多分辨率LK光流联合SURF的跟踪方法

doi:10.11772/j.issn.1001-9081.2017.03.806

计算机应用 ›› 2017, Vol. 37 ›› Issue (3): 806-810.DOI: 10.11772/j.issn.1001-9081.2017.03.806

• 计算机视觉与虚拟现实 • 上一篇下一篇

多分辨率LK光流联合SURF的跟踪方法

厉丹^1,2, 鲍蓉², 孙金萍², 肖理庆², 党向盈²

1. 徐州工程学院江苏省大型工程装备检测与控制重点建设实验室, 江苏徐州 221000;
2. 徐州工程学院信电工程学院, 江苏徐州 221000

收稿日期:2016-08-11 修回日期:2016-10-24 出版日期:2017-03-10 发布日期:2017-03-22
通讯作者: 厉丹
作者简介:厉丹(1981-),女,江苏徐州人,副教授,博士,主要研究方向:图像处理、视频监控;鲍蓉(1968-),女,上海南汇人,教授,博士,主要研究方向:图像处理、网络工程;孙金萍(1980-),女,山东德州人,副教授,硕士,主要研究方向:智能计算;肖理庆(1981-),男,山东青岛人,讲师,博士,主要研究方向:机器视觉、分布式计算;党向盈(1978-),女,江苏徐州人,副教授,硕士,主要研究方向:图像处理。
基金资助:
江苏省高校自然科学研究面上项目（15KJB520033，16KJB510022）；江苏省产学研联合创新资金资助项目（BY2013020）；住房城乡建设部科学技术计划项目（2015-K5-027，2016-R2-060）；江苏省大型工程装备检测与控制重点建设实验室开放课题（JSKLEDC201224）。

Tracking method of multi-resolution LK optical flow combined with SURF

LI Dan^1,2, BAO Rong², SUN Jinping², XIAO Liqing², DANG Xiangying²

1. Jiangsu Key Laboratory of Large Engineering Equipment Detection and Control, Xuzhou Institute of Technology, Xuzhou Jiangsu 221000, China;
2. Information and Electrical Engineering College, Xuzhou Institute of Technology, Xuzhou Jiangsu 221008, China

Received:2016-08-11 Revised:2016-10-24 Online:2017-03-10 Published:2017-03-22
Supported by:
This work is partially supported by the Natural Science Foundation for Colleges and Universities of Jiangsu Province (15KJB520033, 16KJB510022), the Joint Innovation Fund Program of Jiangsu Province (BY2013020), the Ministry of Housing and Urban Construction Science and Technology Plan Project (2014-K5-027,2016-R2-060), the Open Project of Jiangsu Key Laboratory of Large Engineering Equipment Detection and Control (JSKLEDC201224).

摘要/Abstract

摘要： 针对交通监控中运动目标形变、雾霾天气、高速、光照不均、部分遮挡等复杂情况导致Lucas-Kanade（LK）算法跟踪不稳定问题，提出基于多分辨率LK光流算法联合快速鲁棒性特征（SURF）的跟踪算法。所提算法构建图像多分辨率小波金字塔，解决传统LK算法中同一像素点帧间大尺度运动易丢失问题；同时联合SURF尺度不变特征变换算法，提取特征点进行光流跟踪，并制定自适应模板实时更新策略；在减少光流计算量的同时增强运动目标抗复杂环境的能力。实验结果表明，新方法中特征点匹配准确快速，自适应性强，在交通复杂化境中跟踪稳定。

关键词: 光流算法, 特征提取, 快速鲁棒性特征, 多分辨率, 目标跟踪

Abstract: Aiming at the problem of tracking instability of the Lucas-Kanade (LK) algorithm for the complex situation of moving target deformation, fog and haze, high-speed, uneven illumination and partial occlusion in traffic monitoring, a tracking algorithm based on multi-resolution LK optical flow algorithm and Speed Up Robust Features (SURF) was proposed. The problem tracking failure for large-scale motion between frames of same pixel point in the traditional LK algorithm was solved by the proposed method, and the SURF scale invariant feature transformation algorithm was combined, feature points for optical flow tracking were extracted, and an adaptive template real-time update strategy was developed; the amount of optical flow calculation was reduced while enhancing the resistance ability of moving targets against complex environments. The experimental results show that the feature points matching of the new method is accurate and fast, which has strong adaptability and it is stable in the complicated traffic environment.

