四帧间差分与光流法结合的目标检测及追踪

刘鑫, 金晅宏. 四帧间差分与光流法结合的目标检测及追踪[J]. 光电工程, 2018, 45(8): 170665. doi: 10.12086/oee.2018.170665
引用本文: 刘鑫, 金晅宏. 四帧间差分与光流法结合的目标检测及追踪[J]. 光电工程, 2018, 45(8): 170665. doi: 10.12086/oee.2018.170665
Liu Xin, Jin Xuanhong. Algorithm for object detection and tracking combined on four inter-frame difference and optical flow methods[J]. Opto-Electronic Engineering, 2018, 45(8): 170665. doi: 10.12086/oee.2018.170665
Citation: Liu Xin, Jin Xuanhong. Algorithm for object detection and tracking combined on four inter-frame difference and optical flow methods[J]. Opto-Electronic Engineering, 2018, 45(8): 170665. doi: 10.12086/oee.2018.170665

四帧间差分与光流法结合的目标检测及追踪

详细信息
    作者简介:
    通讯作者: 金晅宏(1978-),女,硕士,副教授,主要从事信号的获取与处理、在线检测技术和图像处理技术等研究。E-mail:judithking@vip.sina.com
  • 中图分类号: O436.3;TP391.41

Algorithm for object detection and tracking combined on four inter-frame difference and optical flow methods

More Information
  • 为了能在复杂环境下快速、准确地对多个目标进行检测及追踪,本文结合四帧间差分算法与光流算法,提出了一种更高效的运动目标检测算法。本算法为了提升光流法的处理速度并降低光照等环境所带来的影响,首先对视频序列进行四帧间差分处理,然后将得到的差分视频序列进行光流处理,以实现对视频中目标的准确检测。最后将该算法与粒子滤波、ViBe等算法进行比较,并在不同场景下对不同运动目标、不同个数目标进行捕获处理,结果表明,本方法不仅具有较好的鲁棒性,而且能够更快速、准确的对目标进行检测与追踪。

  • Overview: Tracking the target in video under different conditions is not only the basis of video analysis, but also one of the key research topics in machine vision. Some emergency information can be gotten effectively by analyzing the video. At present, the intelligent processing of video monitoring information has received more and more attention in the fields of computer vision, pattern recognition and machine learning. Some traditional algorithms are introduced into this field for target detection and tracking, such as particle filter algorithm and ViBe algorithm. These algorithms are mature and effective, but these algorithms are less traceable under complex conditions. Other methods, such as Target tracking algorithm based on kernel function, the CodeBook based on clustering algorithm have great performance under complex conditions, but these algorithms' calculations are complex and have difficulties in realizing. Inter-frame differential method is a common algorithm for tracking the moving targets with the advantage of high execution speed, but its disadvantage is that it is easily affected by the environment. Optical flow method is another important way in image processing. It can overcome the shortage of the inter-frame differential method, but the optical flow algorithm is more time-consuming. To solve the problem of multiple targets' detection and tracking under the complex environment, an improved moving objects detection method is proposed based on inter-frame differential method(four inter-frame differential method) and optical flow algorithm(pyramid LK optical flow). In this paper, by discussing the advantages and disadvantages of four inter-frame differential method and optical flow algorithm, a improved combination method is put forward. Firstly, four inter-frame difference method is used to process the of video sequences. The computed area is reduced and therefore the detection speed is improved. Then objects in the video is detected accurately by the optical flow algorithm used on light streaming video sequences. This improved method enhances the processing speed of optical flow method and reduces the effects of environment's illumination. Finally, the paper compares the proposed algorithm with four-frame difference method, optical flow algorithm, particle filter, ViBe algorithm under different scenarios with different moving targets and individual number. Experimental results show that this improved method is proved not only with good robustness, but also can work more quickly and accurately on the target detection and tracking.

  • 加载中
  • 图 1  改进算法流程图

    Figure 1.  Improved algorithm flow chart

    图 2  静态背景下多目标处理效果图。(a)原图;(b)四帧间差分法;(c) ViBe算法;(d)光流法;(e)粒子滤波算法;(f)本文算法

    Figure 2.  Multiple targets processing the renderings in static background. (a) The original image; (b) Four-frame difference method; (c) ViBe algorithm; (d) Optical flow algorithm; (e) Particle filter algorithm; (f) The algorithm in this paper

    图 3  动态背景下单个高速目标处理效果图。(a)原图;(b)四帧间差分法;(c) ViBe算法;(d)光流法;(e)粒子滤波算法;(f)本文算法

    Figure 3.  Single high speed moving target processing the renderings in dynamic background. (a) The original image; (b) Four-frame difference method; (c) ViBe algorithm; (d) Optical flow algorithm; (e) Particle filter algorithm; (f) The algorithm in this paper

