基于低秩-稀疏联合表示的视频序列运动目标检测

doi:10.16182/j.issn1004731x.joss.201812025

系统仿真学报 ›› 2018, Vol. 30 ›› Issue (12): 4693-4702.doi: 10.16182/j.issn1004731x.joss.201812025

基于低秩-稀疏联合表示的视频序列运动目标检测

杨磊¹, 庞芳¹, 胡豁生²

1.上海大学机电工程与自动化学院,上海 200444;
2.埃塞克斯大学计算机科学与电气工程学院,科尔切斯特 CO4 3SQ

收稿日期:2018-04-20 修回日期:2018-07-02 出版日期:2018-12-10 发布日期:2019-01-03

Low-Rank Sparse joint Representation for Moving Object Detection in Video

Yang Lei¹, Pang Fang¹, Hu Huosheng²

1. School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200444, China;
2. School of Computer Science and Electrical Engineering, University of Essex, Colchester, CO4 3SQ, United Kingdom

Received:2018-04-20 Revised:2018-07-02 Online:2018-12-10 Published:2019-01-03
About author:Yang Lei (1976-), male, Shanxi, China, an associate researcher and postgraduate advisor; research interests: artificial intelligence and pattern recognition, computer vision and image processing.
Supported by:
National Natural Science Foundation of China (31100709), Shanghai science and Technology Committee (18411952200)

摘要/Abstract

摘要： 对于固定摄像机的视频序列,假设背景具有低秩特征,动态前景具有稀疏特性,提出了一种基于低秩稀疏联合表示的运动检测方法。思路如下:通过图像预处理降低视频序列的噪声;估计连续帧之间的光流,生成二进制运动掩模作为运动权重矩阵;基于子空间学习理论,建立了低秩背景与稀疏前景的优化模型;利用ADMM-BCD迭代算法得到视频背景和前景。实验结果表明,该方法优于其他同类运动检测方法,对慢速运动目标检测效果良好。

关键词: 鲁棒主成分分析, 子空间学习, 背景-前景建模, 运动检测

Abstract: For the video sequences with fixed cameras, it is a reasonable assumption that the fixed background has low-rank characteristic, and the dynamic foreground has sparse characteristic. A new motion detection method based on low-rank and sparse joint representation is proposed in this paper. The ideas of the proposed method are described as follows: The noise of video sequence is removed by image preprocessing. The optical flow between continuous video sequences is estimated, which is used to generate a binary motion mask as a movement weight matrix. An optimization model with low-rank background and sparse foreground is established based on the idea of subspace learning theory. The background and foreground of each frame are obtained by using the ADMM-BCD iterative algorithm. Experimental results show that the proposed method is super to the other same sort of moving detection methods. The proposed method has perfect effect on slow moving target detection.

Key words: robust principal component analysis (RPCA), subspace learning, background-foreground modeling, motion detection

中图分类号:

TP301

杨磊, 庞芳, 胡豁生. 基于低秩-稀疏联合表示的视频序列运动目标检测[J]. 系统仿真学报, 2018, 30(12): 4693-4702.

Yang Lei, Pang Fang, Hu Huosheng. Low-Rank Sparse joint Representation for Moving Object Detection in Video[J]. Journal of System Simulation, 2018, 30(12): 4693-4702.

