计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (7): 184-186.DOI: 10.3778/j.issn.1002-8331.2010.07.056

• 图形、图像、模式识别 • 上一篇    下一篇

基于视觉注意特征和SVM的镜头边界检测算法

陈 萍1,李秀强2,肖国强2,江健民2   

  1. 1.黄淮学院 计算机科学系,河南 驻马店 463000
    2.西南大学 计算机与信息科学学院,重庆 400715
  • 收稿日期:2008-09-08 修回日期:2009-06-29 出版日期:2010-03-01 发布日期:2010-03-01
  • 通讯作者: 陈 萍

Shot boundary detection using SVM on visual attention features

CHEN Ping1,LI Xiu-qiang2,XIAO Guo-qiang2,JIANG Jian-min2   

  1. 1.Department of Computer Science,Huanghuai University,Zhumadian,Henan 463000,China
    2.College of Computer and Information Science,Southwest University,Chongqing 400715,China
  • Received:2008-09-08 Revised:2009-06-29 Online:2010-03-01 Published:2010-03-01
  • Contact: CHEN Ping

摘要: 镜头边界检测是视频分析的基础。借鉴心理学中有关视觉注意的研究成果,提出了一种采用符合人类视觉注意的特征,并利用支持矢量机进行视频镜头边界检测的算法。通过对TRECVID2007数据库进行实验的结果表明,该算法在查全率和查准率方面都获得了较好的性能。

关键词: 镜头边界检测, 视觉注意特征, 支持矢量机

Abstract: Shot boundary detection is the basis of video analysis.This paper proposes an algorithm of shot boundary detection,which employs Support Vector Machine(SVM) to classify visual attention features based on the research results of psychology.Extensive experiments carried out on TRECVID2007 database show that the proposed approach works well in detecting shot boundary measured by both recall and precision.

Key words: shot boundary detection, visual attention feature, Support Vector Machine(SVM)

中图分类号: