摘要
为了解决复杂背景下运动点目标的检测和跟踪问题,本文提出了一种基于图像差分和聚类的运动目标检测和跟踪算法.该算法首先根据图像配准的方法,对序列图像进行差分运算,提取出候选的运动目标.在此基础上,利用运动目标在空间和时间上的相关性以及运动目标的轨迹所具有的连续性,采用一种特殊的聚类方法,从噪声环境中正确检测出运动目标的轨迹,并实现对运动目标的跟踪.实验表明该算法能快速检测出复杂背景下的运动点目标,并能有效处理轨迹相交和检测过程中出现新目标的情况.
To solve the moving point target detection problem under the condition of complex background, a new algorithm based on image difference and clustering method is presented in this paper. First, the spatial and perspective relations among the sequential images are calculated according to the corresponding feature points. Then by using the method of image registration, the successive two images are differenced to find the possible moving point targets. Since the possible targets are very small in the images, they may be submerged in the noise. Thus, to pick out the targets from the noise, a special clustering method and the corresponding algorithm are introduced, which use the different characters between the targets and noise:compared with noise, the targets move in the space continuously, having spatial and time correlations in the image sequence. The experimental results show that the algorithm can effectively and reliably detect the targets under the complex background. The algorithm can also work very well with some special situations, such as the situations of crossed tracks, new tracks during the tracking process.
出处
《模式识别与人工智能》
EI
CSCD
北大核心
2004年第3期342-346,共5页
Pattern Recognition and Artificial Intelligence
关键词
点目标
运动目标检测
图像配准
Point Target
Moving Target Detection
Image Registration