运动目标跟踪系统的遮挡问题处理.pdfVIP

( ) 计算机工程与设计 Computer Engineering and Design 2816 2009,30 11 计算机工程与设计 Computer Engineering and Design 开发与应用 运动目标跟踪系统的遮挡问题处理 郑可飚， 黄文清， 张佐理， 李艳芳 (浙江理工大学，浙江杭州310018) ( ) 摘 要：遮挡是运动目标跟踪研究中的一个重要问题，介绍了基于Markov Chain Monte Carlo MCMC 的运动物体遮挡问题解 决方法。该方法通过建立贝叶斯模型，确定先验概率和条件概率，将车辆分割问题看成求后验概率最大时的车辆状态；然后 运用MCMC 方法对后验概率进行估计，设计MCMC 标准对后验概率进行采样，用长方形模型来近似车辆外形。实验证明 MCMC 方法在不需对车辆单独初始化的前提下能有效的将相互遮挡的车辆分割出来，检测出车辆之间的相互遮挡。 关键词：运动目标跟踪; MCMC; 分割; 遮挡; 贝叶斯模型; meanshift 中图法分类号：TP391.4 文献标识码：A 文章编号：1000-7024 (2009) 11-2816-03 Occlusion handling in moving obj ect tracking system ZHENG Ke-biao, HUANG Wen-qing, ZHANG Zuo-li, LI Yan-fang (Zhej iang Science Technology University, Hangzhou 310018, China) Abstract ：Occlusion is a significant problem for the research of moving obj ect tracking. A MCMC-based method to solve the problem of occlusion is introduced. At first building Bayesian model, determine the prior probability and likelihood term. Formulate the vehicle segmentation problem as searching for the vehicle state, which maximizes the posterior probability. Then MCMC is applied to posterior probability estimation. A standard MCMC is designed to sample the posterior probability, and rectangle model is used to approximation the vehicle. Tests prove the MCMC-based method can segment multiple merged vehicles into individual effective.