基于模糊聚类的视频对象分割  

Video Object Segmentation Based on Fuzzy Clustering

作　　者：胡双演[1] 李俊山[1] 李旭辉[1] 黄百钢[1] 

机构地区：[1]第二炮兵工程学院,陕西西安710025

出　　处：《微电子学与计算机》2008年第7期1-4,共4页Microelectronics & Computer

基　　金：国家自然科学基金项目(60572080;60772151)

摘　　要：提出了一种基于模糊聚类的视频对象分割方法.首先通过对连续三帧视频图像进行二次差分来得到二次差分图像;然后估计噪声的特征参数滤除背景噪声,提取出视频对象的运动区域;再利用改进的FCM聚类算法对二次帧差图像中的视频对象运动区域进行空域分割,对空域分割结果进行形态学处理,得到视频对象掩模;最终获得较为理想的视频对象.实验结果表明,该算法能够较为准确地分割出视频对象,并且在空间准确度上占优.This paper proposed a method of video object segmentation based on fuzzy clustering. At first, gain the twi - difference image from the three consecutive video frames. Then, eliminate the noise of background with the estimation of the feature parameter and extract the video object motion area. Afterward, employ the improved FCM clustering method to segment the motion area and get the video object mask by processing the prior result with morphological method. Finally, acquire the ideal video object. Experimental results show that the proposed method performs excellently for video object segmentation and preferable the method in spatial accuracy.

关 键 词：变化检测 噪声估计 视频对象分割 模糊C均值聚类 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]