基于张量核范数与3D全变分的背景减除

doi:10.11772/j.issn.1001-9081.2020010005

计算机应用 ›› 2020, Vol. 40 ›› Issue (9): 2737-2742.DOI: 10.11772/j.issn.1001-9081.2020010005

• 虚拟现实与多媒体计算 • 上一篇下一篇

基于张量核范数与3D全变分的背景减除

陈利霞^1,2, 班颖^1,2, 王学文³

1. 桂林电子科技大学数学与计算科学学院, 广西桂林 541004;
2. 广西高校数据分析与计算重点实验室(桂林电子科技大学), 广西桂林 541004;
3. 桂林电子科技大学计算机与信息安全学院, 广西桂林 541004

收稿日期:2020-01-15 修回日期:2020-04-24 出版日期:2020-09-10 发布日期:2020-05-06
通讯作者: 王学文
作者简介:陈利霞(1979-),女,湖北浠水人,教授,博士,主要研究方向:数字图像处理中的数学理论与算法;班颖(1995-),女,河北廊坊人,硕士研究生,主要研究方向:数字图像处理中的数学理论与算法;王学文(1979-),男,湖北浠水人,讲师,硕士,主要研究方向:数字图像处理。
基金资助:
国家自然科学基金资助项目（11961010）；广西自然科学基金资助项目（2018GXNSFAA138169）。

Background subtraction based on tensor nuclear norm and 3D total variation

CHEN Lixia^1,2, BAN Ying^1,2, WANG Xuewen³

1. School of Mathematics and Computing Science, Guilin University of Electronic Technology, Guilin Guangxi 541004, China;
2. Guangxi Colleges and Universities Key Laboratory of Data Analysis and Computation(Guilin University of Electronic Technology), Guilin Guangxi 541004, China;
3. School of Computer Science and Information Security, Guilin University of Electronic Technology, Guilin Guangxi 541004, China

Received:2020-01-15 Revised:2020-04-24 Online:2020-09-10 Published:2020-05-06
Supported by:
This work is partially supported by the National Natural Science Foundation of China (11961010), the Natural Science Foundation of Guangxi (2018GXNSFAA138169).

摘要/Abstract

摘要： 针对常用背景减除方法忽略前景时空连续性的问题，以及动态背景对前景提取的干扰问题，基于张量鲁棒主成分分析（TRPCA）提出了一种改进的背景减除模型。该模型利用改进的张量核范数对背景进行约束，加强了背景的低秩性，保留了视频的空间信息；然后用3D全变分（3D-TV）对前景进行正则化约束，考虑了目标在时空上的连续性，有效地抑制了动态背景和目标移动对前景提取造成的干扰。实验结果表明，所提算法能有效地分离视频中的前景和背景，且与高阶鲁棒主成分分析（HoRPCA）、带有新核范数的张量鲁棒主成分分析（TRPCA-TNN）和基于克罗内克基的鲁棒主成分分析（KBR-RPCA）等方法相比，综合评判指标F-measure值均处于最优或次优状态。由此可见，所提算法有效地提高了前景背景分离的准确度，抑制了复杂天气和目标移动对前景提取的干扰。

关键词: 背景减除, 张量鲁棒主成分分析, 张量核范数, 3D全变分, 交替方向乘子法

Abstract: Concerning the fact that common background subtraction methods ignore the spatio-temporal continuity of foreground and the disturbance of dynamic background to foreground extraction, an improved background subtraction model was proposed based on Tensor Robust Principal Component Analysis (TRPCA). The improved tensor nuclear norm was used to constrain the background, which enhanced the low rank of background and retained the spatial information of videos. Then the regularization constraint was performed to the foreground by 3D Total Variation (3D-TV), so as to consider the spatio-temporal continuity of object and effectively suppress the interference of dynamic background and target movement on the foreground extraction. Experimental results show that the proposed model can effectively separate the foreground and background of videos. Compared with High-order Robust Principal Component Analysis (HoRPCA), Tensor Robust Principal Component Analysis with Tensor Nuclear Norm (TRPCA-TNN) and Kronecker-Basis-Representation based Robust Principal Component Analysis (KBR-RPCA), the proposed algorithm has the F-measure values all optimal or sub-optimal. It can be seen that, the proposed model effectively improves the accuracy of foreground and background separation, and suppresses the interference of complex weather and target movement on foreground extraction.

Key words: background subtraction, Tensor Robust Principal Component Analysis (TRPCA), tensor nuclear norm, 3D Total Variation (3D-TV), Alternating Direction Multiplier Method (ADMM)

中图分类号:

TP391.413

陈利霞, 班颖, 王学文. 基于张量核范数与3D全变分的背景减除[J]. 计算机应用, 2020, 40(9): 2737-2742.

CHEN Lixia, BAN Ying, WANG Xuewen. Background subtraction based on tensor nuclear norm and 3D total variation[J]. Journal of Computer Applications, 2020, 40(9): 2737-2742.

