首页 > 论文 > 激光与光电子学进展 > 55卷 > 6期(pp:61503--1)

结合BRISK与区域预估的改进长时跟踪算法

Improved Long Time Tracking Algorithm by Combining BRISK and Region Estimation

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

鉴于传统的跟踪学习检测(TLD)算法存在稳健性差、跟踪成功率低以及运算效率低等问题, 提出一种结合二进制稳健不变可扩展关键点(BRISK)特征点与区域预估的TLD跟踪算法。在跟踪器中引入BRISK特征点, 将其与传统的用于跟踪的普通像素点相结合, 共同用于目标跟踪, 由于BRISK特征点提取较快, 从而使得跟踪器部分的总体运算时间降低;在检测器部分采用了卡尔曼滤波器与马尔可夫模型方向预测器相结合的方式, 该方式使得最终送入到检测器的子图像块数量大幅缩减, 且对相似目标的辨别能力增强, 进而提升了检测器的速度和精度。实验结果表明, 相比于传统TLD算法, 所提TLD算法的跟踪精度提高约64.4%, 运行速度提升约39.6%, 并具有更好的稳健性。

Abstract

In view of the fact that the traditional tracking learning detection (TLD) algorithm has poor robustness, low tracking success rate and low computing efficiency, a TLD tracking algorithm combining binary robust invariant scalable keypoints (BRISK) feature points and region prediction is proposed. In the tracker, the BRISK feature point is combined with the conventional pixel points used for tracking, and they are used for target tracking together. Due to the fast extraction of BRISK feature points, the total computing time of the tracker is reduced. In the detector part, the combination of Kalman filter and Markov model direction predictor greatly reduces the number of sub-image blocks sent to the detector, and enhances the identification ability for similar targets, thereby improving the speed and accuracy of the detector. The experimental results show that, compared with the traditional TLD algorithm, the tracking accuracy of the proposed TLD algorithm is improved by about 64.4%, and the running speed is increased by about 39.6%, and its robustness is better.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TP391.4

DOI:10.3788/lop55.061503

所属栏目:机器视觉

基金项目:黔科合LH字[2014]7630号、国家国际科技合作专项(2014DFA00670)、黔科合外G字[2015]7002号

收稿日期:2017-12-01

修改稿日期:2018-01-05

网络出版日期:--

作者单位    点击查看

康海林:贵州大学大数据与信息工程学院, 贵州 贵阳 550025
赵婷:贵州大学大数据与信息工程学院, 贵州 贵阳 550025
周骅:贵州大学大数据与信息工程学院, 贵州 贵阳 550025
刘桥:贵州大学大数据与信息工程学院, 贵州 贵阳 550025
张正平:贵州大学大数据与信息工程学院, 贵州 贵阳 550025

联系人作者:刘桥(liuqiao1955@163.com)

备注:康海林(1990-), 男, 硕士研究生, 主要从事视频图像处理和目标跟踪方面的研究。E-mail: hailin3288@sina.com

【1】Dong Y K, Wang C X, Xue L J, et al. Pedestrian detection and tracking based on TLD framework [J]. Journal of Huazhong University of Science and Technology (Natural Science Edition), 2013, 41(s1): 226-228.
董永坤, 王春香, 薛林继, 等. 基于TLD框架的行人检测和跟踪[J]. 华中科技大学学报(自然科学版), 2013, 41(s1): 226-228.

【2】Kalal Z, Mikolajczyk K, Matas J. Tracking-learning-detection[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2012, 34(7):1409-1422.

【3】Qin F, Wang R G, Liang Q X, et al. A Improved TLD target tracking algorithm based on key feature points[J]. Computer Engineering and Applications, 2016, 52(4): 181-187.
秦飞, 汪荣贵, 梁启香, 等. 基于关键特征点的改进TLD目标跟踪算法研究[J]. 计算机工程与应用, 2016, 52(4): 181-187.

【4】Yi S, Lin F Q, Zhou S Y. Automatic tracking method based on improved TLD[J]. Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition), 2016, 28(6): 892-896.
易诗, 林凡强, 周姝颖. 基于改进TLD的自动目标跟踪方法[J]. 重庆邮电大学学报(自然科学版), 2016, 28(6): 892-896.

【5】Jin Z, Liu C C. Accelerated TLD algorithm and its application in multiple target tracking[J]. Computer System and Applications, 2016, 25(6): 196-201.
金哲, 刘传才. 加速的TLD算法及其在多目标跟踪中的应用[J]. 计算机系统应用, 2016, 25(6): 196-201.

