期刊文献+

基于透视投影变换与动态候选策略的目标交接算法 被引量:2

Target handoff algorithm based on perspective projection transformation and dynamic candidate strategy
下载PDF
导出
摘要 提出使用透视投影变换与动态候选目标策略相结合的算法来实现多摄像机可靠的目标交接。首先提出使用透视变换来获得多摄像机画面的对应关系,然后采用最短距离法计算出多摄像机之间的对应目标。但由于摄像机镜头存在畸变系数,或受光照、图像背景干扰使得目标提取不准确,使目标投影到另一个摄像机画面的位置发生误差;特别当多个目标之间的距离较近时,采用最短距离法进行目标交接会产生多个候选目标。因此提出新的动态候选策略,根据目标运动信息的先验知识和颜色外观模型建立最大似然相似函数,获得特征相似度值最大的目标作为对应目标。实验结果证明,该适应场景复杂度的动态候选策略具有较强的鲁棒性。 Reliable target handoff is a challenging task. A novel target-handoff based on perspective projection transformation and dynamic candidate strategy was proposed. Firstly, the relationship between two cameras was computed for projection transformation. Then the shortest distance method for indentifying the corresponding targets among multi-cameras was designed. However, there is more than one candidate in case that multiple targets on the camera are closer or perspective projection matrix is inaccurate. A dynamic candidate strategy was proposed for recognizing the corresponding targets by the target-motion information of prior knowledge and color appearance model. Experimental results demonstrate that the dynamic candidate strategy was adapted to complexity scene, having stronger robustness.
出处 《计算机应用》 CSCD 北大核心 2013年第A01期170-173,共4页 journal of Computer Applications
基金 广西自然科学基金资助项目(2010GXNSFC013014) 2010广西科学研究与技术开发计划项目(桂科攻10100001-9)
关键词 多摄像机 透视投影变换 动态候选策略 目标交接 multi-camera perspective projection transformation dynamic candidate strategy target handoff
  • 相关文献

参考文献15

  • 1FANG Y, WANG H Y,MAO S, et al. Multi-object tracking basedon region corresponding and improved color-histogram matching[C] // 2007 IEEE International Symposium on Signal Processing andInformation Technology. Piscataway: IEEE,2007:1 -4.
  • 2CHITALIYA N G,TRJVEDI A I. Novd block matching algorithm usingpredictive motion vector for video object tracking based on color histo-gram [C]// 2001 3rd International Conference on ELectronics ComputerTechnology. Piscataway: IEEE*2011,5:81 -85.
  • 3MAALOUF A, LARABI M. Robust foveal wavelet-based objecttracking[ C ]// 2012 IEEE Intematicmal Conference on Acoustics,Speech and Signal Processing. Piscataway: IEEE, 2012: 1489 -1492.
  • 4KIM T, LEE S,PAIK J. Combined shape and feature-based videoanalysis and its application to non-rigid object trackingf J]. IET Im-age Processing, 2011,5( 1): 87 -100.
  • 5KONIGS A. Fast visual people tracking using a feature-based peopledetector[ C] // 2011 IEEE/RSJ International Conference on Intelli-gent Robots and Systems. Pitscataway: IEEE, 2011: 3614 -3619.
  • 6LEE D, CHOI S. Multisensor fusion-based object detection andtracking using active shape model[ C] // 2011 6th International Con-ference on Digital Information Management. Pitscataway: IEEE,2011:108-144.
  • 7JAVED O, KHAN S, RASHEED Z. Camera handoff: Tracking in mul-tiple uncalibrated stationary cameras[ C]// Proceedings of Workshop onHuman Motion, 2000. Pitscataway: IEEE, 2000: 113 -118.
  • 8UN D-T. Collaborative pedestrian tracking and data fusion withmultiple cameras[ J]. Information Forensics and Security, 2011,6(4): 1432-1444.
  • 9牛恩国,程咏梅,潘泉,赵春晖,杨涛.多摄像机视野分界线快速自动生成算法[J].计算机应用研究,2007,24(3):297-299. 被引量:6
  • 10LI J R. The application and modeling of the Levenberg-Marquardtalgorithm[C]// 2010 2nd International Conference on E-Businessand Information System Security. Pitscataway: IEEE, 2010:1-3.

二级参考文献11

  • 1KHAN S,SHAH M. Consistent labeling of tracked objects in multiple cameras with overlapping fields of view[ J]. IEEE Transactions on Pattem Analysis and Machine Intelligence,2003,25 ( 10 ) : 1355-1360.
  • 2VELIPASALAR S, WOLF W. Multiple object tracking and occlusion landling by information exchange between uncalibrated cameras:IEEE International Conference on Image Processing[ C ]. [ S. I. ] :[ s.I. ] ,2005:418-421.
  • 3KHAN S, JAVED O, RASHEED Z. Human tracking in multiple cameras: proceedings of the 8th IEEE International Conference on Computer Vision [ C ]. Vancouver: [ s. n. ],2001:331 - 336.
  • 4VELIPASALAR S, WOLF W. Recovering field of view lines by using projective invadants :2004 International Conference on Image Precesing[C]. [S.I. ]:[s.n. ],2004:3069-3072.
  • 5GULER S, GRIFFITH J M, PUSHEE I A. Tracking and handoff beween multiple perspective camera views:proceedings of the 32nd Applied Imagery Pattern Recognition Workshop [ C ]. [ S. 1. ] : [ s. n. ],2003:275-281.
  • 6HAN Mei,XU Wei,TAO Hai,et al. An algorithm for multiple object tajectory tracking:proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition[ C]. [ S. 1. ] : [ s.n. ] ,2004:864-871.
  • 7FERRER C C, CSSAS J R, PARDAS M. Towards a Bayesian approach to robust finding correspondence in multiple view geometry enironments. International Conference on Computational Science [ C ].[S.1.]:[s.n.],2005:281-289.
  • 8ZHAO Tao, AGGARWAL M, KUMAR R, et al. Real-time wide area lulti-camera stereo tracking: proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition [ C ].[S. 1. ] :[s. n. ] ,2005:976-983.
  • 9KELLY P H, KATKERE A, KURAMURA D Y, et al. An architecture for multiple perspective interactive video: proc. of the 3nd ACM International Conference on Multimedia [ C ]. San Francisco: [ s. n. ],1995:201-212.
  • 10CAI Q, AGGARWAL J K. Tracking human motion in structured envirsonments using a distributed-camera system [ J]. IEEE Trnnsactions on Pattern Analysis and Machine Intelligence, 1999, 21(11 ) :1241-1247.

共引文献5

同被引文献14

引证文献2

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