计算机应用 ›› 2013, Vol. 33 ›› Issue (10): 2918-2921.

• 多媒体技术 • 上一篇    下一篇

基于前—后向光流点匹配运动熵的视频抖动检测算法

江爱文,刘长红,王明文   

  1. 江西师范大学 计算机信息工程学院, 南昌 330022
  • 收稿日期:2013-04-25 修回日期:2013-06-12 出版日期:2013-10-01 发布日期:2013-11-01
  • 通讯作者: 江爱文
  • 作者简介: 
    江爱文(1984-),男,江西景德镇人,副教授,博士,CCF会员,主要研究方向:模式识别、图像处理;刘长红(1977-),女,江西南丰人,博士,主要研究方向:智能视频分析;王明文(1964-),男,江西南康人,教授,博士,CCF高级会员,主要研究方向:智能信息处理。
  • 基金资助:
    国家自然科学基金资助项目;国家自然科学基金资助项目;江西省教育厅科技项目

Video jitter detection algorithm based on forward-backward optical flow point matching motion entropy

JIANG Aiwen,LIU Changhong,WANG Mingwen   

  1. School of Computer and Information Engineering, Jiangxi Normal University, Nanchang Jiangxi 330022, China
  • Received:2013-04-25 Revised:2013-06-12 Online:2013-11-01 Published:2013-10-01
  • Contact: JIANG Aiwen

摘要: 目前大部分视频监控系统面临着高效实时性智能分析与低效滞后的人工故障排查的矛盾。视频质量智能诊断系统可以为此提供有效的解决方案。针对视频质量诊断系统中的画面抖动异常检测问题,提出一种简单有效的实用算法。该算法通过有效融合图像的稀疏光流与特征点匹配算法,根据前向-后向误差标准估计图像帧的全局运动参数,引入连续帧的运动熵用于衡量视频画面片段运动的混乱程度,判断是否存在视频抖动现象。算法在不同分辨率的实际监控录像数据集上进行了测试和比较。实验证明,该算法在一定程度上克服了大位移抖动的影响,具备良好的实时特性以及较高的检测精度,能够满足实际工作的需求。

关键词: 视频监控, 抖动检测, 前-后向误差, 金字塔Lucas-Kanade光流, 运动熵

Abstract: The conflicts between the real-time, efficient intelligent analysis and the inefficient, laborious trouble shooting, which are faced by most of video surveillance systems, can be resolved by Intelligent Video Quality Detection System (IVQDS). As a part of IVQDS, video jitter detection algorithm was focused in this paper. In the proposed method, sparse optical flow features were fused together with interest point matching algorithm; correctly matched point-set which was reliably detected according to the forward-backward error criterion, was used to estimate the global motion parameters, from which motion entropy was computed to measure the motion homogeneity of the video fragment. The experimental results tested on realistic surveillance video records have shown that the proposed algorithm can work under real-time environment against the effects from big movements with high detection performance.

Key words: video surveillance, jitter detection, forward-backward error, pyramid Lucas-Kanade optical flow, motion entropy

中图分类号: