高级搜索

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

动态背景下基于低秩及稀疏分解的动目标检测方法

王洪雁 张海坤

王洪雁, 张海坤. 动态背景下基于低秩及稀疏分解的动目标检测方法[J]. 电子与信息学报, 2020, 42(11): 2788-2795. doi: 10.11999/JEIT190452
引用本文: 王洪雁, 张海坤. 动态背景下基于低秩及稀疏分解的动目标检测方法[J]. 电子与信息学报, 2020, 42(11): 2788-2795. doi: 10.11999/JEIT190452
Hongyan WANG, Haikun ZHANG. Moving Object Detection Method Based on Low-Rank and Sparse Decomposition in Dynamic Background[J]. Journal of Electronics and Information Technology, 2020, 42(11): 2788-2795. doi: 10.11999/JEIT190452
Citation: Hongyan WANG, Haikun ZHANG. Moving Object Detection Method Based on Low-Rank and Sparse Decomposition in Dynamic Background[J]. Journal of Electronics and Information Technology, 2020, 42(11): 2788-2795. doi: 10.11999/JEIT190452

动态背景下基于低秩及稀疏分解的动目标检测方法

doi: 10.11999/JEIT190452
基金项目: 国家自然科学基金(61301258, 61271379),中国博士后科学基金(2016M590218),重点实验室基金(61424010106),河南省高等学校重点科研项目支持计划(14A520079),河南省科技攻关计划(162102210168)
详细信息
    作者简介:

    王洪雁:男,1979年生,副教授,博士,主要研究方向为MIMO雷达信号处理,毫米波通信,机器视觉

    张海坤:男,1995年生,硕士生,主要研究方向为图像处理,计算机视觉

    通讯作者:

    王洪雁 gglongs@163.com

  • 中图分类号: TN911.73; TP391

Moving Object Detection Method Based on Low-Rank and Sparse Decomposition in Dynamic Background

Funds: The National Natural Science Foundation of China (61301258, 61271379), China Postdoctoral Science Foundation (2016M590218), The Key Laboratory Foundation (61424010106), The Henan Province Support Plans for Key Scientific Research Projects of Colleges and Universities (14A520079), The Henan Province Plans for Science and Technology Development (162102210168)
  • 摘要: 针对背景运动引起动目标检测精度显著下降的问题,该文提出一种基于低秩及稀疏分解的动目标检测方法。所提方法首先引入伽马范数($\gamma {\rm{ - norm}}$)近乎无偏地逼近秩函数以解决核范数过度惩罚较大奇异值从而导致所得最小化问题无法获得最优解进而降低检测性能的问题,而后利用${L_{{1 / 2}}}$范数抽取稀疏前景目标以增强对噪声的稳健性,同时基于虚警像素所具有稀疏且空间不连续特性提出空间连续性约束以抑制动态背景像素,进而构建目标检测模型。最后利用基于交替方向最小化(ADM)策略扩展的增广拉格朗日乘子(ALM)法对所得优化问题求解。实验结果表明,与现有主流算法对比,所提方法可显著改善动态背景情况下动目标检测精度。
  • 图  1  不同场景下F-measure随$\gamma $取值变化曲线

    图  2  检测结果对比

    图  3  动目标检测定量分析对比

    图  4  各部分性能提升对比

    表  1  低秩与稀疏分解动目标检测方法

     算法:使用ADM策略扩展的ALM法求解问题式(7)
     输入:观测矩阵${{Z}}$,参数$\gamma $, ${\lambda _1}$, ${\lambda _2}$, ${\mu _1}$, ${\mu _2}$和$\varphi $。
     输出:${{H}}$, ${{K}}$和${{G}}$。
     (1):固定其他变量,计算式(12)以更新变量${{H}}$;
     (2):固定其他变量,由式(17)更新变量${{K}}$;
     (3):固定其他变量,计算式(22)以更新变量${{G}}$;
     (4):由式(23)和式(24)更新拉格朗日乘子${{{Y}}_1}$和${{{Y}}_2}$;
     (5):重复步骤(1)—(4),直至满足收敛条件。
    下载: 导出CSV

