摘要
运动目标的检测与跟踪是计算机视觉和图像编码研究的主要内容,有着广泛的应用领域。基于光流场的检测与跟踪是其常用的方法之一,但其计算结果和效率是不能令人满意的。提出了基于帧间差阈值法和光流场相结合的快速运动目标检测与跟踪算法,计算结果表明该算法简单实用、运算速度快,克服了单纯光流场方法的不足。
Detection and tracking of moving objects are the main contents of researching for computer vision and image encoding,which has an extensive applicative prospect.One of their common methods is based on the detection and tracking of optical flow field,whose calculation results and efficiency are unsatisfying.This paper proposes a fast algorithm for detection and tracking of the moving objects based on the combination of inter-frame difference and optical flow field,whose calculation results show that the algorithm has the advantages of simplicity,practicality and fast operating speed,which overcomes the shortcoming of pure optical flow field.
出处
《上海电机学院学报》
2005年第6期52-55,共4页
Journal of Shanghai Dianji University
关键词
目标跟踪
目标检测
跟踪算法
光流场
target tracking
target detection
tracking algorithm
optical flow field