    图 4  晃动摄像头下多目标处理效果图。(a)原图;(b)四帧间差分法;(c) ViBe算法;(d)光流法;(e)粒子滤波算法;(f)本文算法

    Figure 4.  Multiple targets processing the renderings under the shaking camera. (a) The original image; (b) Four-frame difference method; (c) ViBe algorithm; (d) Optical flow algorithm; (e) Particle filter algorithm; (f) The algorithm in this paper

    表 1  对静态背景下多个运动目标性能对比

    Table 1.  Performance comparison of multiple moving targets in a static background

    处理方法 平均速率/(f·s-1) 追踪目标总数量 实际追踪数量
    四帧差分法 104.17 16 9
    ViBe算法 780.00 16 7
    光流算法 23.67 16 10
    粒子滤波算法 26.67 16 11
    本算法 46.05 16 10
    下载: 导出CSV

    表 2  动态背景下单个高速运动目标性能对比

    Table 2.  Performance comparison of single high speed moving target in dynamic background

    处理方法 平均速率/(f·s-1) 追踪目标总数量 实际追踪数量
    四帧差分法 86.91 1 无法识别
    ViBe算法 780.00 1 1
    光流算法 25.00 1 1
    粒子滤波算法 29.70 1 1
    本算法 41.67 1 1
    下载: 导出CSV

    表 3  晃动摄像头下多个目标性能对比

    Table 3.  Performance comparison of multiple targets under the shaking camera

    处理方法 平均速率/(f·s-1) 追踪目标总数量 实际追踪数量
    四帧差分法 84.27 2 无法识别
    ViBe算法 757.58 2 无法识别
    光流算法 24.00 2 2
    粒子滤波算法 29.86 2 2
    本算法 41.03 2 2
    下载: 导出CSV
  • [1]

    袁国武. 智能视频监控中的运动目标检测和跟踪算法研究[D]. 昆明: 云南大学, 2012.

    Yuan G W. Research on moving objects detection and tracking methods in intelligent visual surveillance system[D]. Kunming: Yunnan University, 2012.http://cdmd.cnki.com.cn/article/cdmd-10673-1012420011.htm

    [2]

    唐迅. 基于稀疏编码的群体异常行为检测[D]. 哈尔滨: 哈尔滨工业大学, 2013.

    Tang X. Crowd abnormal behavior detection based on sparse coding[D]. Harbin: Harbin Institute of Technology, 2013.http://cdmd.cnki.com.cn/Article/CDMD-10213-1014003385.htm

    [3]

    白晓方, 杨卫, 陈佩珩.一种改进的运动目标检测与跟踪方法[J].电视技术, 2014, 38(1): 180-182. http://www.cnki.com.cn/Article/CJFDTOTAL-XXWX201303050.htm

    Bai X F, Yang W, Chen P H. Improved moving object detection and tracking method[J]. Video Engineering, 2014, 38(1): 180-182. http://www.cnki.com.cn/Article/CJFDTOTAL-XXWX201303050.htm

    [4]

    许俊波. 目标检测与跟踪方法研究[D]. 武汉: 中国地质大学, 2007.

    Xu J B. The research of detection and tracking of moving object[D]. Wuhan: China University of Geosciences, 2007.http://cdmd.cnki.com.cn/Article/CDMD-10491-2007142922.htm

    [5]

    何黄凯, 唐宁九, 李征, 等.基于LBP核密度估计的动态目标分割模型研究[J].计算机应用研究, 2012, 29(7): 2719-2721. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjyyyj201207087

    He H K, Tang N J, Li Z, et al. Research of model for dynamic object segmentation based on LBP kernel density estimation[J]. Application Research of Computers, 2012, 29(7): 2719-2721. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjyyyj201207087

    [6]

    田绪婷, 郭丹.基于Codebook的运动补偿内插方法[J].计算机工程, 2016, 42(9): 214-219. http://www.cqvip.com/QK/91657X/201601/668131467.html

    Tian X T, Guo D. Motion-compensated interpolation method based on codebook[J]. Computer Engineering, 2016, 42(9): 214-219. http://www.cqvip.com/QK/91657X/201601/668131467.html

    [7]

    张金敏, 王斌.光照快速变化条件下的运动目标检测[J].光电工程, 2016, 43(2): 14-21. doi: 10.3969/j.issn.1003-501X.2016.02.003

    Zhang J M, Wang B. Moving object detection under condition of fast illumination change[J]. Opto-Electronic Engineering, 2016, 43(2): 14-21. doi: 10.3969/j.issn.1003-501X.2016.02.003

    [8]

