基于序贯蒙特卡罗的多线索目标跟踪算法

冯桂兰; 田维坚; 黄昌清; 林盘; 张帆

光电工程, 2010, 37 (8): 5, 网络出版: 2010-09-07

基于序贯蒙特卡罗的多线索目标跟踪算法

Object Tracking Algorithm Based on Multi-cue and Sequential Monte Carlo

论文大纲

冯桂兰 ^1,*田维坚 ¹黄昌清 ¹林盘 ²张帆 ¹

作者单位

¹ 中国计量学院光学与电子科技学院，杭州 310018

² 特伦多大学脑科学中心，特伦多 38060，意大利

图像序列跟踪序贯蒙特卡罗方法多线索融合抗遮挡 image sequence tracking SMC method multi-cue fusion anti-occlusions

摘要

颜色直方图对噪声和部分遮挡不敏感，当背景颜色与目标颜色相近时，会影响跟踪效果。本文提出一种有效的基于多线索融合的序贯蒙特卡罗图像序列跟踪方法，采用颜色直方图和边缘直方图与序贯蒙特卡罗算法结合起来进行视频跟踪。颜色直方图和边缘直方图一起构建目标观测似然函数。在序贯蒙特卡罗方法的框架下，采用观测模型函数获取图像序列中目标位置的后验概率分布。实验结果表明，结合图像颜色与边缘特征，在序贯蒙特卡罗的框架下可以取得更为有效和稳健的跟踪效果。

Abstract

The color-based histogram is robust against noise and partial occlusion, but suffers from the presence of the confusing colors in the background. An efficient image sequence tracking method was presented based on multiple cues in Sequential Monte Carlo (SMC). The background-weighted color histogram was combined with edge histogram into SMC for tracking. Color histograms and edge histograms were used to model the object observations likelihoods function. The observations were used to obtain a posterior probability distribution for the location of the object in the sequence images based on SMC. It can be seen from the experiment that the combination of color histogram and edge histogram based on SMC can achieve more robustness and efficient tracking.

参考文献

[1] Comaniciu D，Ramesh V，Meer P. Real-Time Tracking of Non-Rigid Objects using Mean Shift [C]// IEEE Conference on Computer Vision and Pattern Recognition，Hilton Head，SC，USA，June 13-15，2000：142-151.

[2] . A Tutorial on Particle Filters for On-line Non-linear Non-gaussian Bayesian Tracking[J]. IEEE Transactions on Signal Processing(S1053-587X), 2002, 50(2): 174-188.

[3] . An Introduction to MCMC for Machine Learning[J]. Machine Learning (S0885-6125), 2003, 50: 5-43.

[4] . Evolutionary Optimization of Dynamics Models in Sequential Monte Carlo Target Tracking[J]. IEEE Transactions on Evolutionary Computation(S1089-778X), 2009, 13(4): 879-894.

[5] . Sequential Monte Carlo Methods for Multiple Target Tracking and Data Fusion[J]. IEEE Trans. on Signal Processing(S1053-587X), 2002, 50(2): 309-325.

[6] . Sequential Monte Carlo Methods for Tracking Multiple Targets with Deterministic and Stochastic Constraints[J]. IEEE Transactions on Signal Processing(S1053-587X), 2008, 56(3): 937-948.

[7] . An Adaptive Color-based Particle Filter[J]. Image and Vision Computing (S0262-8856), 2003, 21(1): 99-110.

[8] Perez P，Hue C，Vermaak J，et al. Color-Based Probabilistic Tracking [C]// Proceedings of the 7th European Conference on Computer Vision-Part I，Copenhagen，May 27-30，2002：661-675.

[9] . Data Fusion for Visual Tracking with Particles[J]. Proceedings of the IEEE(S0018-9219), 2004, 92(3): 495-513.

[10] Rui Y，Chen Y. Better Proposal Distributions: Object Tracking Using Unscented Particle Filter [C]// Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition，2001，2：786-793.

[11] . Visual Contour Tracking Based on Particle Filters[J]. Image and Vision Computing(S0262-8856), 2003, 21(1): 111-123.

[12] Okuma K，Taleghani A，De Freitas N. A Boosted Particle Filter：Multitarget Detection and Tracking [C]// European Conference on Computer Vision，Prague，May 11-14，2004：28-39.

[13] Shen C，Hengel A，Dick A. Probabilistic Multiple Cue Integration for Particle Filter Based Tracking [C]// DICTA，Sydney， Dec 10-12，2003：399-408.

[14] . Toward robust multi-cue integration for visual tracking[J]. Machine Vision and Applications (S1432-1769), 2001, 14: 50-58.

[15] . Incremental Learning for Robust Visual Tracking[J]. International Journal of Computer Vision (S0920-5691), 2008, 77: 125-141.

[16] . Partial Linear Gaussian Models for Tracking in Image Sequences Using Sequential Monte Carlo Methods[J]. International Journal of Computer Vision (S0920-5691), 2007, 74(1): 75-102.

[17] . Sequential Monte Carlo Tracking by Fusing Multiple Cues in Video Sequences[J]. Image and Vision Computing(S0262-8856), 2007, 28(1): 1217-1227.

[18] Islam M Z，Oh Chi-min，Lee Chil-Woo. Real Time Moving Object Tracking by Particle Filter [C]// International Symposium on Computer Science and its Applications，Hobart，Oct 13-15，2008：347-352.

[19] Birchfield S，Rangarajan S. Spatiograms versus Histograms for Region-Based Tracking [C]// Computer Vision and Pattern Recognition，San Diego，June 20-26，2005：1152-1157

冯桂兰, 田维坚, 黄昌清, 林盘, 张帆. 基于序贯蒙特卡罗的多线索目标跟踪算法[J]. 光电工程, 2010, 37(8): 5. FENG Gui-lan, TIAN Wei-jian, HUANG Chang-qing, LIN Pan, ZHANG Fan. Object Tracking Algorithm Based on Multi-cue and Sequential Monte Carlo[J]. Opto-Electronic Engineering, 2010, 37(8): 5.

基于序贯蒙特卡罗的多线索目标跟踪算法

关于本站 Cookie 的使用提示

全站搜索

基于序贯蒙特卡罗的多线索目标跟踪算法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索