利用区域变形和背景更新实现运动对象跟踪.pdfVIP

第 17 卷第5 期 计算机辅助设计与图形学学报 Vo1. 17 , No.5 2005 年 5 月 JOURNAL OF COMPUTER-AIDED DESIGN COMPU丁ER GRAPHICS May , 2005 利用区域变形和背景更新实现运动对象跟踪 袁晓彤郭礼华杨树棠 (上海交通大学信息、安全工辍学院 200030) (yxt@sjtu. edu. cn) 摘 要 从时域统计的角度出发，提出了一种结合自适应混合背景更新模型的区域变形跟踪算法.该算法以模型更 新得到的前景/背景二值分割掩膜作为区域特征，将跟踪问题抽象为一个水平集(Leve! S巳t) 偏微分方程的数值求解 问题，并分析了算法的自适应性.为了进一步提高算法的实现效率，寻|入了窄带跟踪方案.实验表明，该算法可以对 视频序列中的指定运动对象进行快速精确的跟踪. 关键诩 运动对象跟踪;区域变形;背景建模1 动态轮廓;水平集;偏微分方程;窄带法 中图法分类号丁P391.41 Moving Object Tracking Based on Region Deformation and Background Modeling Yuan Xiaotong Guo Lihua Yang Shutang (Schoo[ of lnformation Security Engineering , ShangHai Jiaοtong Uni四月ity ， Shanghai 200030) Abstract We address the pr咄lem of tracking moving objects through a video sequence using deformable regions. 50 far , numerous tracking approaches based on region deformation have been proposed and the chosen region-based features are mostly special visual information such as mearνcolor and texture. This kind of information may be misleading in some dense visual clutter. ln this paper , we pr巳sent a temporal statistical tracking algorithm that extracts the region-based features via adaptive background mixture model. Our contribution is to define a new tracking criterion combining geometrical and motional features of the region. The resultant algorithm is expressed in the form of a level set partial differential equation. We also analyzed the adaptive ness of our algorithm. To further improve the computational efficiency , we introduce the narrowband method into our tracking scheme. Finally , highly promising experimental results are provided to illustrate the efficiency and accuracy of our method. Key words moving object tracking; region deformation; background modeling; active contours; l