秩约束的快速鲁棒主成分分析算法及应用
Fast Robust Component Analysis with Rank Constraint and Applications
- 电子学报 2023年51卷第6期 页码：1448-1457
- 作者机构：
1.武汉科技大学信息科学与工程学院，湖北武汉 430081
2.武汉科技大学机械自动化学院，湖北武汉 430081
3.新乡学院机电工程学院，河南新乡 453003
- 作者简介：
[ "何锐 男，1995年出生于四川广安.现为武汉科技大学信息科学与工程学院控制科学与工程硕士研究生.主要研究方向为低秩矩阵分解和图像处理.E-mail: herui6543@163.com" ]
[ "伍世虔 男，1964年，江西赣州人.现为武汉科技大学信息科学与工程学院教授，博士生导师.主要研究领域包括计算机视觉、模式识别、机器学习及智能机器人.E-mail: shiqian.wu@wust.edu.cn" ]
- 基金信息：国家自然科学基金(61775172;61371190);河南省高校科技创新人才支持计划(21HASTIT021)
DOI：10.12263/DZXB.20210557
中图分类号： TN391.41;收稿：2021-04-30，
修回：2022-05-15，
纸质出版：2023-06-25
- 稿件说明：
移动端阅览
何锐,徐正勤,伍世虔等.秩约束的快速鲁棒主成分分析算法及应用[J].电子学报,2023,51(06):1448-1457.
HE Rui,XU Zheng-qin,WU Shi-qian,et al.Fast Robust Component Analysis with Rank Constraint and Applications[J].ACTA ELECTRONICA SINICA,2023,51(06):1448-1457.
何锐,徐正勤,伍世虔等.秩约束的快速鲁棒主成分分析算法及应用[J].电子学报,2023,51(06):1448-1457. DOI： 10.12263/DZXB.20210557.
HE Rui,XU Zheng-qin,WU Shi-qian,et al.Fast Robust Component Analysis with Rank Constraint and Applications[J].ACTA ELECTRONICA SINICA,2023,51(06):1448-1457. DOI： 10.12263/DZXB.20210557.
摘要
鲁棒主成分分析被广泛应用于计算机视觉领域，然而现有鲁棒主成分分析方法难以针对各种场景准确分离出低秩信息，而且计算成本高导致算法的实时性不足.针对这两个问题，本文提出了一种新型鲁棒主成分分析算法.一方面基于先验秩信息提出了低秩约束改进模型，提高算法在不同场景中的泛化性能；另一方面引入了黎曼优化理论，将目标矩阵投影到低维子空间上求解，减少算法的运算复杂度.各种实验结果表明，与现有算法相比，改进算法在速度上有非常大的优势，同时能够保证稳定的恢复能力.
Abstract
Robust principal component analysis (RPCA) is widely used in the field of computer vision. The current RPCA method is facing the issue of difficulty to accurately separate low-rank information
as well as low efficiency due to high computational cost. To solve both problems
a novel RPCA optimization is proposed. Specifically
a low-rank constraint model is presented based on the prior rank information to improve the recovery performance. On the other hand
Riemannian optimization
which projects the matrix into a low-dimensional subspace to reduce the complexity
is employed. Various experiments show that the proposed algorithm has significant advantages in terms of efficiency and recovery accuracy in comparison to the existing algorithms.
关键词
Keywords
references
常侃 , 张智勇 , 陈诚 , 等 . 采用低秩与加权稀疏分解的视频前景检测算法 [J]. 电子学报 , 2017 , 45 ( 9 ): 2272 - 2280 .
CHANG K , ZHANG Z Y , CHEN C , et al . Video foreground detection by low-rank and reweighted sparse decomposition [J]. Acta Electronica Sinica , 2017 , 45 ( 9 ): 2272 - 2280 . (in Chinese)
JAVED S , MAHMOOD A , AL-MAADEED S , et al . Moving object detection in complex scene using spatiotemporal structured-sparse RPCA [J]. IEEE Transactions on Image Processing , 2019 , 28 ( 2 ): 1007 - 1022 .
LI L H , HU Q H , LI X . Moving object detection in video via hierarchical modeling and alternating optimization [J]. IEEE Transactions on Image Processing , 2019 , 28 ( 4 ): 2021 - 2036 .
ZHANG W M , ZHAO X , MORVAN J M , et al . Improving shadow suppression for illumination robust face recognition [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2019 , 41 ( 3 ): 611 - 624 .
XUE Z C , DONG J , ZHAO Y X , et al . Low-rank and sparse matrix decomposition via the truncated nuclear norm and a sparse regularizer [J]. The Visual Computer , 2019 , 35 ( 11 ): 1549 - 1566 .
AHN H , KANG M . Dynamic background subtraction with masked RPCA [J]. Signal, Image and Video Processing , 2021 , 15 ( 3 ): 467 - 474 .
