期刊文献+

一种结合光流法与三帧差分法的运动目标检测算法 被引量:81

A Moving Object Detection Algorithm Based on a Combination of Optical Flow and Three-Frame Difference
下载PDF
导出
摘要 运动目标的检测是计算机视觉研究的重要内容之一,光流法是其中的一种重要方法.由于计算光流的算法复杂,限制了它的使用.本文提出一种结合光流法与三帧差分法的运动目标检测算法,该算法简化了光流的计算,选择图像中具有代表性的Harris角点,只对这些像素点计算光流信息,有效地减少了复杂度,由于检测得到的运动目标区域不够完整,引入了三帧差分法作为简化光流法的补充.经过实验,该方法使光流法达到了实时性要求,取得了好的效果,优于单独运用两种方法中的任何一种取得的效果. Moving objects detection is an important research of computer vision.Optical flow method is an important way,but it is limited to use because of its complexity.A moving object detection algorithm based on a combination of optical flow and the three-frame difference is proposed.The calculation of optical flow is simplified.Harris corners are detected,then the pixels are selected to compute optical flow information,which reduce the algorithm′s complexity.Because the detected moving target area is not complete,three-frame difference method is introduced as a supplement.The experimental results show that the algorithm can achieve real-time and have better results than anyone in these two separate algorithms.
出处 《小型微型计算机系统》 CSCD 北大核心 2013年第3期668-671,共4页 Journal of Chinese Computer Systems
基金 国家自然科学基金项目(61163024)资助 云南省应用基础研究计划项目(2011FB019)资助 云南省教育厅科学研究基金项目(2012C105)资助 云南省2012年大学生创新创业训练计划项目(2012022)资助 2011年云南大学本科生科研项目(2011005)资助
关键词 运动目标检测 光流法 帧差法 角点检测 moving targets detection optical flow frame difference corner detection
  • 相关文献

参考文献9

  • 1Yuan Guo-wu. Research on moving objects detection and tracking methods in intelligent visual surveillance system [ D]. Kunming: Yunnan University ,2012.
  • 2Huang Chao-qun. Motion detection and tracking based on gaussian mixture model and kalman filter [ D ]. Kunming : Yunnan Universi- ty, 2010.
  • 3Harris C, Stephens M. A combined comer and edge detector [ C ]. Proceedings of the 4th Alvey Vision Conference, August 31-Sep- tember 2,1988:147-151.
  • 4Shi Jian-bo,Carlo tomasi. Good features to track[ C ]. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition ( CVPR94), 1994:593-600.
  • 5Lucas B D, Kanade T. An iterative image registration technique with an application in stereo vision E C 1- Proceeding of 7th Interna- tional Joint Conference on Artificial Intelligence, 1981:674-679.
  • 6Gary Bradski, Adrian Kaehler. Learning OpenCV-computer vision with the OpenCV library[ EB/OL]. Free online edition, http ://sa- fail. oreilly, corn ,2011.
  • 7I Computer vision and robotics research laboratory test bed data[ EB/ OL ]. http ://cvrr. ucsd. edu/aton/shadow/, 2011.
  • 8Performance evaluation of tracking and surveillance ( PETS2000 ) [ EB/OL ]. http ://ftp. pets. rdg. ac. uk/pub/PETS2000/,2011.
  • 9The second IEEE international workshop on performance evaluation of tracking and surveillance( PETS2001 ) [- EB/OL ]. http ://www. cvg. cs. rdg. ac. uk/PETS2001/,2011.

同被引文献578

引证文献81

二级引证文献482

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部