Key words: optical flow algorithm, feature extraction, Speed Up Robust Features (SURF), multi-resolution, target tracking

中图分类号:

TP751

厉丹, 鲍蓉, 孙金萍, 肖理庆, 党向盈. 多分辨率LK光流联合SURF的跟踪方法[J]. 计算机应用, 2017, 37(3): 806-810.

LI Dan, BAO Rong, SUN Jinping, XIAO Liqing, DANG Xiangying. Tracking method of multi-resolution LK optical flow combined with SURF[J]. Journal of Computer Applications, 2017, 37(3): 806-810.

参考文献

[1] 王为,姚明海.基于计算机视觉的智能交通监控系统[J].浙江工业大学学报,2010,38(5):574-579.(WANG W,YAO M H. Intelligent transportation monitoring system based on computer vision[J]. Journal of Zhejiang University of Technology, 2010, 38(5):574-579.)
[2] 李乐平,高杨.一种高准确率的交通监控视频车辆检测算法[J].计算机测量与控制,2015,23(3):852-854.(LI L P, GAO Y. A high accuracy and low compute cost approach in traffic video vehicles classification[J]. Computer Measurement and Control, 2015, 23(3):852-854.)
[3] 石祥滨,王萌,张德园,等.种持续光流跟踪的运动目标检测方法[J].小型微型计算机系统,2014,35(3):642-647.(SHI X B, WANG M, ZHANG D Y, et al. An approach for moving object detection using continuing tracking optical flow[J]. Journal of Chinese Computer Systems, 2014, 35(3):642-647.)
[4] HORN B K P, SCHUNCK B G. Determining optical flow[J].Artificial Intelligence, 1981, 17(1/2/3):185-203.
[5] 宋爽,杨健,王涌天.全局光流场估计技术及展望[J].计算机辅助设计与图形学学报,2014,26(5):841-850.(SONG S, YANG J, WANG Y T. Technology and prospect of global optical flow[J]. Journal of Computer-Aided Design and Computer Graphics, 2014, 26(5):841-850.)
[6] LUCAS B D, KANADE T. An iterative image registration technique with an application to stereo vision[J]. Proceedings of the 7th International Joint Conference on Artificial Intelligence. San Francisco, CA:Morgan Kaufmann Publishers, 1981,2:674-679.
[7] NAGEL H H, ENKELMANN W. An investigation of smoothness constraints for the estimation of displacement vector fields from image sequences[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, 8(5):565-593.
[8] TAO M, BAI J, KOHLI P, et al. SimpleFlow:a non-iterative, sublinear optical flow algorithm[J]. Computer Graphics Forum, 2012, 31(2pt1):345-353.
[9] PARVATHY R, THILAKAN S, JOY M, et al. Anomaly detection using motion patterns computed from optical flow[C]//Proceedings of the 2013 Third International Conference on Advances in Computing and Communications. Washington, DC:IEEE Computer Society, 2013:58-61.
[10] 陈添丁,胡鉴,吴涤.稀疏光流快速计算的动态目标检测与跟踪[J].中国图象图形学报,2013,18(12):1593-1600.(CHEN T D, HU J, WU D. Dynamic target detection and tracking based on fast computation using sparse optical flow[J]. Journal of Image and Graphics, 2013, 18(12):1593-1600.)
[11] 郑驰,项志宇,刘济林.融合光流与特征点匹配的单目视觉里程计[J].浙江大学学报(工学版),2014,48(2):279-284.(ZHENG C, XIANG Z Y, LIU J L. Monocular vision odometry based on the fusion of optical flow and feature points matching[J]. Journal of Zhejiang University (Engineering Science), 2014, 48(2):279-284.)
[12] 张志禹,刘耀阳.基于SURF与光流法的增强现实跟踪注册[J].计算机工程与应用,2015,51(3):166-170.(ZHANG Z Y, LIU Y Y. Tracking and AR registration based on SURF and optical flow method[J]. Computer Engineering and Applications, 2015, 51(3):166-170.)
[13] BAY H, TUYTELAARS, VAN GOOL L. SURF:speeded up robust features[C]//Computer Vision-ECCV 2006, LNCS 3951. Berlin:Springer, 2006:404-417.