    表  2  不同场景下6种算法评价指标平均值

    评价指标PCPMoGPRMFDECBRPCA本文算法
    Precision0.47150.48960.55560.69380.79080.8967
    Recall0.78880.79780.81930.91990.89530.9181
    F-measure0.54400.58850.63870.76430.83330.9022
    下载: 导出CSV

    表  3  不同动目标检测算法平均运行时间对比(s)

    算法PCPMoGPRMFDECBRPCA本文算法
    运行时间541.55177.70105.31288.366161.29498.42
    下载: 导出CSV
  • [1] LUO Yuan, ZHOU Hanxing, TAN Qin, et al. Key frame extraction of surveillance video based on moving object detection and image similarity[J]. Pattern Recognition and Image Analysis, 2018, 28(2): 225–231. doi:  10.1134/S1054661818020190
    [2] SENTHIL M A, SUGANYA D K, SIVARANJANI A, et al. A study on various methods used for video summarization and moving object detection for video surveillance applications[J]. Multimedia Tools and Applications, 2018, 77(3): 23273–23290. doi:  10.1007/s11042-018-5671-8
    [3] CHEN Changhong, LIANG Jimin, ZHAO Heng, et al. Frame difference energy image for gait recognition with incomplete silhouettes[J]. Pattern Recognition Letters, 2009, 30(11): 977–984. doi:  10.1016/j.patrec.2009.04.012
    [4] KE Ruimin, LI Zhibin, TANG Jinjun, et al. Real-time traffic flow parameter estimation from UAV video based on ensemble classifier and optical flow[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 20(1): 54–64. doi:  10.1109/TITS.2018.2797697
    [5] YONG Hongwei, MENG Deyu, ZUO Wangmeng, et al. Robust online matrix factorization for dynamic background subtraction[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(7): 1726–1740. doi:  10.1109/TPAMI.2017.2732350
    [6] LIU Hongkun, DAI Jialun, WANG Ruchen, et al. Combining background subtraction and three-frame difference to detect moving object from underwater video[C]. OCEANS 2016, Shanghai, China, 2016: 1–5.
    [7] SEVILLA-LARA L, SUN Deqing, JAMPANI V, et al. Optical flow with semantic segmentation and localized layers[C]. The IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, USA, 2016: 3889–3898.
    [8] PRASAD D K, PRASATH C K, RAJAN D, et al. Object detection in a maritime environment: Performance evaluation of background subtraction methods[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 20(5): 1787–1802. doi:  10.1109/tits.2018.2836399
    [9] ZENG Zhi, JIA Jianyuan, YU Dalin, et al. Pixel modeling using histograms based on fuzzy partitions for dynamic background subtraction[J]. IEEE Transactions on Fuzzy Systems, 2017, 25(3): 584–593. doi:  10.1109/TFUZZ.2016.2566811
    [10] STAUFFER C and GRIMSON W E L. Adaptive background mixture models for real-time tracking[C]. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Fort Collins, USA, 1999: 246–252.
    [11] ZIVKOVIC Z. Improved adaptive Gaussian mixture model for background subtraction[C]. The 17th International Conference on Pattern Recognition, Cambridge, UK, 2004: 28–31.
