[1]杨欣,沈雷,费树岷,等.基于证据理论和多核函数融合的目标跟踪[J].东南大学学报(自然科学版),2015,45(5):861-864.[doi:10.3969/j.issn.1001-0505.2015.05.009]
 Yang Xin,Shen Lei,Fei Shumin,et al.Target tracking method based on evidence theory and multiple kernel function[J].Journal of Southeast University (Natural Science Edition),2015,45(5):861-864.[doi:10.3969/j.issn.1001-0505.2015.05.009]
点击复制

基于证据理论和多核函数融合的目标跟踪()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
45
期数:
2015年第5期
页码:
861-864
栏目:
计算机科学与工程
出版日期:
2015-09-20

文章信息/Info

Title:
Target tracking method based on evidence theory and multiple kernel function
作者:
杨欣1沈雷1费树岷2周大可1
1南京航空航天大学自动化学院, 南京210016; 2东南大学自动化学院, 南京210096
Author(s):
Yang Xin1 Shen Lei1 Fei Shumin2 Zhou Dake1
1College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
2School of Automation, Southeast University, Nanjing 210096, China
关键词:
目标遮挡 目标跟踪 证据理论 均值漂移
Keywords:
target occlusion target tracking evidence theory Mean-Shift
分类号:
TP391.41
DOI:
10.3969/j.issn.1001-0505.2015.05.009
摘要:
针对目标长时间大面积被物体遮挡所导致的跟踪位置偏差或丢失,提出一种基于证据理论和多核函数融合的目标跟踪方法.首先,采用Mean-Shift方法分别计算边缘方向直方图模型和颜色直方图模型;其次,使用特征融合的方法将2种特征模型进行融合;再采用多核函数计算出多个相对应的目标位置估计;运用基于证据理论的方法对已获得的不同目标位置估计进行判断取舍和计算,计算出最理想的目标区域作为最终的目标位置.经过这样的处理和计算可以保证在遮挡的情况下能够实时准确地跟踪上原始目标.实验结果表明,所提算法在处理目标大面积遮挡、光照变化等问题时具有更好的性能和鲁棒性.
Abstract:
When an object is occluded for long the tracking position deviation or loss may occur. To solve this problem a target tracking method based on evidence theory and multiple kernel function is proposed. First, edge direction histogram model and color histogram model are calculated by using Mean-Shift method, and then the two feature models are fused together by the feature fusion method. Secondly, multiple kernel function is adopted to obtain several estimated positions of target. Thirdly, according to evidence theory, judgment and calculation are implemented to get the final target location through the obtained multiple target position estimates. By these processing and calculation, it can be ensured that the original target can be tracked in real time and accurately in the case of occlusion. Experimental results show that the proposed algorithm has better performance and robustness for resolving the problem of occlusion and illumination change during target tracking.

参考文献/References:

