序列图像中运动目标检测与跟踪方法分析  

Analysis of Moving Object Detection and Tracking in Sequence Image

作　　者：罗志升[1] 王黎[1] 高晓蓉[1] 王泽勇[1] 赵全轲[1] 

机构地区：[1]西南交通大学理学院,四川成都610031

出　　处：《现代电子技术》2009年第11期125-128,共4页Modern Electronics Technique

摘　　要：运动目标检测与跟踪是计算机视觉研究和图像编码研究的主要内容,有着广泛的应用领域。研究了目前运动对象检测与跟踪的一些常用方法,包括时域差分法、背景差分法、基于光流场的检测方法和卡尔曼滤波、特征光流法的跟踪方法,并对各种方法进行了比较,指出其优缺点及适用范围,并给出了时域差分及背景差分方法的实验结果。Detection and tracking of moving objects are the main contents of researching for computer vision and image encoding,which has an extensive applicative prospect. In this paper, some present commonly used moving object detection and tracking methods including the detection methods of the time domain differential method, the background differential method, optical flow method and tracking methods of the Kalman filter,feature optical flow method are studied. The comparison to each method is carried out,its advantages and disadvantages as well as applicable scope are pointed out. And the experimental results of the time domain differential method and the background differential method are given.

关 键 词：运动目标 时域差分 背景差分 光流法 卡尔曼滤波 特征光流法 

分 类 号：TN919.8[电子电信—通信与信息系统]