摘要
提出了一种新的基于时空信息的视频分割算法,即先将原始图像标记成不同的区域,然后以帧间差分得到的对象运动信息作为评判准则,将这些区域分别归类于前景对象和背景,达到对象分割的目的。特别是在区域标记的过程中,采用了一种新的基于分水岭的区域标识技术。通过对标准图像序列的实验结果可以看到,利用该算法能够较精确地分割出视频对象。
In this paper, we propose a new segmentation method to separate the moving objects from the background in a generic video sequence. It is to support the content-based functionalities focused by the MPEG-4. The proposed algorithm discriminates background from foreground by means of spatio-temporal information. First, the original image is labeled into different regions. Then, some regions are merged to form video object according to the inter-frame difference information as judgement criterion, and others are discarded. Experimental results obtained on some typical MPEG-4 sequences are shown in order to illustrate the segmentation algorithm performance.
出处
《电视技术》
北大核心
2004年第11期4-7,共4页
Video Engineering
基金
国家自然科学基金项目(60172020)
关键词
视频对象分割
形态学
分水岭算法
帧间差分
video objects segmentation
morphology
watershed algorithm
inter-frame difference
