Hierarchical Graph Augmented Deep Collaborative Dictionary Learning for Classification

被引:28
作者
Gou, Jianping [1 ,2 ]
Yuan, Xia [1 ,2 ]
Du, Lan [3 ]
Xia, Shuyin [4 ]
Yi, Zhang [5 ]
机构
[1] Jiangsu Univ, Sch Comp Sci & Commun Engn, Zhenjiang 212013, Jiangsu, Peoples R China
[2] Jiangsu Univ, Jiangsu Key Lab Secur Technol Ind Cyberspace, Zhenjiang 212013, Jiangsu, Peoples R China
[3] Monash Univ, Fac Informat Technol, Clayton, Vic 3800, Australia
[4] Chongqing Univ Posts & Telecommun, Chongqing Key Lab Computat Intelligence, Chongqing 400065, Peoples R China
[5] Sichuan Univ, Sch Comp Sci, Chengdu 610017, Peoples R China
基金
中国国家自然科学基金;
关键词
Dictionaries; Machine learning; Collaboration; Task analysis; Encoding; Sparse matrices; Testing; Deep dictionary learning; representation learning; graph construction; pattern classification; FACE RECOGNITION; SPARSE;
D O I
10.1109/TITS.2022.3177647
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
Recently, deep dictionary learning (DDL) has aroused attention due to its abilities of learning multiple different dictionaries and extracting multi-level abstract feature representations for samples. It has been applied to many intelligent recognition tasks, such as vehicle detection, traffic sign recognition and driver monitoring. Nevertheless, the off-the-shelf DDL-based methods ignore the essential structural information of data in multi-layer dictionary learning. The learned hierarchical data representations are less discriminative. To address this issue, we develop a new DDL framework, called the hierarchical graph augmented deep collaborative dictionary learning (HGDCDL). Firstly, we propose a new deep collaborative dictionary learning (DCDL) that applies collaborative representation to the deepest-level representation learning. Most importantly, equipped with a simple yet effective hierarchal graph construction mechanism, our HGDCDL uses the structure of data to regularize dictionary learning, and generates more informative dictionaries and discriminative representations at different levels. Extensive experiments show that our HGDCDL performs significantly better than the state-of-the-art shallow and deep representation learning methods for classification.
引用
收藏
页码:25308 / 25322
页数:15
相关论文
共 59 条
[1]   K-SVD: An algorithm for designing overcomplete dictionaries for sparse representation [J].
Aharon, Michal ;
Elad, Michael ;
Bruckstein, Alfred .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2006, 54 (11) :4311-4322
[2]  
[Anonymous], 1996, Tech. Rep. CUCS-006-96
[3]   Graph Anomaly Detection Using Dictionary Learning [J].
Baltoiu, Andra ;
Patrascu, Andrei ;
Irofti, Paul .
IFAC PAPERSONLINE, 2020, 53 (02) :3551-3558
[4]   Learning Deep Architectures for AI [J].
Bengio, Yoshua .
FOUNDATIONS AND TRENDS IN MACHINE LEARNING, 2009, 2 (01) :1-127
[5]   Dictionary learning for clustering on hyperspectral images [J].
Bruton, Joshua ;
Wang, Hairong .
SIGNAL IMAGE AND VIDEO PROCESSING, 2021, 15 (02) :255-261
[6]   Graph-Regularized Discriminative Analysis-Synthesis Dictionary Pair Learning for Image Classification [J].
Chang, Heyou ;
Tang, Hui ;
Zhang, Fanlong ;
Chen, Yang ;
Zheng, Hao .
IEEE ACCESS, 2019, 7 :55398-55406
[7]   GraFT: Graph Filtered Temporal Dictionary Learning for Functional Neural Imaging [J].
Charles, Adam S. ;
Cermak, Nathan ;
Affan, Rifqi O. ;
Scott, Benjamin B. ;
Schiller, Jackie ;
Mishne, Gal .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2022, 31 :3509-3524
[8]   Vehicle Detection in High-Resolution Aerial Images Based on Fast Sparse Representation Classification and Multiorder Feature [J].
Chen, Ziyi ;
Wang, Cheng ;
Luo, Huan ;
Wang, Hanyun ;
Chen, Yiping ;
Wen, Chenglu ;
Yu, Yongtao ;
Cao, Liujuan ;
Li, Jonathan .
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2016, 17 (08) :2296-2309
[9]   Driver Monitoring Using Sparse Representation With Part-Based Temporal Face Descriptors [J].
Chiou, Chien-Yu ;
Wang, Wei-Cheng ;
Lu, Shueh-Chou ;
Huang, Chun-Rong ;
Chung, Pau-Choo ;
Lai, Yun-Yang .
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2020, 21 (01) :346-361
[10]   Joint dictionary and graph learning for unsupervised feature selection [J].
Ding, Dediong ;
Xia, Fei ;
Yang, Xiaogao ;
Tang, Chang .
APPLIED INTELLIGENCE, 2020, 50 (05) :1379-1397