BOUWMANS T , JAVED S , ZHANG H Y , et al . On the applications of robust PCA in image and video processing [J]. Proceedings of the IEEE , 2018 , 106 ( 8 ): 1427 - 1457 .
LIKASSA H T . New robust principal component analysis for joint image alignment and recovery via affine transformations, frobenius and L 2, 1 norms [J]. International Journal of Mathematics and Mathematical Sciences , 2020 , 2020 : 1 - 9 .
CANDÈS E J , LI X D , MA Y , et al . Robust principal component analysis? [J]. Journal of the ACM , 2011 , 58 ( 3 ): 11.1 - 11.37 .
LI C J . A new approximation of the matrix rank function and its application to matrix rank minimization [J]. Journal of Optimization Theory and Applications , 2014 , 163 ( 2 ): 569 - 594 .
LIN Z C , LIU R S , SU Z X . Linearized alternating direction method with adaptive penalty for low-rank representation [C]// Proceedings of the 24th International Conference on Neural Information Processing Systems . New York : ACM , 2011 : 612 - 620 .
TOH K C , YUN S . An accelerated proximal gradient algorithm for nuclear norm regularized linear least squares problems [J]. Pacific Journal of Optimization , 2010 , 6 ( 3 ): 615 - 640 .
CAI J F , CANDÈS E J , SHEN Z W . A singular value thresholding algorithm for matrix completion [J]. SIAM Journal on Optimization , 2010 , 20 ( 4 ): 1956 - 1982 .
GU S H , XIE Q , MENG D Y , et al . Weighted nuclear norm minimization and its applications to low level vision [J]. International Journal of Computer Vision , 2017 , 121 ( 2 ): 183 - 208 .
OH T H , TAI Y W , BAZIN J C , et al . Partial sum minimization of singular values in robust PCA: Algorithm and applications [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2016 , 38 ( 4 ): 744 - 758 .
RODRÍGUEZ P , WOHLBERG B . Fast principal component pursuit via alternating minimization [C]// 2013 IEEE International Conference on Image Processing . Melbourne : IEEE , 2013 : 69 - 73 .
WEI K , CAI J F , CHAN T F , et al . Guarantees of Riemannian optimization for low rank matrix completion [J]. Inverse Problems & Imaging , 2020 , 14 ( 2 ): 233 - 265 .
NGUYEN L T , KIM J , KIM S , et al . Localization of IoT networks via low-rank matrix completion [J]. IEEE Transactions on Communications , 2019 , 67 ( 8 ): 5833 - 5847 .
VANDEREYCKEN B . Low-rank matrix completion by Riemannian optimization [J]. SIAM Journal on Optimization , 2013 , 23 ( 2 ): 1214 - 1236 .
XUE Z C , DONG J , ZHAO Y X , et al . Low-rank and sparse matrix decomposition via the truncated nuclear norm and a sparse regularizer [J]. The Visual Computer , 2019 , 35 ( 11 ): 1549 - 1566 .
XU Z Q , LU Y , WU J X , et al . Adaptive weighted robust principal component analysis [C]// 2020 15th IEEE Conference on Industrial Electronics and Applications . Kristiansand : IEEE , 2020 : 19 - 24 .
GALLO O , GELFANDZ N , CHEN W C , et al . Artifact-free high dynamic range imaging [C]// 2009 IEEE International Conference on Computational Photography . San Francisco : IEEE , 2009 : 1 - 7 .
DONG J , XUE Z C , WANG W W . Robust PCA using nonconvex rank approximation and sparse regularizer [J]. Circuits, Systems, and Signal Processing , 2020 , 39 ( 6 ): 3086 - 3104 .
LEE K C , HO J , KRIEGMAN D J . Acquiring linear subspaces for face recognition under variable lighting [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2005 , 27 ( 5 ): 684 - 698 .
PENG Y G , GANESH A , WRIGHT J , et al . RASL: Robust alignment by sparse and low-rank decomposition for linearly correlated images [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2012 , 34 ( 11 ): 2233 - 2246 .
ZHANG X Q , WANG D , ZHOU Z Y , et al . Robust low-rank tensor recovery with rectification and alignment [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2021 , 43 ( 1 ): 238 - 255 .
0
浏览量
12
下载量
1
CSCD
- 网络PDF下载
- Meta-XML下载
- BibTeX下载
- Endnote下载
- RefMan 下载
- NoteExpress下载
- NoteFirst下载
关联资源
相关文章
相关作者
相关机构
- 技术支持由北京北大方正电子有限公司提供 京ICP备09064830号-19
京公网安备11010802024621 - 本系统建议在Chrome、 IE9+ 以上版本浏览器阅读本站内容，360浏览器请切换至极速模式
- Cookies帮助我们提供服务并提供个性化体验。使用本网站，即表示您同意我们使用Cookies