    袁国武, 陈志强, 龚健, 等.一种结合光流法与三帧差分法的运动目标检测算法[J].小型微型计算机系统, 2013, 34(3): 668-671. http://www.cnki.com.cn/Article/CJFDTOTAL-XXWX201303050.htm

    Yuan G W, Chen Z Q, Gong J, et al. A Moving object detection algorithm based on a combination of optical flow and three-frame difference[J]. Journal of Chinese Computer Systems, 2013, 34(3): 668-671. http://www.cnki.com.cn/Article/CJFDTOTAL-XXWX201303050.htm

    [9]

    王凯, 吴敏, 姚辉, 等.多帧背景差与Cauchy模型融合的目标检测[J].光电工程, 2016, 43(10): 12-17. doi: 10.3969/j.issn.1003-501X.2016.10.003

    Wang K, Wu M, Yao H, et al. Target detection method based on multi-frame background subtractionand cauchy model[J]. Opto-Electronic Engineering, 2016, 43(10): 12-17. doi: 10.3969/j.issn.1003-501X.2016.10.003

    [10]

    庾鹏, 王旭, 仝天乐, 等.基于GrabCut算法和四帧差分法的目标提取算法[J].微型机与应用, 2016, 35(11): 40-42. http://edu.wanfangdata.com.cn/Periodical/Detail/wxjyyy201611003

    Yu P, Wang X, Tong T L, et al. A target extraction algorithm based on GrabCut segmentation algorithm and four frame differencing[J]. Microcomputer & its Applications, 2016, 35(11): 40-42. http://edu.wanfangdata.com.cn/Periodical/Detail/wxjyyy201611003

    [11]

    Enkelmann W. Investigations of multigrid algorithms for the estimation of optical flow fields in image sequences[J]. Computer Vision, Graphics, and Image Processing, 1988, 43(2): 150-177. doi: 10.1016/0734-189X(88)90059-X

    [12]

    任克强, 余启明, 罗会兰.一种改进的混合高斯模型运动目标检测算法[J].电视技术, 2012, 36(23): 168-171. doi: 10.3969/j.issn.1002-8692.2012.23.047

    Ren K Q, Yu Q M, Luo H L. Improved algorithm of moving objects detection based on gaussian mixture model[J]. Video Engineering, 2012, 36(23): 168-171. doi: 10.3969/j.issn.1002-8692.2012.23.047

    [13]

    袁宝红. 基于视频的运动目标检测与跟踪研究[D]. 合肥: 安徽大学, 2014.

    Yuan B H. Research on video moving object detection and tracking[D]. Hefei: Anhui University, 2014.http://cdmd.cnki.com.cn/Article/CDMD-10357-1014229726.htm

    [14]

    胡社教, 葛西旺, 陈宗海.基于角点特征的KLT跟踪全景图像拼接算法[J].系统仿真学报, 2007, 19(8): 1742-1744. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=xtfzxb200708022

    Hu S J, Ge X W, Chen Z H. Based on corner feature KLT track panoramic mosaic algorithm[J]. Journal of System Simulation, 2007, 19(8): 1742-1744. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=xtfzxb200708022

    [15]

    Schmidt R A, Cathey W T. Optical representations for artificial intelligence problems[C]//O-E/lase'86 Symp, Los Angeles: SPIE, 1986: 226-233.

    [16]

    吴新根, 罗立民.一种改进的光流场计算方法[J].电子学报, 2000, 28(1): 130-131. https://www.cnki.com.cn/lunwen-1016792965.html

    Wu X G, Luo L M. An improved method of optical flow estimation[J]. Acta Electronica Sinica, 2000, 28(1): 130-131. https://www.cnki.com.cn/lunwen-1016792965.html

    [17]

    刘丹. 视频运动目标检测与跟踪算法的GPU并行优化[D]. 长沙: 国防科学技术大学, 2013.

    Liu D. parallel optimization for video moving object detection and tracking algorithm based on GPU[D]. Changsha: National University of Defense Technology, 2013.

    [18]

    蔡瑞初, 谢伟浩, 郝志峰, 等.基于多尺度时间递归神经网络的人群异常检测[J].软件学报, 2015, 26(11): 2884-2896. http://www.cnki.com.cn/Article/CJFDTotal-RJXB201511012.htm

    Cai R C, Xie W H, Hao Z F, et al. Abnormal crowd detection based on multi-scale recurrent neural network[J]. Journal of Software, 2015, 26(11): 2884-2896. http://www.cnki.com.cn/Article/CJFDTotal-RJXB201511012.htm

  • 加载中

(4)

(3)

计量
  • 文章访问数:  10721
  • PDF下载数:  3153
  • 施引文献:  0
出版历程
收稿日期:  2017-12-05
修回日期:  2018-04-11
刊出日期:  2018-08-01

目录

/

返回文章
返回