基于光流场与EM算法的运动目标检测

Moving target detection based on optical flow field and EM algorithm

针对光流法对检测目标的不准确性、抗噪能力弱等缺陷,提出了光流场和EM算法相结合的方法,利用两帧图片求得光流场,并在此基础上利用EM算法对属于背景和目标运动场的光流进行聚类,得到完整的目标,区分出前景和背景,消除噪声.最后通过仿真实验可知,所提出的改进算法对单目标、多目标以及被遮挡的目标的检测都具有良好的效果.

Due to the fact that the optical flow method for the target detection is not accurate,and weak in anti-noise,we propose a method based on a combination with optical flow field and the EM algorithm. We can obtain optical flow field from two pictures. On the basis of the optical flow field,we use the EM algorithm to cluster the optical flow which belong to the background and the motion field of objectives,as a result,we can obtain a whole objective and distinguish the foreground from background,eliminate noise. Through the experimental results,it shows that the modified algorithm has a good effect on the detection of single object,multi-object or covered-object.