期刊文献+

基于边缘帧差和高斯混合模型的行人目标检测 被引量:12

Detection method for pedestrian target based on Gaussian mixture model and edge frame difference
在线阅读 下载PDF
导出
摘要 针对高斯混合模型存在背景更新收敛性差、易受环境噪声和光照突变影响、易产生虚假目标等问题,提出一种基于高斯混合模型的改进算法,用于视频中行人目标检测。通过将帧差法引入高斯混合模型,快速区分背景区域和运动目标区域,从而提取前景中完整的行人目标。结合视频帧边缘和边缘帧差信息,采用多种模型更新率,提高高斯混合模型对复杂背景的自适应性和快速收敛性,从而消除环境噪声和光照突变的影响,避免检测出虚假目标。实验结果表明,相比传统高斯混合模型,该方法可以有效去除噪声和光照的干扰,收敛性更佳、行人检测效果更鲁棒。 For the Gaussian mixture model(GMM)was weak in convergence,sensitive to ambient noise and sudden light change,and prone to detect false target,this paper proposed an improved algorithm based on GMM for moving pedestrian detection in videos.It introduced the information of frame difference into GMM background model,which quickly distinguished the background region and the moving object region,to extract the complete moving pedestrian in the foreground.Then,the background model mixed with the edge detection and edge frame difference in the video,and adopted different rates of background updating.It improved the adaptability and accelerated the speed of convergence,to eliminate the affection of noise and illumination change,and also removed the false target.The experimental results show that the improved algorithm is more efficient than traditional GMM to remove the influence of noise and illumination,and has better convergence and robustness in the pedestrian object detection.
作者 苏剑臣 李策 杨峰 Su Jianchen;Li Ce;Yang Feng(Dept.of Computer,School of Mechanical Electrical&Information Engineering,China University of Mining&Technology,Beijing 100083,China;State Key Laboratory of Coal Resources&Safe Mining,China University of Mining&Technology,Beijing 100083,China)
出处 《计算机应用研究》 CSCD 北大核心 2018年第4期1246-1249,共4页 Application Research of Computers
基金 煤炭资源与安全开采国家重点实验室开放课题项目(SKLCRSM16KFD04) 国家自然科学基金青年基金资助项目(61601466) 中央高校基本科研业务费专项资助项目(2016QJ04) 国家大学生创新训练资助项目(规划项目)
关键词 高斯混合模型 虚假目标 目标检测 边缘检测 帧差法 Gaussian mixture model false target object detection edge detection frame difference
作者简介 苏剑臣(1992-),男,湖南岳阳人,硕士,主要研究方向为图像处理.;李策(1988-),女(通信作者),山西运城人,讲师,主要研究方向为计算机视觉(celi@cumtb.edu.cn).;杨峰(1968-),男,北京人,教授,主要研究方向为计算机图形图像.
  • 相关文献

参考文献5

二级参考文献32

  • 1王典,程咏梅,杨涛,潘泉,赵春晖.基于混合高斯模型的运动阴影抑制算法[J].计算机应用,2006,26(5):1021-1023. 被引量:20
  • 2Stauffer C, Grmson W E L.Adaptive background mixture mod- els for real-time tracking[C]//Proceeding of IEEE International Conference on Computer Vision and Pattern Recognition,Fort Co- lins, Colorado,USA.[S.l.] : IEEE Computer Society, 1999,2: 246-252,.
  • 3Pavlidis I, Morellas V, Tsiamyrtzis P, et al.Urban surveillance system: From the laboratory to the commercial world[J].Pro- ceedmgs of the IEEE,2001,89(10) : 1478-1497.
  • 4Power P W, Schoonees J A.Understanding background mixture models for foregrounds segmentation[C]//Proceedings of lmage and Vision Computing, New Zealand, Auckland, 2002: 267-271.
  • 5Zhang Yunchu, Liang Zize, Hou Zengguang, et aLAn adaptive mixture gaussian background model with online background re- construction and adjustable foreground .mergence time for mo- tion segmentation[C]//IEEE International Conference on Industri- al Technology,ICIT 2005,14-17 Dec 2005:23-27.
  • 6Duda R O,Hart P E,Stork D G.Pattcm classification[M].2nd cd. New York:John Wiley & Sons,lnc,2001:521-528.
  • 7Chris S, Grimson W E L. Learning Patterns of Activity Using Real-time Tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(8): 747-757.
  • 8Zoran Z. Improved Adaptive Gaussian Mixture Model for Back- ground Subtraction[C] //Proc of the 17th International Conference on Pattern Recognition. Washington D. C., USA: IEEE Computer Society, 2004.
  • 9Lee D S. Effective Gaussian Mixture Learning for Video Back- ground Subtraction[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(5): 827-832.
  • 10Julien P, Christoph S, Pascal F. Making Background Subtraction Robust to Sudden Illumination Changes[C] //Proc. of the 10th European Conference on Computer Vision. Marseille, France: Springer-Verlag, 2008.

共引文献155

同被引文献107

引证文献12

二级引证文献52

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