基于多帧边缘差异的视频运动对象的分割与跟踪算法

刘龙; 刘贵忠; 王占辉; 王黎明

基于多帧边缘差异的视频运动对象的分割与跟踪算法

计量
- 文章访问数: 2283
- HTML全文浏览量: 125
- PDF下载量: 699
- 被引次数: 0
出版历程
- 收稿日期: 2002-12-08
- 修回日期: 2003-05-19
- 刊出日期: 2004-05-19

Segmentation and Tracking of Video Object of Interest Based on Change of Multi-frames Edge

摘要

摘要: 从视频场景中分割和跟踪感兴趣的视频对象对于MPEG-4等基于对象的视频编码来说是关键性的技术之一。针对目前大部分视频对象分割和追踪算法相当复杂但仍不能有效地去除背景噪声的问题,该文提出用于分割和跟踪视频运动对象的一种基于多帧边缘差异的算法。该算法利用一组帧的边缘差异来提取运动对象区域,通过聚类方法去除背景像素点,利用形态学算子得到对象分割模板,同时通过建立前帧感兴趣对象与当前帧运动对象的帧间向量跟踪当前帧的感兴趣视频对象。不同标准视频测试序列的测试结果表明,该算法能够实现对感兴趣的视频运动对象更为精确、快速和有效地分割和跟踪。
- 边缘差异; 视频对象; 分割; 跟踪
Abstract: It is one of the key technologies to MPEG-4 grade codes based on target to segment and track the concerned video objects from the video scene. Most current seg-mentation and tracking algorithms are of high complexity but not effective in getting rid of background noise. One algorithm is put forward to segment and trace video objects based on edge difference among multiple frames. According to the proposed algorithm, edge difference between a group of frames is used to draw the area of moving objects; then, background pixels are removed through setting up pixel-measuring window and threshold value; the area of objects is set up by morphology operator; at the same time, vectors between the last concerned objects per frame and moving objects of present frame are established to follow current concerned objects. The result of various standard video test sequences shows that the proposed algorithm offers more accurate, faster and more effective segmentation and tracking of concerned video moving objects.

HTML全文

参考文献(1)

Sikora T. The MPEG-4 video standard verification model[J].IEEE Trans. on Circuits and Systems for Video Technology.1997, 7(1):19-30[2]Marques F, Molina C. Object tracking for content-based functionalities, in SPIE Vis. Commun.Image Processing, 1997, vol. 3024: 190-199.[3]C. Gu, Lee M C. Semantic segmentation and tracking of semantic video objects. IEEE Trans.on Circuits and Systems for Video Technology, 1998, 8(5): 572-584.[4]C. Gu. Tracking of multiple semantic video objects for internet applications. SPIE, 1999, vol.3653:806-820,.[5]Thomas M, King N. Automatic segmentation of moving objects for video object plane generation[J].IEEE Trans. on Circuits and Systems for Video Technology.1998, 8(5):525-538[6]Mech R, Wollborn M. Automatic segmentation of movi ng objects (partial results of core experiment N2), ISO/IEC JTC1 /SC2 9/WG11 MPEG98/m3187, 1998.[7]Mech R, Wollborn M. A noise robust method for 2D shape estimation of moving objects in video sequences considering a moving camera[J].Signal Processing.1998, 66(2):203-217[8]Kim M.[J].Choi J G, Lee M H. Performance analysis of an ETRIs global motion compensation and scene cut detection algorithms for automatic segmentation. ISO/ IEC JTC1/SC29/WG11 MPEG97/m238.1997,:-[9]Bouthemy P, Francois E. Motion segmentation and qualitative dynamic scene analysis from an image sequence[J].Int. Journal of Computer Vision.1993, 10(2):157-182[10]Meyer F, Bouthemy P. Region-based tracking in an image sequence, in Proc.European Conf.Computer Vision (ECCV), Rennes, May 1992: 476-484.[11]Canny J F. A computational approach to edge detection. IEEE Trans. on PAMI, 1986, 6(3):679-698.

施引文献

资源附件(0)

访问统计