    [12] CANDÈS E J, LI Xiaodong, MA Yi, et al. Robust principal component analysis?[J]. Journal of the ACM, 2011, 58(3): 11. doi:  10.1145/1970392.1970395
    [13] DING Xinghao, HE Lihan, and CARIN L. Bayesian robust principal component analysis[J]. IEEE Transactions on Image Processing, 2011, 20(12): 3419–3430. doi:  10.1109/TIP.2011.2156801
    [14] GAO Junbin. Robust L1 principal component analysis and its Bayesian variational inference[J]. Neural Computation, 2008, 20(2): 555–572. doi:  10.1162/neco.2007.11-06-397
    [15] WANG Naiyan, YAO Tiansheng, WANG Jingdong, et al. A probabilistic approach to robust matrix factorization[C]. The 12th European Conference on Computer Vision, Florence, Italy, 2012: 126–139.
    [16] ZHOU Xiaowei, YANG Can, and YU Weichuan. Moving object detection by detecting contiguous outliers in the low-rank representation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(3): 597–610. doi:  10.1109/TPAMI.2012.132
    [17] WANG Yi, JODOIN P M, PORIKLI F, et al. CDnet 2014: An expanded change detection benchmark dataset[C]. 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops, Columbus, USA, 2014: 393–400.
    [18] WANG Shusen, LIU Dehua, and ZHANG Zhihua. Nonconvex relaxation approaches to robust matrix recovery[C]. The 23rd International Joint Conference on Artificial Intelligence, Beijing, China, 2013: 1764–1770.
    [19] JIA Sen, ZHANG Xiujun, and LI Qingquan. Spectral–spatial hyperspectral image classification using 1/2 regularized low-rank representation and sparse representation-based graph cuts[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 8(6): 2473–2484. doi:  10.1109/JSTARS.2015.2423278
    [20] ZHANG Hengmin, YANG Jian, XIE Jianchun, et al. Weighted sparse coding regularized nonconvex matrix regression for robust face recognition[J]. Information Sciences, 2017, 394–395: 1–17. doi:  10.1016/j.ins.2017.02.020
    [21] CAO Wenfei, SUN Jian, and XU Zongben. Fast image deconvolution using closed-form thresholding formulas of regularization[J]. Journal of Visual Communication and Image Representation, 2013, 24(1): 31–41. doi:  10.1016/j.jvcir.2012.10.006
    [22] CAO Xiaochun, YANG Liang, and GUO Xiaojie. Total variation regularized RPCA for irregularly moving object detection under dynamic background[J]. IEEE Transactions on Cybernetics, 2016, 46(4): 1014–1027. doi:  10.1109/TCYB.2015.2419737
    [23] CHAMBOLLE A. An algorithm for total variation minimization and applications[J]. Journal of Mathematical Imaging and Vision, 2004, 20(1/2): 89–97. doi:  10.1023/b:jmiv.0000011325.36760.1e
    [24] XUE Yawen, GUO Xiaojie, and CAO Xiaochun. Motion saliency detection using low-rank and sparse decomposition[C]. 2012 IEEE International Conference on Acoustics, Speech and Signal Processing, Kyoto, Japan, 2012: 1485–1488.
