Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (14): 175-177.

• 图形、图像、模式识别 • Previous Articles     Next Articles

Compressing method of monitored video based on ridgelet motion detection

LI Zong-jian1,2,ZENG Li1,3,GUO Yan1,3   

  1. 1.ICT Research Center,Key Lab of Optoelectronic Tech.and Sys.of the Education Ministry of China,Chongqing Univ.,Chongqing 400044,China
    2.College of Optoelectronic Engineering,Chongqing Univesity,Chongqing 400044,China
    3.College of Mathematics and Physics,Chongqing University,Chongqing 400044,China
  • Received:2007-12-25 Revised:2008-02-22 Online:2008-05-11 Published:2008-05-11
  • Contact: LI Zong-jian

基于脊波运动检测的监控视频压缩方法

李宗剑1,2,曾 理1,3,郭 燕1,3   

  1. 1.重庆大学 光电技术及系统教育部重点实验室 ICT研究中心,重庆 400044
    2.重庆大学 光电工程学院,重庆 400044
    3.重庆大学 数理学院,重庆 400044
  • 通讯作者: 李宗剑

Abstract: In this paper,the authors research a compressing method of monitored video based on ridgelet motion detection.First,judge the frame whether with interesting motion change or not.If having,compress the frame with low loss or lossless.Else,compress the frame with high rate.Judging the motion change,research the hybrid method of ridgelet and background subtraction method.The bybrid method can position the frame with interesting motion change and having the following advantages:less compute complexity,less manual intervening,less illumination changes influence.

Key words: monitored video compression, motion detection, ridgelet, background subtraction method

摘要: 研究了一种基于运动检测的监控视频压缩方法,即先判断出相对于背景场景来说具有感兴趣运动变化发生的图像帧,再对判断出的图像帧进行低损失或无损失的压缩,剩余的图像帧进行高比率的压缩。为判断视频帧有无感兴趣运动变化发生,研究出基于新的多分辨率分析工具——脊波和背景减除法的混合方法,该方法能有效地定位出具有感兴趣运动变化发生的图像帧,并具有计算量较小、人工干预较少、受光照变化影响较小的优点。

关键词: 监控视频压缩, 运动检测, 脊波, 背景减除法