基于压缩传感的多样本目标跟踪算法

doi:10.3969/j.issn.1006-2475.2014.07.013

计算机与现代化 ›› 2014, Vol. 0 ›› Issue (7): 58-62+67.doi: 10.3969/j.issn.1006-2475.2014.07.013

基于压缩传感的多样本目标跟踪算法

（江南大学数字媒体学院，江苏无锡 214122）

收稿日期:2014-04-23 出版日期:2014-07-16 发布日期:2014-07-17
作者简介:陈茜(1988-)，女，浙江嘉善人，江南大学数字媒体学院硕士研究生，研究方向：计算机视觉; 狄岚(1965-)，女，副教授，硕士，研究方向：数字图像处理，计算机视觉。
基金资助:
江苏省六大人才高峰项目(DZXX-028); 江南大学教师卓越工程项目(JGC2013145)

Multiple Instance Target Tracking Algorithm Based on Compressed Sense

(School of Digital Media, Jiangnan University, Wuxi 214122, China)

Received:2014-04-23 Online:2014-07-16 Published:2014-07-17

摘要/Abstract

摘要： 针对实时目标跟踪会产生跟踪不稳定、易漂移、被遮挡就丢失的问题，提出改进的多样本跟踪算法。在压缩传感实时跟踪中，通过增加随机测量矩阵产生新的压缩感知特征，融合多个正负样本。结合boosting学习方法更新特征权值并改进置信图估计，解决目标漂移和丢失问题。实验结果表明，该方法在目标运动、纹理和环境显著变化以及被部分遮挡的情况下，跟踪的鲁棒性依旧很高，能达到稳定、实时的目标跟踪。

关键词: 目标跟踪, 压缩传感, 多样本, 实时跟踪, 漂移, 目标遮挡

Abstract: Aiming at the problems of unstable tracking, easy to drift, obscured loss, which are produced in real-time target tracking, we propose an improved tracking algorithm for multiple instances. In the compressed sensing and real-time tracking, by adding random measurement matrix to produce new features, multiple positive and negative instances are integrated. By combining with the boosting learning method to update the feature weights and improve the confidence map estimation, we solve the problems of target drift and loss. Experimental results show that the proposed algorithm achieves better robustness and stable real-time tracking when the target moves quickly, or in conditions that the textures and lightings change seriously, as well as it is partially covered.

Key words: target tracking, compressive sensing, multiple instance, real-time tracking, drift, target occlusion

中图分类号:

TP391

陈茜，狄岚. 基于压缩传感的多样本目标跟踪算法[J]. 计算机与现代化, 2014, 0(7): 58-62+67.

CHEN Xi, DI Lan. Multiple Instance Target Tracking Algorithm Based on Compressed Sense[J]. Computer and Modernization, 2014, 0(7): 58-62+67.

参考文献

[1] 马俊. 基于图像和深度信息融合的目标跟踪[J]. 计算机与现代化, 2013(9):91-94.

[2] 徐培智,徐贵力,王彪,等. 基于立体视觉的非合作目标位姿测量[J]. 计算机与现代化, 2013(8):85-91.

[3] Dee H, Velastin S. How close are we to solving the problem of automated visual surveillance?[J]. Machine Vision and Applications, 2008,19(5-6):329-343.

[4] 刘国营,陈秀宏,庄甘霖. 检测识别跟踪分离的在线多样本视频目标跟踪[J]. 计算机工程, 2012,38(23):194-197.

[5] Grabner H, Bischof H. On-line boosting and vision[C]// Proceedings of the 2006 IEEE Conference on Computer Vision and Pattern Recognition. 2006,1:260-267.

[6] Stalder S, Grabner H, Van Gool L. Beyond semi-supervised tracking: Tracking should be as simple as detection, but not simpler than recognition[C]// Proceedings of the 12th IEEE International Conference on Computer Vision Workshops. 2009:1409-1416.

[7] Li Hanxi, Shen Chunhua, Shi Qinfeng. Real-time visual tracking using compressive sensing[C]// Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition. 2011:1305-1312.

[8] Mei Xue, Ling Haibin. Robust visual tracking and vehicle classification via sparse representation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011,33(11):2259-2272.

[9] Donoho D L. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006,52(4):1289-1306.

[10] Zhang Kaihua, Zhang Lei, Yang Ming-Hsuan. Real-time compressive tracking[C]// Proceedings of the 13th European Conference on Computer Vision. 2012:866-879.

[11] 朱秋平,颜佳,张虎,等. 基于压缩感知的多特征实时跟踪[J]. 光学精密工程, 2013,21(2):437-443.

[12] Viola P, Jones M. Rapid object detection using a boosted cascade of simple features[C]// Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2001,1:511-518.

[13] Babenko B, Yang Ming-Hsuan, Belongie S. Visual tracking with online multiple instance learning[C]// Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. 2009:983-990.

[14] 顾幸方,茅耀斌,李秋洁. 基于Mean Shift的视觉目标跟踪算法综述[J]. 计算机科学, 2012,39(12):16-24.

[15] Patras I, Hancock E R. Coupled prediction classification for robust visual tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010,32(9):1553-1567.

[16] Olson C F. Maximum-likelihood template matching[C]// Proceedings of the 2000 IEEE Conference on Computer Vision and Pattern Recognition. 2000,2:52-57.

[17] Glocker B, Heibel T H, Navab N, et al. TriangleFlow: Optical flow with triangulation-based higher-order likelihoods[C]// Proceedings of the 11th European Conference on Computer Vision. 2010:272-285.

[18] Lin Zhe, Hua Gang, Davis L S. Multiple instance feature for robust part-based object detection[C]// Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. 2009:405-412.

[19] Fang Jiang-xiong, Yang Jie, Liu Hua-xiang. Efficient and robust fragments-based multiple kernels tracking[J]. International Journal of Electronics and Communications, 2011,65(11):915-923.

基于压缩传感的多样本目标跟踪算法

Multiple Instance Target Tracking Algorithm Based on Compressed Sense

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 9

编辑推荐

Metrics

本文评价

[1]	章恒, 鞠时光. 基于概念漂移检测的网络数据流分类[J]. 计算机与现代化, 2021, 0(07): 107-114.
[2]	郑沫利1,赵艳轲1,闫敏2,孙俊2,刘雍容1. 基于RDPSO-SVM的粮食产后储藏环节损耗智能评估方法[J]. 计算机与现代化, 2020, 0(03): 72-.
[3]	梁慈1,陈世平2. 基于包簇映射架构的包漂移策略研究[J]. 计算机与现代化, 2019, 0(11): 112-.
[4]	钟志威. 基于自适应核回归和代数重建法的低剂量CT图像重建[J]. 计算机与现代化, 2016, 0(11): 38-42.
[5]	王晶一1，高航1，沈春林2. 一种基于压缩传感的光场采集与重建方法[J]. 计算机与现代化, 2015, 0(4): 61-64,69.
[6]	郭文锋1,王勇2. 基于累积正样本的偏斜数据流集成分类方法[J]. 计算机与现代化, 2015, 0(3): 41-47.
[7]	朱学慧;李享元. 基于DSP的心电信号实时滤波算法的实现[J]. 计算机与现代化, 2012, 198(2): 35-37.
[8]	刘洪江李文英. 基于传感压缩的量子层析成像技术的研究[J]. 计算机与现代化, 2011, 12(12): 121-125.
[9]	谢琰嵩;朱晓莲. 数据流分类技术分析[J]. 计算机与现代化, 2010, 1(01): 32-36.