参考文献

[1] T B Moeslund, A Hilton, V Kruger.A survey of advances in vision-based human motion capture and analysis[J]. Computer Vision & Image Understanding (S1077-3142), 2006, 104(2): 90-126.
[2] R Poppe.A survey on vision-based human action recognition[J]. Image & Vision Computing (S0262-8856), 2010, 28(6): 976-990.
[3] A Ilyas, M Scuturici, S Miguet.Real-time foreground-background segmentation using a modified codebook model[C]. IEEE International Conference on Advanced Video and Signal Based Surveillance, 2009, 10: 454-459.
[4] A Mittal, N Paragios.Motion-based background subtraction using adaptive kernel density estimation[C]. IEEE Computer Society Conference on Computer Vision & Pattern Recognition (S1063-6919), 2004, 2: 302-309.
[5] O Barnich, M Van Droogenbroeck.ViBe: A universal background subtraction algorithm for video sequences[J]. IEEE Transactions on Image Processing (S1057-7149), 2011, 20(6): 1709-1724.
[6] E J Candès, X Li, Y Ma, et al.Robust principal component analysis?. [J]. ACM (S0004-5411), 2011, 58(3): 1-37.
[7] Y Peng, A Ganesh, J Wright, et al.RASL: Robust Alignment by Sparse and Low-Rank Decomposition for Linearly Correlated Images[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence (S0162-8828), 2012, 34(11): 2233-2246.
[8] T Zhou, D Tao.GoDec: randomized low-rank and sparse matrix decomposition in noisy case[C]. International Conference on Machine Learning, 2011, 33-40.
[9] Q Ke, T Kanade.Robust l-Norm Factorization in the Presence of Outliers and Missing Data by Alternative Convex Programming[C]. IEEE Computer Society Conference on Computer Vision & Pattern Recognition:(S1063-6919), 2005: 739-746.
[10] Y Shen, Z Wen, Y Zhang.Augmented lagrangian alternating direction method for matrix separation based on low-rank factorization[J]. Optimization Methods & Software(S1055-6788), 2011, 29(2): 239-263.
[11] Xianbiao Shu, Fatih Porikli, Narendra Ahuja.Robust Orthonormal Subspace Learning: Efficient Recovery of Corrupted Low-Rank Matrices[C]. 2014 IEEE Conference on Computer Vision and Pattern Recognition:(S1063-6919), 2014, 1: 3874-3881.
[12] Xinchen Ye, Jingyu Yang, Xin Sun, et al.Foreground-Background Separation From Video Clips via Motion-Assisted Matrix Restoration[J]. IEEE Transactions on Circuits and Systems for Video Technology (S1051-8215), 2015, 25(11): 1721-1734.
[13] S Javed, A Mahmood, T Bouwmans, et al.Background- Foreground Modeling Based on Spatiotemporal Sparse Subspace Clustering[J]. IEEE Transactions on Image Processing (S1057-7149), 2017, 26(12): 5840-5854.
[14] N Shahid, V Kalofolias, X Bresson, et al.Robust Principal Component Analysis on Graphs[C]. 2015 IEEE International Conference on Computer Vision (ICCV) :(S1550-5499), 2015: 2812-2820.
[15] S Boyd, N Parikh, E Chu, et al.Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers[J]. Foundations and Trends in Machine Learning:(S1935-8237), 2011, 3(1): 1-122.
[16] S Bengio, F Pereira, Y Singer, et al.Group sparse coding[C]. International Conference on Neural Information Processing Systems (S1049-5258), 2009, 22(1): 82-89.
[17] C Guyon, T Bouwmans, E H Zahzah.Moving Object Detection by Robust PCA Solved via a Linearized Symmetric Alternating Direction Method[C]. International Symposium on Visual Computing:(S0302-9743), 2013, 7431: 427-436.

基于低秩-稀疏联合表示的视频序列运动目标检测

Low-Rank Sparse joint Representation for Moving Object Detection in Video

PDF (PC)

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

本文评价

[1]	邓向阳, 张立民, 方伟, 汤淼. 基于双向汇聚引导蚁群算法的机器人路径规划[J]. 系统仿真学报, 2022, 34(5): 1101-1108.
[2]	宁涛, 苟涛, 刘向东. 考虑低碳约束的生鲜农产品冷链物流策略仿真研究[J]. 系统仿真学报, 2022, 34(4): 797-805.
[3]	李建锋, 卢迪, 李贺香. 一种改进的原子搜索算法[J]. 系统仿真学报, 2022, 34(3): 490-502.
[4]	董君, 叶春明. 半导体晶圆节能分布式制造与预维护联合优化[J]. 系统仿真学报, 2022, 34(3): 584-602.
[5]	孙小晴, 程昊, 张潞瑶, 季伟东, 王旭. 一种基于空间分割搜索策略的自然计算方法[J]. 系统仿真学报, 2021, 33(11): 2589-2605.
[6]	王付宇, 汤涛. 双种群鱼群算法在分布式投资组合的应用[J]. 系统仿真学报, 2021, 33(9): 2074-2084.
[7]	段红, 邱晓刚. 网络化仿真及其发展趋势[J]. 系统仿真学报, 2021, 33(7): 1526-1533.
[8]	段红, 邱晓刚, 谢旭, 鞠儒生. 仿真学科知识领域构成研究[J]. 系统仿真学报, 2021, 33(4): 763-772.
[9]	刘景森, 袁蒙蒙, 李煜. 基于改进樽海鞘群算法求解工程优化设计问题[J]. 系统仿真学报, 2021, 33(4): 854-866.
[10]	顾菊平, 程天宇, 王建平, 华亮, 赵凤申, 蒋凌. 基于几何特征不变量的快速3D医学图像配准[J]. 系统仿真学报, 2020, 32(11): 2105-2111.
[11]	倪婉璐, 季伟东, 孙小晴. 自适应动态控制种群分组的自然计算方法[J]. 系统仿真学报, 2020, 32(10): 1884-1894.
[12]	陆湛文, 程新功, 张永峰. 基于共识粒子群的全局优化求解方法[J]. 系统仿真学报, 2020, 32(10): 1936-1942.
[13]	张潞瑶, 季伟东, 程昊. 基于LLE降维思想的自然计算方法[J]. 系统仿真学报, 2020, 32(10): 1943-1955.
[14]	何奕涛, 李珺, 郝丽艳. 具有引力机制的细菌觅食算法[J]. 系统仿真学报, 2020, 32(9): 1724-1735.
[15]	刘长良, 王梓齐. 基于严格二叉树编码和GA的模糊树结构学习[J]. 系统仿真学报, 2020, 32(8): 1473-1480.