    [25] XU Zongben, CHANG Xiangyu, XU Fengmin, et al. L1/2 regularization: A thresholding representation theory and a fast solver[J]. IEEE Transactions on Neural Networks and Learning Systems, 2012, 23(7): 1013–1027. doi:  10.1109/TNNLS.2012.2197412
    [26] BECK A and TEBOULLE M. Fast gradient-based algorithms for constrained total variation image denoising and deblurring problems[J]. IEEE Transactions on Image Processing, 2009, 18(11): 2419–2434. doi:  10.1109/TIP.2009.2028250
    [27] MAHADEVAN V and VASCONCELOS N. Spatiotemporal saliency in dynamic scenes[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(1): 171–177. doi:  10.1109/TPAMI.2009.112
    [28] GOLUB G H and REINSCH C. Singular value decomposition and least squares solutions[J]. Numerische Mathematik, 1970, 14(5): 403–420. doi:  10.1007/BF02163027
  • [1] 张思乾, 于美婷, 匡纲要.  一种低秩张量约束的下视稀疏线阵SAR三维成像算法, 电子与信息学报. doi: 10.11999/JEIT200274
    [2] 邹鲲, 来磊, 骆艳卜, 李伟.  子空间约束的一阶秩1干扰下的自适应检测, 电子与信息学报. doi: 10.11999/JEIT200259
    [3] 唐红梅, 白梦月, 韩力英, 梁春阳.  基于低秩背景约束与多线索传播的图像显著性检测, 电子与信息学报. doi: 10.11999/JEIT200193
    [4] 孟智超, 卢景月, 张帅钦, 张磊, 王虹现.  单通道合成孔径雷达抗调频斜率失配干扰新方法, 电子与信息学报. doi: 10.11999/JEIT190687
    [5] 王洪雁, 于若男.  基于稀疏和低秩恢复的稳健DOA估计方法, 电子与信息学报. doi: 10.11999/JEIT190263
    [6] 汪玲, 朱栋强, 马凯莉, 肖卓.  空间目标卡尔曼滤波稀疏成像方法, 电子与信息学报. doi: 10.11999/JEIT170319
    [7] 杨依忠, 汪鹏飞, 胡雄楼, 伍能举.  基于鲁棒主成分分析的运动目标检测优化算法, 电子与信息学报. doi: 10.11999/JEIT170789
    [8] 刘凯, 冯辉, 杨涛, 胡波.  上行3D-MIMO中利用结构稀疏低秩特性的信道估计算法, 电子与信息学报. doi: 10.11999/JEIT170399
    [9] 张涛, 唐振民.  一种基于非负低秩稀疏图的半监督学习改进算法, 电子与信息学报. doi: 10.11999/JEIT160559
    [10] 杨爱萍, 张莉云, 曲畅, 王建.  基于加权L1正则化的水下图像清晰化算法, 电子与信息学报. doi: 10.11999/JEIT160481
    [11] 欧伟枫, 杨春玲, 戴超.  一种视频压缩感知中两级多假设重构及实现方法, 电子与信息学报. doi: 10.11999/JEIT161142
    [12] 李海洋, 王恒远.  基于TL1范数约束的子空间聚类方法, 电子与信息学报. doi: 10.11999/JEIT170193
    [13] 李少东, 杨军, 陈文峰, 马晓岩.  基于压缩感知理论的雷达成像技术与应用研究进展, 电子与信息学报. doi: 10.11999/JEIT150874
    [14] 王斯琪, 冯象初, 张瑞, 李小平.  基于最大范数的低秩稀疏分解模型, 电子与信息学报. doi: 10.11999/JEIT150468
    [15] 唐琳, 焦淑红, 齐欢, 吴如煊.  一种单脉冲雷达多通道L1正则化波束锐化方法, 电子与信息学报. doi: 10.3724/SP.J.1146.2013.01792
    [16] 占荣辉, 胡杰民, 张军.  基于压缩感知的二维GTD模型参数估计方法, 电子与信息学报. doi: 10.3724/SP.J.1146.2012.00780
    [17] 席峰, 陈胜垚, 刘中.  混沌模拟信息转换基于多射法的稀疏信号重构, 电子与信息学报. doi: 10.3724/SP.J.1146.2012.00905
    [18] 李拥军, 曾标, 徐克付, 李阳.  复杂背景下基于贝叶斯-全概率联合估计的前景检测, 电子与信息学报. doi: 10.3724/SP.J.1146.2011.00626
    [19] 陈胜垚, 席峰, 刘中.  基于混沌压缩感知的稀疏时变信号在线估计, 电子与信息学报. doi: 10.3724/SP.J.1146.2011.00620
    [20] 寇波, 江海, 刘磊, 张冰尘.  基于压缩感知的SAR抑制旁瓣技术研究, 电子与信息学报. doi: 10.3724/SP.J.1146.2010.00103
  • 加载中
  • 图(4) / 表ll (3)
    计量
    • 文章访问数:  75
    • HTML全文浏览量:  47
    • PDF下载量:  29
    • 被引次数: 0
    出版历程
    • 收稿日期:  2019-06-20
    • 修回日期:  2020-04-20
    • 网络出版日期:  2020-08-29
    • 刊出日期:  2020-11-10

    目录

      /

      返回文章
      返回

      官方微信,欢迎关注