期刊文献+

基于帧间差分和水平集的运动目标探测跟踪方法 被引量:7

Moving Targets Detection and Tracking Based on Frame Difference and Level Set
下载PDF
导出
摘要 针对运动目标探测中,帧间差分法容易出现目标拉伸和空洞的问题,提出了基于帧间差分和水平集的运动目标探测跟踪方法。该方法首先利用帧差法获取运动目标的大概位置,再利用水平集得到目标的完整区域,最后利用直方图匹配的方式对目标进行关联,实现实时目标跟踪,这样既解决了帧间差分法中目标拉伸和空洞问题,又克服了水平集自身无法利用目标运动信息和初始化不准确时收敛速度慢的问题,提高了探测的准确性和跟踪的稳定性。实验结果表明,该方法能够在实时条件下,实现目标的精确探测和稳定跟踪。 Stretching and hole of moving targets may occur during the targets detection by frame difference method.This paper proposed a detection and tracking method of moving targets based on the frame difference and level set.Firstly,frame difference was used to get the approximate location of a moving target;then,the complete area was obtained by using the level set;finally,histogram matching helped associating with the targets,achieving the real-time tracking.This solution not only solved stretching and hole of frame difference,but also overcame the problems that the level set itself could not use target motion information and slow convergence speed of initializing the inaccurate.The detection accuracy and the stability of targets tracking was improved.The experimental results showed that the method could achieve accurate detection and stable tracking in realtime.
出处 《探测与控制学报》 CSCD 北大核心 2015年第1期27-30,共4页 Journal of Detection & Control
关键词 目标探测 跟踪方法 帧间差分 水平集 直方图匹配 targets detection tracking method frame difference level set histogram matching
  • 相关文献

参考文献8

二级参考文献28

  • 1邵文坤,黄爱民,韦庆.动态场景下的运动目标跟踪方法研究[J].计算机仿真,2006,23(5):181-184. 被引量:30
  • 2万缨,韩毅,卢汉清.运动目标检测算法的探讨[J].计算机仿真,2006,23(10):221-226. 被引量:121
  • 3李俊.基于曲线演化的图像分割方法及应用:博士学位认文[M].上海:上海交通大学,2001.. 被引量:1
  • 4Fukunaga, Keinosuke, Hostetler, D Larry. Estimation of the gradient of a density function and its applications in pattern recognition [J]. American Society of Mechanical Engineers (Paper) , 1973: 149 - 158. 被引量:1
  • 5BY oers, H Driessen. A particle filter based detection scheme[ J].IEEE SP Letters, 2003,10 ( 10 ) : 300 - 302. 被引量:1
  • 6Eduardo Parrilla, etc. Handling occlusion in optical flow algorithms for object tracking[ J]. Computers & Mathematics with Applications, 2008,56 (3) :733 - 742. 被引量:1
  • 7B K P Horn, B G Schunck. Determining optical flow [ J ]. Artificial Intelligence, 1981,17( 1 -3) :85 -203. 被引量:1
  • 8B D Lukas, T Kanade. An iterative image registration technique with an application to stereovision[ C ]. Proceedings of Imaging Understanding Workshop, 1981:121 -130. 被引量:1
  • 9Fabrice Bourel, Claude C Chibelushi, Adrian A Low. Recognition of facial expressions in the presence of occlusion[ C ]. Proceedings of the Twelfth British Machine Vision conference, 2001,1 ( 1 ) :213 - 222. 被引量:1
  • 10E Parrilla, J Riera, Marcos H Gimenez, Juan R Torregrosa. Vehicle tracking in urban routes by means of Lucas & Kanade algorithm[ C]. Proceedings of the International Conference on Computational and Mathematical Methods in Science and Engi- neering, 2005:438 -445. 被引量:1

共引文献233

同被引文献49

  • 1邓玉春,姜昱明,张建荣.视频序列图像中运动对象分割综述[J].计算机应用研究,2005,22(1):8-11. 被引量:12
  • 2赵天飙,陶波尔,董希超,唐蒙昌.啮齿动物种群数量调查方法及其评价[J].中国媒介生物学及控制杂志,2007,18(4):332-334. 被引量:16
  • 3蔡梅艳,吴庆宪,姜长生.改进Otsu法的目标图像分割[J].电光与控制,2007,14(6):118-119. 被引量:48
  • 4Koshelev V I, Kozlov D N. Wire recognition in image within aerial inspection application. 2015 4th Mediterranean Conference on Em- bedded Computing ( MECO 2015 ) Montenegro, Budva: IEEE, 2015 : 159-162. 被引量:1
  • 5Lowe D G. Distinctive image features from seale-invariant keypoints. International Journal of Computer Vision, 2004; 60 (2) : 91-110. 被引量:1
  • 6Bay H,Van G L, Tuytelaars T. SURF: speeded up robust features. 9th European Conference on Computer Vision ( ECCV 2006). Graz, Austria: IEEE, 2006 : 404-417. 被引量:1
  • 7Rublee E, Rabaud V, Konolige K, et al. ORB: an efficient alterna- tive to SIFT or SURF. 2011 IEEE International Conferenee on Com- puter Vision( ICCV 2011 ). Barcelona, Spain : IEEE, 2011 : 2564- 2571. 被引量:1
  • 8Kumar N S, Shobha G, Balaji S. Key frame extraction algorithm for video abstraction application in underwater videos. 2015Underwater Technology(UT 2015). Kamataka, India: IEEE, 2015. 被引量:1
  • 9Tao S Y, Mei D X, Hong Z. Implement of neural network predictive controller based on Matlab and VC + + mixed programming. 2015 Control and Decision Conference ( CCDC 2015 ). Qingdao China: IEEE, 2015. 被引量:1
  • 10陈磊,邹北骥.基于动态阈值对称差分和背景差法的运动对象检测算法[J].计算机应用研究,2008,25(2):488-490. 被引量:28

引证文献7

二级引证文献42

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