【6】Jiao P F, Qin P L, Miao Q G, et al. Improved TLD algorithm based on multi-innovation Kalman filter[J]. Journal of Data Acquisition and Processing, 2016, 31(3): 592-598.
焦蓬斐, 秦品乐, 苗启广, 等. 基于多新息Kalman滤波的TLD改进算法[J]. 数据采集与处理, 2016, 31(3): 592-598.

【7】Xiao Q G, Ye Q W, Zhou Y, et al. Long-time video tracking algorithm of optimized TLD based on Mean-Shife[J]. Computer Application Research, 2015, 32(3): 925-928.
肖庆国, 叶庆卫, 周宇, 等. 基于Mean-Shift优化的TLD视频长时间跟踪算法[J]. 计算机应用研究, 2015, 32(3): 925-928.

【8】Sun B J, Zhang B, song C, et al. Improved TLD object tracking algorithm based on corner reinforced[J]. Chinese Journal of Liquid Crystals and Displays, 2016, 31(9): 921-928.
孙保基, 张葆, 宋策, 等. 基于角点增强改进的TLD目标跟踪算法[J]. 液晶与显示, 2016, 31(9): 921-928.

【9】Shi H, Lin Z, Tang W, et al. A robust hand tracking approach based on modified tracking-learning-detection algorithm[J]. Lecture Notes in Electrical Engineering, 2014, 308: 9-15.

【10】Zhu X T, Shi F H. Improved tracking learning detection method based on BRISK keypoints[J]. Computer Engineering, 2017, 43(2): 268-272.
祝贤坦, 石繁槐. 基于BRISK特征点改进的跟踪学习检测方法[J]. 计算机工程, 2017, 43(2): 268-272.

【11】Meng Y, Zhang B. A multiple-object tracking algorithm using TLD-based adaptive adjustment of detection areas[J]. Journal of Northeastern University (Natural Science Edition), 2017, 38(2): 214-218.
孟煜, 张斌. 检测区域自适应调整的TLD多目标跟踪算法[J]. 东北大学学报(自然科学版), 2017, 38(2): 214-218.

【12】Lu B, Gu S H. Object tracking algorithm based on hidden Markov model and block feature matching[J]. Laser & Optoelectronics Progress, 2017, 54(9): 091006.
陆兵, 顾苏杭. 基于隐马尔可夫模型和分块特征匹配的目标跟踪算法[J]. 激光与光电子学进展, 2017, 54(9): 091006.

【13】Zhou X, Qian Q M, Ye Y Q, et al. Improved TLD visual target tracking algorithm[J]. Journal of Image and Graphics, 2013, 18(9): 1115-1123.
周鑫, 钱秋朦, 叶永强, 等. 改进后的TLD视频目标跟踪方法[J]. 中国图象图形学报, 2013, 18(9): 1115-1123.

【14】Cai R T, Zhu P. Face tracking with multi-feature based on Markov random field[J]. Laser & Optoelectronics Progress, 2017, 54(2): 021002.
蔡荣太, 朱鹏. 基于马尔科夫随机场的多特征人脸跟踪算法[J]. 激光与光电子学进展, 2017, 54(2): 021002.

【15】Huang C, Li Y, Nevatia R. Multiple target tracking by learning-based hierarchical association of detection responses[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2013, 35(4): 898-910.

【16】Kang Z, Landry S J. Using scanpaths as a learning method for a conflict detection task of multiple target tracking[J]. Pattern Recognition and Artificial Intelligence, 2014, 56(6): 1150-1162.
周治平, 周明珠, 李文慧. 基于混合粒子滤波和稀疏表示的目标跟踪算法[J]. 模式识别与人工智能, 2016, 29(1): 22-30.

【17】Zhang B, Long H. Visual target tracking algorithm based on image signature algorithm[J]. Laser & Optoelectronics Progress, 2017, 54(9): 091504.
张博, 龙慧. 基于图像签名算法的视觉目标跟踪算法[J]. 激光与光电子学进展, 2017, 54(9): 091504.

【18】Zhou Z P, Zhou M Z, Li W H. Object tracking algorithm based on hybrid particle filter and sparse representation[J]. Pattern Recognition and Artificial Intelligence, 2016, 29(1): 22-30.

【19】Hare S, Saffari A, Torr P H S. Struck:structured output tracking with kernels[C]. IEEE International Conference on Computer Vision, 2012: 263-270.

引用该论文

Kang Hailin,Zhao Ting,Zhou Hua,Liu Qiao,Zhang Zhengping. Improved Long Time Tracking Algorithm by Combining BRISK and Region Estimation[J]. Laser & Optoelectronics Progress, 2018, 55(6): 061503

康海林,赵婷,周骅,刘桥,张正平. 结合BRISK与区域预估的改进长时跟踪算法[J]. 激光与光电子学进展, 2018, 55(6): 061503

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF