, shadow and noise from the reconstructed frames, constructs basic background with the method of binary mask images, uses multi-frame combination of non-uniform noise to filter noise in basic background, applies mutual information to detect the situation of adjacent changes. If the background has a gradual change, weighted superposition of multi background modeling images with time will be applied to update the background. If the background has a major or sudden change, the background will remodel from this frame.

[1]  Visvanathan Ramesh,et al.  Error analysis of background subtraction , 2000, CVPR 2000.

[2]  Jelena Kovacevic,et al.  Nonseparable multidimensional perfect reconstruction filter banks and wavelet bases for Rn , 1992, IEEE Trans. Inf. Theory.

[3]  W. Eric L. Grimson,et al.  Learning Patterns of Activity Using Real-Time Tracking , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Naoya Ohta,et al.  A statistical approach to background subtraction for surveillance systems , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[5]  Ming-Jun Lai,et al.  Examples of bivariate nonseparable compactly supported orthonormal continuous wavelets , 2000, IEEE Trans. Image Process..

[6]  Jake K. Aggarwal,et al.  Simultaneous tracking of multiple body parts of interacting persons , 2006, Comput. Vis. Image Underst..

[7]  Larry S. Davis,et al.  Hydra: multiple people detection and tracking using silhouettes , 1999, Proceedings 10th International Conference on Image Analysis and Processing.

[8]  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).

[9]  Mubarak Shah,et al.  A hierarchical approach to robust background subtraction using color and gradient information , 2002, Workshop on Motion and Video Computing, 2002. Proceedings..

[10]  Azriel Rosenfeld,et al.  Tracking Groups of People , 2000, Comput. Vis. Image Underst..

[11]  Roman Goldenberg,et al.  A real-time system for classification of moving objects , 2002, Object recognition supported by user interaction for service robots.