Effective method for tracking multiple objects in real-time visual surveillance systems

作     者：Wang Yaonan Wan Qin Yu Hongshan 

作者机构：Coll. of Electrical and Information Engineering Hunan Univ. Changsha 410082 P. R. China 

出 版 物：《Journal of Systems Engineering and Electronics》 (系统工程与电子技术（英文版）)

年 卷 期：2009年第20卷第6期

页      面：1167-1178页

核心收录：

学科分类：08[工学] 080401[工学-精密仪器及机械] 080203[工学-机械设计及理论] 0804[工学-仪器科学与技术] 080402[工学-测试计量技术及仪器] 0838[工学-公安技术] 0802[工学-机械工程] 

基　　金：supported by the National Natural Science Foundation of China(60835004 60775047 60872130) the National High Technology Research and Development Program of China(863 Program)(2007AA04Z244 2008AA04Z214) 

主　　题：visual surveillance multiple object tracking object model matching matrix. 

摘      要：An object model-based tracking method is useful for tracking multiple objects, but the main difficulties are modeling objects reliably and tracking objects via models in successive frames. An effective tracking method using the object models is proposed to track multiple objects in a real-time visual surveillance system. Firstly, for detecting objects, an adaptive kernel density estimation method is utilized, which uses an adaptive bandwidth and features combining colour and gradient. Secondly, some models of objects are built for describing motion, shape and colour features. Then, a matching matrix is formed to analyze tracking situations. If objects are tracked under occlusions, the optimal ＂visual＂ object is found to represent the occluded object, and the posterior probability of pixel is used to determine which pixel is utilized for updating object models. Extensive experiments show that this method improves the accuracy and validity of tracking objects even under occlusions and is used in real-time visual surveillance systems.