参考文献

[1] 孙吉花. 背景减除的算法研究[D]. 长沙:国防科技大学,2006:1-10.(SUN J H. Research on algorithm of background subtraction[D]. Changsha:National University of Defense Technology,2006:1-10.)
[2] ZHOU X,YANG C,ZHAO H,et al. Low-rank modeling and its applications in image analysis[J]. ACM Computing Surveys,2014, 47(2):No. 36.
[3] LIN Z,CHEN M,MA Y. The augmented Lagrange multiplier method for exact recovery of corrupted low-rank matrices[J]. Journal of Structural Biology,2010,181(2):116-127.
[4] 陈利霞, 刘俊丽, 王学文. 结合加权Schatten-p范数与3D全变分的前景检测[J]. 计算机应用,2019,39(4):1170-1175.(CHEN L X,LIU J L,WANG X W. Foreground detection with weighted Schatten-p norm and 3D total variation[J]. Journal of Computer Applications,2019,39(4):1170-1175.)
[5] GOLDFARB D,QIN Z. Robust low-rank tensor recovery:models and algorithms[J]. SIAM Journal on Matrix Analysis and Applications,2014,35(1):225-253.
[6] XIE Q,ZHAO Q,MENG D,et al. Kronecker-basis-representation based tensor sparsity and its applications to tensor recovery[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018,40(8):1888-1902.
[7] LIU Y, CHEN L, ZHU C. Improved robust tensor principal component analysis via low-rank core matrix[J]. IEEE Journal of Selected Topics in Signal Processing,2018,12(6):1378-1389.
[8] LU C, FENG J, CHEN Y, et al. Tensor robust principal component analysis with a new tensor nuclear norm[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2020, 42(4):925-938.
[9] YUAN M,LIN Y. Model selection and estimation in regression with grouped variables[J]. Journal of the Royal Statistical Society:Series B(Statistical Methodology),2006,68(1):49-67.
[10] XU Y,WU Z,CHANUSSOT J,et al. Joint reconstruction and anomaly detection from compressive hyperspectral images using Mahalanobis distance regularized tensor RPCA[J]. IEEE Transactions on Geoscience and Remote Sensing,2018,56(6):2919-2930.
[11] CAO W,WANG Y,SUN J,et al. Total variation regularized tensor RPCA for background subtraction from compressive measurements[J]. IEEE Transactions on Image Processing, 2016,25(9):4075-4090.
[12] WANG Y,YIN W,ZENG J. Global convergence of ADMM in non-convex non-smooth optimization[J]. Journal of Scientific Computing,2019,78(1):29-63.
[13] BOYD S,PARIKH N,CHU E,et al. Distributed optimization and statistical learning via the alternating direction method of multipliers[J]. Foundations and Trends in Machine Learning, 2011,3(1):1-122.
[14] TOM A J, GEORGE S N. Tensor total variation regularized moving object detection for surveillance videos[C]//Proceedings of the 2018 International Conference on Signal Processing and Communications. Piscataway:IEEE,2018:327-331.
[15] WANG Y,JODOIN P M,PORIKLI F,et al. CDnet 2014:an expanded change detection benchmark dataset[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops. Piscataway:IEEE,2014:393-400.
[16] 刘仲民, 何胜皎, 胡文瑾, 等. 基于背景减除法的视频序列运动目标检测[J]. 计算机应用,2017,37(6):1777-1781.(LIU Z M,HE S J,HU W J,et al. Moving object detection based on background subtraction for video sequence[J]. Journal of Computer Applications,2017,37(6):1777-1781.)

基于张量核范数与3D全变分的背景减除

Background subtraction based on tensor nuclear norm and 3D total variation

PDF

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

[1]	李辉, 吴传生, 刘俊, 刘文. 基于梯度曲面面积与稀疏约束的图像平滑方法[J]. 计算机应用, 2021, 41(7): 2039-2047.
[2]	沈马锐, 李金城, 张亚, 邹健. 基于非凸全变差正则的核磁共振图像重构算法[J]. 计算机应用, 2020, 40(8): 2358-2364.
[3]	耿源谦, 吴传生, 刘文. 混合非凸非光滑正则化约束的模糊图像盲复原[J]. 计算机应用, 2020, 40(4): 1171-1176.
[4]	陈利霞, 刘俊丽, 王学文. 结合加权Schatten-p范数与3D全变分的前景检测[J]. 计算机应用, 2019, 39(4): 1170-1175.
[5]	刘仲民, 何胜皎, 胡文瑾, 李战明. 基于背景减除法的视频序列运动目标检测[J]. 计算机应用, 2017, 37(6): 1777-1781.
[6]	许浩锋, 凌青. 分布式在线交替方向乘子法[J]. 计算机应用, 2015, 35(6): 1595-1599.
[7]	高乾坤王玉军王惊晓. 基于交替方向乘子法的非光滑损失坐标优化算法[J]. 计算机应用, 2013, 33(07): 1912-1916.
[8]	陈雷张荣国胡静刘焜. 改进的混合高斯模型及阴影消除方法[J]. 计算机应用, 2013, 33(05): 1394-1400.
[9]	张艳郭继昌王琛. 运动目标检测中的环境感知与自适应研究[J]. 计算机应用, 2011, 31(07): 1827-1830.