提出了基于递归高阶统计的运动目标分割方案 ,主要用于会议电视图像序列 .与传统的高阶统计方法相比 ,基于递归高阶统计的分割方法从高斯噪声背景中提取信息 ,由于利用了前 n帧图像的信息冗余度 ,递归高阶统计方法可以更有效地减少噪声、检测小目标 ,并且提高了目标的分割准确度 .模拟试验结果显示 ,分割结果较好 This paper proposes a motion target segmentation scheme based on recursive high-order statistics, which is mainly used in the video sequence of conferencing video.Compared with the traditional high-order statistics methods, the segmentation method based on recursive high-order statistics extracts information from Gaussian noise background, The information redundancy of the first n frames and the recursive high-order statistics method can reduce the noise more effectively, detect the small target and improve the segmentation accuracy of the target.The simulation results show that the segmentation result is better