文献类型：期刊文献

英文题名：Joint Optimal Transport With Convex Regularization for Robust Image Classification

作者：Qian, Jianjun[1,2,3];Wong, Wai Keung[3];Zhang, Hengmin[4,5];Xie, Jin[1,2];Yang, Jian[1,2]

机构：[1]Nanjing Univ Sci & Technol, PCA Lab, Key Lab Intelligent Percept & Syst High Dimens In, Minist Educ, Nanjing 210094, Peoples R China;[2]Nanjing Univ Sci & Technol, Sch Comp Sci & Engn, Jiangsu Key Lab Image & Video Understanding Socia, Nanjing 210094, Peoples R China;[3]Hong Kong Polytech Univ, Inst Text & Clothing, Hong Kong, Peoples R China;[4]East China Univ Sci & Technol, Key Lab Adv Control & Optimizat Chem Proc, Minist Educ, Shanghai 200237, Peoples R China;[5]Tongji Univ, Shanghai Inst Intelligent Sci & Technol, Shanghai 200092, Peoples R China

年份：2022

卷号：52

期号：3

起止页码：1553

外文期刊名：IEEE TRANSACTIONS ON CYBERNETICS

收录：;EI(收录号：20221511960163);WOS:【SCI-EXPANDED(收录号:WOS:000787531500006)】；

基金：This work was supported in part by the National Science Fund of China under Grant 61876083, Grant U1713208, Grant 61906067, and Grant 61876084, in part by the Research Grant of Hong Kong Scholars Program, in part by the Hong Kong Polytechnic University under Project YZ2K, and in part by the China Postdoctoral Science Foundation under Grant 2019M651415.

语种：英文

外文关键词：Image classification; nuclear norm; optimal transport (OT); robust regression

摘要：The critical step of learning the robust regression model from high-dimensional visual data is how to characterize the error term. The existing methods mainly employ the nuclear norm to describe the error term, which are robust against structure noises (e.g., illumination changes and occlusions). Although the nuclear norm can describe the structure property of the error term, global distribution information is ignored in most of these methods. It is known that optimal transport (OT) is a robust distribution metric scheme due to that it can handle correspondences between different elements in the two distributions. Leveraging this property, this article presents a novel robust regression scheme by integrating OT with convex regularization. The OT-based regression with L-2 norm regularization (OTR) is first proposed to perform image classification. The alternating direction method of multipliers is developed to handle the model. To further address the occlusion problem in image classification, the extended OTR (EOTR) model is then presented by integrating the nuclear norm error term with an OTR model. In addition, we apply the alternating direction method of multipliers with Gaussian back substitution to solve EOTR and also provide the complexity and convergence analysis of our algorithms. Experiments were conducted on five benchmark datasets, including illumination changes and various occlusions. The experimental results demonstrate the performance of our robust regression model on biometric image classification against several state-of-the-art regression-based classification methods.