Journal of Beijing University of Posts and Telecommunications

  • EI核心期刊

JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM ›› 2009, Vol. 32 ›› Issue (5): 109-113.doi: 10.13190/jbupt.200905.109.gaojx

• Reports • Previous Articles     Next Articles

Moving Shadow Detection by Ellipsoidal Method Using Illumination Invariants

GAO Jun-xiang,DU Hai-qing,LIU Yong   

  1. (School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China)
  • Received:2008-12-23 Revised:2009-05-31 Online:2009-10-28 Published:2009-10-28
  • Contact: GAO Jun-xiang

Abstract:

To detect moving shadow in intelligent video surveillance system, a method using illumination invariants c1c2c3is presented. Firstly, the coordinate
systems are defined on the basis of c1c2c3ratios of foreground and b
ackground pixels at the same position. Secondly, the moving shadow detection can be treated as a classification problem of shadows and moving objects in abovementioned coordinate systems. An ellipsoid is further constructed according to the scattering feature of shadow pixels. All the pixels inside the ellipsoid are extracted as shadows. Finally, the postprocessing is performed by exploiting the geometrical property of shadows. Experiments show that the proposed algorithm is robust and effective in detecting shadows for a variety of scenes and moving objects; moreover, it is suited for varying illumination conditions.

Key words: pattern recognition, moving shadow detection, intelligent video surveillance