[1] 杨欣,刘加,周鹏宇,等.基于多特征融合的粒子滤波自适应目标跟踪算法 [J]. 吉林大学学报:工学版, 2015,45(2): 533-539.
  Yang Xin, Liu Jia, Zhou Pengyu, et al. Adaptive particle filter for object tracking based on fusing multiple features[J]. Journal of Jilin University:Engineering and Technology Edition, 2015,45(2): 533-539.(in Chinese)
[2] Comaniciu D, Ramesh V, Meer P. Kernel-based object tracking [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(5): 564-577.
[3] 杨欣,费树岷,李刚,等.基于复杂特征融合的改进Mean Shift目标跟踪 [J].控制与决策,2014, 29(7):1297-1300.
  Yang Xin, Fei Shumin,Li Gang, et al. Improved Mean Shift tracking algorithm based on complicated feature fusion [J]. Control and Decision, 2014, 29(7):1297-1300.(in Chinese)
[4] 何志勇,蔡乐才,许继家.基于 Mean Shift 算法跟踪视频中运动目标 [J].郑州大学学报:理学版,2010,42(1): 38-40.
  He Zhiyong, Cai Lecai, Xu Jija. Tracking moving objects based on Mean Shift algorithm in video [J]. Journal of Zhengzhou University:Natural Science Edition, 2010,42(1): 38-40.(in Chinese)
[5] 易建政,汪金军,张俊坤,等. DS 证据理论在信息融合中的应用研究 [J].国外电子测量技术,2010(12):31-34.
  Yi Jianzheng, Wang Jinjun, Zhang Junkun, et al. Application research on D-S evidence theory in data fusion [J]. Foreign Electronic Measurement Technology, 2010(12):31-34.(in Chinese)
[6] Comaniciu D, Meer P. Mean shift: a robust approach toward feature space analysis [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(5): 603-619.
[7] 刘惟锦,章毓晋.基于 Kalman 滤波和边缘直方图的实时目标跟踪 [J].清华大学学报:自然科学版,2008,48(7):1104-1107.
  Liu Weijin, Zhang Yujin. Edge-color-histogram and Kalman filter-based real-time object tracking[J]. Journal of Tsinghua University:Science and Technology, 2008,48(7):1104-1107.(in Chinese)
[8] Pylkko H, Riekki J, Roning J. Real-time color-based tracking via a marker interface[C]//IEEE International Conference on Robotics and Automation. Seoul, Korea, 2001, 2: 1214-1219.
[9] Gao L P, Jia J, Dong L B. An improved fusion algorithm of evidence theory [C]//Conference on Cross Strait Quad-Regional Radio Science and Wireless Technology. Harbin, China, 2011,2:1495-1498.
[10] Li Y B, Wang N, Zhou C. Based on D-S evidence theory of information fusion improved method [C]//International Conference on Computer Application and System Modeling. Taiyuan, China,2010, 1:416-419.

相似文献/References:

[1]葛云,章东.活体细胞图像斑点的自动提取和跟踪方法[J].东南大学学报(自然科学版),2009,39(3):464.[doi:10.3969/j.issn.1001-0505.2009.03.009]
 Ge Yun,Zhang Dong.Automated methods for particle tracking of the intracellular transport of clathrin-coated pits and vesicles[J].Journal of Southeast University (Natural Science Edition),2009,39(5):464.[doi:10.3969/j.issn.1001-0505.2009.03.009]
[2]王立功,于甬华,姜晓彤,等.基于Snake模型的图像目标轮廓自动跟踪方法[J].东南大学学报(自然科学版),2003,33(2):215.[doi:10.3969/j.issn.1001-0505.2003.02.025]
 Wang Ligong,Yu Yonghua,Jiang Xiaotong,et al.Snake model based automatic tracking method for image object contour[J].Journal of Southeast University (Natural Science Edition),2003,33(5):215.[doi:10.3969/j.issn.1001-0505.2003.02.025]
[3]程旭,郭海燕,李拟珺,等.一种基于超像素的局部判别式跟踪算法[J].东南大学学报(自然科学版),2014,44(6):1105.[doi:10.3969/j.issn.1001-0505.2014.06.002]
 Cheng Xu,Guo Haiyan,Li Nijun,et al.Local discriminative tracking algorithm based on superpixel[J].Journal of Southeast University (Natural Science Edition),2014,44(5):1105.[doi:10.3969/j.issn.1001-0505.2014.06.002]
[4]程旭,周琳,张毅锋.基于多损失的生成式对抗目标跟踪算法[J].东南大学学报(自然科学版),2018,48(3):400.[doi:10.3969/j.issn.1001-0505.2018.03.004]
 Cheng Xu,Zhou Lin,Zhang Yifeng.Object tracking algorithm based on multiple loss generative adversary[J].Journal of Southeast University (Natural Science Edition),2018,48(5):400.[doi:10.3969/j.issn.1001-0505.2018.03.004]

备注/Memo

备注/Memo:
收稿日期: 2014-11-20.
作者简介: 杨欣(1978—),男,博士,副教授,yangxin@nuaa.edu.cn.
基金项目: 国家自然科学基金资助项目(61573182,61305011)、光电控制技术重点实验室和航空科学基金联合资助项目(20145152027)、南京航空航天大学青年科技创新基金资助项目(NS2014035)、南京航空航天大学研究生创新基地(实验室)开放基金资助项目(kfjj201426).
引用本文: 杨欣,沈雷,费树岷,等.基于证据理论和多核函数融合的目标跟踪[J].东南大学学报:自然科学版,2015,45(5):861-864. [doi:10.3969/j.issn.1001-0505.2015.05.009]
更新日期/Last Update: 2015-09-20