TY - JOUR
T1 - Robust High-Order Manifold Constrained Sparse Principal Component Analysis for Image Representation
AU - Zhou, Nan
AU - Cheng, Hong
AU - Qin, Jing
AU - Du, Yuanhua
AU - Chen, Badong
N1 - Funding Information:
Manuscript received December 23, 2017; revised May 30, 2018; accepted June 29, 2018. Date of publication July 17, 2018; date of current version July 1, 2019. This work was supported in part by the 973 Program under Grants 2015CB351703 and 2015CB351706, in part by the NSFC under Grants U1613223, 6157021026 and 61305033, in part by the National Key R&D Program of China under Grant 2017YFB1302300, and in part by the Fundamental Research Funds for Central Universities under Grant ZYGX2014Z009. This paper was recommended by Associate Editor J. Yuan. (Corresponding author: Hong Cheng.) N. Zhou is with the College of Control Engineering, Chengdu University of Information Technology, Chengdu 610225, China (e-mail: [email protected]).
Publisher Copyright:
© 1991-2012 IEEE.
PY - 2019/7
Y1 - 2019/7
N2 - In order to efficiently utilize the information in the data and eliminate the negative effects of outliers in the principal component analysis (PCA) method, in this paper, we propose a novel robust sparse PCA method based on maximum correntropy criterion (MCC) with high-order manifold constraints called the RHSPCA. Compared with the traditional PCA methods, the proposed RHSPCA has the following benefits: 1) the MCC regression term is more robust to outliers than the MSE-based regression term; 2) thanks to the high-order manifold constraints, the low-dimensional representations can preserve the local relations of the data and greatly improve the clustering and classification performance for image processing tasks; and 3) in order to further counteract the adverse effects of outliers, the MCC-based samples' mean is proposed to better centralize the data. We also propose a new solver based on the half-quadratic technique and accelerated block coordinate update strategy to solve the RHSPCA model. Extensive experimental results show that the proposed method can outperform the state-of-the-art robust PCA methods on a variety of image processing tasks, including reconstruction, clustering, and classification, on outliers contaminated datasets.
AB - In order to efficiently utilize the information in the data and eliminate the negative effects of outliers in the principal component analysis (PCA) method, in this paper, we propose a novel robust sparse PCA method based on maximum correntropy criterion (MCC) with high-order manifold constraints called the RHSPCA. Compared with the traditional PCA methods, the proposed RHSPCA has the following benefits: 1) the MCC regression term is more robust to outliers than the MSE-based regression term; 2) thanks to the high-order manifold constraints, the low-dimensional representations can preserve the local relations of the data and greatly improve the clustering and classification performance for image processing tasks; and 3) in order to further counteract the adverse effects of outliers, the MCC-based samples' mean is proposed to better centralize the data. We also propose a new solver based on the half-quadratic technique and accelerated block coordinate update strategy to solve the RHSPCA model. Extensive experimental results show that the proposed method can outperform the state-of-the-art robust PCA methods on a variety of image processing tasks, including reconstruction, clustering, and classification, on outliers contaminated datasets.
KW - Correntropy
KW - high-order manifold
KW - principal component analysis (PCA)
UR - http://www.scopus.com/inward/record.url?scp=85068470559&partnerID=8YFLogxK
U2 - 10.1109/TCSVT.2018.2856827
DO - 10.1109/TCSVT.2018.2856827
M3 - Journal article
AN - SCOPUS:85068470559
SN - 1051-8215
VL - 29
SP - 1946
EP - 1961
JO - IEEE Transactions on Circuits and Systems for Video Technology
JF - IEEE Transactions on Circuits and Systems for Video Technology
IS - 7
M1 - 8412267
ER -