基于颜色纹理特征的均值漂移目标跟踪算法

被引:19
作者
戴渊明
韦巍
林亦宁
机构
[1] 浙江大学电气工程学院
基金
中央高校基本科研业务费专项资金资助;
关键词
目标跟踪; 均值漂移; 联合相似度; 颜色特征; 纹理特征;
D O I
暂无
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
针对经典均值漂移跟踪算法采用单一的颜色特征对目标进行跟踪检测存在的不足,提出一种将纹理特征与颜色特征相结合的改进均值漂移目标跟踪算法.该算法首次提出特征联合相似度的概念,通过均值漂移算法联合相似度的最大化计算,正确快速地获取新一帧图像跟踪目标的位置.实验结果表明,该算法具有更高的可靠性,同时满足一般目标跟踪任务的实时性要求.
引用
收藏
页码:212 / 217
页数:6
相关论文
共 13 条
[1]  
A nonparametric adaptive trackingalgorithm based on multiple feature distributions. Ediz Polat,Mustafa Ozden. IEEETransactions on Multimedia .
[2]  
Kernel-based object tracking. Comaniciu D,Ramesh V,Meer P. IEEE Transactions on Pattern Analysis and Machine Intelligence . 2003
[3]  
Texture features forbrowsing andretrieval of large imagedata. MANJUNATH B S,MA W Y. IEEETransactions on Pattern Analysis and Machine Intelli-gence . 1996
[4]  
An EM-like algorithm forcolor-histogram-based object tracking. ZIVKOVIC Z,KROSE B. Proceedingsof IEEE Conference on Computer Vision and Pattern Rec-ognition . 2004
[5]  
A filter designtechnique for steerable pyramid image transforms. KARASARIDIS A,SIMONCELLI E. 1996IEEE International Conference on Acoustics,Speech,and Signal Processing,ICASSP-96 . 2002
[6]  
Real-time Tracking Using Trust-region Methods. Liu T.L,Chen H.T. IEEE Transactions on Pattern Analysis and Machine Intelligence . 2004
[7]  
Integrating shape and color features for adaptive real-time object tracking. Junqiu Wang,Yagi Yasushi. IEEE International Confon Robotics and Biomimetics . 2006
[8]  
Adaptive color space switching for face tracking in multi-colored lighting environments. Stern, H.,Efros, B. Proceedings of the 5th IEEE International Conference on Automatic Face and Gesture Recognition . 2002
[9]  
Real-time visual tracking of complex structures. T. Drummond,R. Cipolla. IEEE Transactions on Pattern Analysis and Machine Intelligence . 2002
[10]  
Online selection of discriminative tracking features. Collins RT,Liu YX,Leordeanu M. IEEE Transactions on Pattern Analysis and Machine Intelligence . 2005