[1]	姬张建, 任兴旺. 带旋转与尺度估计的全卷积孪生网络目标跟踪算法[J]. 计算机应用, 2021, 41(9): 2705-2711.
[2]	郑志强, 胡鑫, 翁智, 王雨禾, 程曦. 基于改进DenseNet的牛眼图像特征提取方法[J]. 计算机应用, 2021, 41(9): 2780-2784.
[3]	管其杰, 张挺, 李德亚, 周绍景, 杜奕. 基于多分辨率生成对抗网络的空间数据不确定性重建方法[J]. 计算机应用, 2021, 41(8): 2306-2311.
[4]	钟莎, 黄玉清. 基于孪生区域候选网络的无人机指定目标跟踪[J]. 计算机应用, 2021, 41(2): 523-529.
[5]	佘玉龙, 张晓龙, 程若勤, 邓春华. 基于边缘关注模型的语义分割方法[J]. 计算机应用, 2021, 41(2): 343-349.
[6]	赵津, 宋文爱, 邰隽, 杨吉江, 王青, 李晓丹, 雷毅, 邱悦. 儿童阻塞性睡眠呼吸暂停计算机人脸辅助诊断综述[J]. 《计算机应用》唯一官方网站, 2021, 41(11): 3394-3401.
[7]	朱新成, 何坤金, 倪娜, 郝博. 基于改进迭代最近点算法的接骨板贴合性快捷计算方法[J]. 计算机应用, 2021, 41(10): 3033-3039.
[8]	尹春勇, 何苗. 基于改进胶囊网络的文本分类[J]. 计算机应用, 2020, 40(9): 2525-2530.
[9]	周云, 陈淑荣. 基于双流非局部残差网络的行为识别方法[J]. 计算机应用, 2020, 40(8): 2236-2240.
[10]	张杰, 常天庆, 戴文君, 郭理彬, 张雷. 基于相关滤波与颜色概率模型的目标跟踪算法[J]. 计算机应用, 2020, 40(6): 1774-1782.
[11]	张家岗, 李达平, 杨晓东, 邹茂扬, 吴锡, 胡金蓉. 基于深度卷积特征光流的形变医学图像配准算法[J]. 计算机应用, 2020, 40(6): 1799-1805.
[12]	徐代, 岳璋, 杨文霞, 任潇. 基于改进的三向流Faster R-CNN的篡改图像识别[J]. 计算机应用, 2020, 40(5): 1315-1321.
[13]	刘尚旺, 刘承伟, 张爱丽. 基于深度可分卷积神经网络的实时人脸表情和性别分类[J]. 计算机应用, 2020, 40(4): 990-995.
[14]	沈亮, 王鑫, 陈曙晖. 面向移动应用识别的结构化特征提取方法[J]. 计算机应用, 2020, 40(4): 1109-1114.
[15]	郭志强, 胡永武, 刘鹏, 杨杰. 基于特征融合的室外天气图像分类[J]. 计算机应用, 2020, 40(4): 1023-1029.

多分辨率LK光流联合SURF的跟踪方法

Tracking method of multi-resolution LK optical flow combined with SURF

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics