Similarity Measure and Shadow Casting Method for Object Tracking

A real-time surveillance system is especially important in providing accurate information by extracting features by recognizing objects from given images. This paper presents a shadow casting method to extract an exact feature vertor about an object, and a color relationship feature vector that can precisely track objects in occlusive and disocclusive situations. The shadow casting method is applied to improved objects recognized as a person. Also, Reliable color relationship feature vector extraction is inevitable for event detection and the precise tracking of objects. The proposed tracking method had an outstanding performance using to count any weakness of the vector. As a result of the experimental images, the proposed method allows for precise trajectory tracking even in cases of the occlusive and disocclusive situations of multi-objects.

[1]  A. Murat Tekalp,et al.  Non-rigid object tracking using performance evaluation measures as feedback , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

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

[3]  G. T. Kogut,et al.  Maintaining the identity of multiple vehicles as they travel through a video network , 2001, Proceedings 2001 IEEE Workshop on Multi-Object Tracking.

[4]  Mubarak Shah,et al.  Tracking and Object Classification for Automated Surveillance , 2002, ECCV.

[5]  Hironobu Fujiyoshi,et al.  Moving target classification and tracking from real-time video , 1998, Proceedings Fourth IEEE Workshop on Applications of Computer Vision. WACV'98 (Cat. No.98EX201).

[6]  Kentaro Toyama,et al.  Wallflower: principles and practice of background maintenance , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[7]  Larry S. Davis,et al.  W4: Real-Time Surveillance of People and Their Activities , 2000, IEEE Trans. Pattern Anal. Mach. Intell..