低秩矩阵和结构化稀疏分解的视频背景差分方法

刘鑫 1; 张钊强 1; 姚佳文 1; 郭莉莉 2; 齐春 1

doi:10.7652/xjtuxb201606004

您当前的位置：

首页 >

文章列表页 >

低秩矩阵和结构化稀疏分解的视频背景差分方法

更新时间：2025-07-09

- 低秩矩阵和结构化稀疏分解的视频背景差分方法
- A Background Subtraction Method Based on Decomposition of Low Rank and Sparsity Matrices
- 西安交通大学学报 2016年50卷第6期页码：23-29
- 作者机构：
  
  西安交通大学电子与信息工程学院,西安,710049
- 作者简介：
- 基金信息：
  
  : 2015-12-25。作者简介: 刘鑫(1981—),男,博士生;齐春(通信作者),男,教授,博士生导师。基金项目: 国家自然科学基金资助项目(61572395)
- DOI：10.7652/xjtuxb201606004
  中图分类号： TP391
- 网络首发：2016-06-10，
  
  纸质出版：2016
- 稿件说明：
移动端阅览
刘鑫 1, 张钊强 1, 姚佳文 1, 等. 低秩矩阵和结构化稀疏分解的视频背景差分方法[J]. 西安交通大学学报, 2016,50(6):23-29.

A Background Subtraction Method Based on Decomposition of Low Rank and Sparsity Matrices[J]. 2016, 50(6): 23-29.
刘鑫 1, 张钊强 1, 姚佳文 1, 等. 低秩矩阵和结构化稀疏分解的视频背景差分方法[J]. 西安交通大学学报, 2016,50(6):23-29. DOI： 10.7652/xjtuxb201606004.

A Background Subtraction Method Based on Decomposition of Low Rank and Sparsity Matrices[J]. 2016, 50(6): 23-29. DOI： 10.7652/xjtuxb201606004.

摘要

针对基于矩阵分解的视频前景检测传统算法中忽视前景元素之间相关性会导致检测结果容易受噪声干扰和运动目标检测不完整等问题

提出了一个低秩矩阵和结构化稀疏分解的视频背景差分算法。该算法充分考虑到视频前景区域的结构化分布特征

利用结构化稀疏范数对前景进行约束; 针对矩阵分解方法中参数选择的难题

采用了一种基于运动显著性判定的两步法来实现动态背景去除和正则化参数的自适应选择

即第一步利用低秩和结构化稀疏分解获得运动候选块

第二步对运动候选块进行显著性分析并利用自适应正则化参数的块稀疏分解进行前景检测。实验结果表明:与现有的基于矩阵分解的前景检测方法相比

该算法能够更加适应复杂多变的视频环境

在I2R测试库中检测出的前景有较高的精确度和召回率。

Abstract

A background subtraction method based on decomposition of low-rank and structured sparsity matrices is proposed to solve the problem that detection results are sensitive to noise and incomplete caused by ignoring the relationship between foreground pixels in traditional foreground detection methods based on matrix decomposition. The method takes the structural distribution of the foreground into account

and a structured sparsity constraint is used on the foreground pixels. Moreover

a two-stage framework based on motion saliency is introduced to address the parameter setting issue in dynamic background videos and to tune regularization parameters adaptively. Motion block candidates are obtained by using the low rank and structured sparsity decomposition in the first step. Then

motion saliency analysis is applied to these candidates and the adapt block sparsity decomposition is used to detect the foreground in the second step. Experimental results show that the performance of the proposed method is more adaptive than the existing foreground detection methods based on matrix decomposition in complex videos

and that the proposed approach outperforms the state-of-the-art methods according to the precision and recall results on dataset I2R.

关键词

Keywords

references

王伟嘉, 刘辉, 沙莉, 等. 滞留与偷窃物体实时检测与分类算法 [J]. 计算机应用, 2007, 27(10): 2591-2594.

WANG Weijia, LIU Hui, SHA Li, et al. Real time detection and classification algorithm for abandoned and stolen objects [J]. Computer Applications, 2007, 27(10): 2591-2594.

STAUFFER C, GRIMSON W E L. Adaptive background mixture models for real-time tracking [C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 1999: 246-252.

刘鑫, 刘辉, 强振平, 等. 混合高斯模型和帧间差分相融合的自适应背景模型 [J]. 中国图象图形学报, 2008, 13(4): 729-734.

LIU Xin, LIU Hui, QIANG Zhengping, et al. Adaptive background modeling based on mixture Gaussian model and frame subtraction [J]. Journal of Image and Graphic, 2008, 13(4): 729-734.

LI L Y, HUANG W M, GU I H, et al. Statistical modeling of complex backgrounds for foreground object detection [J]. IEEE Transactions on Image Processing, 2004, 13(11): 1459-1472.

SHEIKH Y, SHAH M. Bayesian modeling of dynamic scenes for object detection [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(11): 1778-1792.

KIM K, CHALIDABHONGSE T H, HARWOOD D, et al. Realtime foreground-background segmentation using codebook model [J]. Real-Time Imagine, 2005, 11(3): 172-185.

BARNICH O, VAN DROOGENBROECK M. ViBe: a universal background subtraction algorithm for video sequences [J]. IEEE Transactions on Image Processing, 2011, 20(6): 1709-1724.

CANDES E J, LI X, MA Y, et al. Robust principal component analysis? [J]. Journal of the ACM, 2011, 58(3): 11-47.

WRIGHT J, YANG A Y, GANESH A, et al. Robust face recognition via sparse representation [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(2): 210-227.

TIBSHIRANI R. Regression shrinkage and selection via the lasso [J]. Journal of the Royal Statistical Society: Series B Methodological, 1996, 58(1): 267-288.

MAIRAL J, JENATTON R, BACH F R, et al. Network Flow Algorithms for Structured Sparsity [C]∥Proceedings of Neural Information Processing Systems. Vancouver, Canada: NIPS, 2010: 1558-1566.

GUYON C, BOUWMANS T, ZAHZAH E H. Foreground detection based on low-rank and block-sparse matrix decomposition [C]∥Proceedings of the IEEE International Conference on Image Processing. Piscataway, NJ, USA: IEEE, 2012: 1225-1228.

TANG G, NEHORAI A. Robust principal component analysis based on low-rank and block-sparse matrix decomposition [C]∥Proceedings of the Annual Conference on Information Sciences and Systems. Piscataway, NJ, USA: IEEE, 2011: 1-5.

ZHOU X W, YANG C, YU W. 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.

GAO Z, CEONG L F, SHAN M. Block-sparse RPCA for consistent foreground detection [C]∥Proceedings of the European Conference on Computer Vision. Heidelberg, Germany: Springer, 2012: 690-703.

CUI X Y, LIU Q S, ZHANG S T, et al. Temporal Spectral Residual for fast salient motion detection [J]. Neurocomputing, 2012, 86: 24-32.

LIN Z, CHEN M, MA Y. The augmented Lagrange multiplier method for exact recovery of corrupted low-rank matrices [EB/OL].(2013-10-18)[2015-11-20]. http: ∥arxiv.org/pdf/1009.5055v3.pdf.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

消除阴影和高亮噪声的双梯柱体码本前景检测模型

电动车辆间接热系统快速除湿响应迟滞机理及优化控制策略

融合物理信息与深度算子网络的变压器绕组热-流场快速计算方法

超临界水异常区域热力学特性研究

不同板叠材料热声热机能量转换过程数值模拟研究