上海交通大学学报(英文版) ›› 2017, Vol. 22 ›› Issue (1): 77-081.doi: 10.1007/s12204-017-1803-y

• • 上一篇    下一篇

Weak Correlation Dictionary Construction Method for Sparse Coding

LONG Haixia (龙海霞), ZHUO Li* (卓 力), QU Panling (屈盼玲), ZHANG Jing (张 菁)   

  1. (Signal & Information Processing Laboratory, Beijing University of Technology, Beijing 100124, China)
  • 出版日期:2017-02-28 发布日期:2017-04-04
  • 通讯作者: ZHUO Li* (卓 力) E-mail:zhuoli@bjut.edu.cn

Weak Correlation Dictionary Construction Method for Sparse Coding

LONG Haixia (龙海霞), ZHUO Li* (卓 力), QU Panling (屈盼玲), ZHANG Jing (张 菁)   

  1. (Signal & Information Processing Laboratory, Beijing University of Technology, Beijing 100124, China)
  • Online:2017-02-28 Published:2017-04-04
  • Contact: ZHUO Li* (卓 力) E-mail:zhuoli@bjut.edu.cn

摘要: For sparse coding, the weaker the correlation of dictionary atoms is, the better the representation capacity of dictionary will be. A weak correlation dictionary construction method for sparse coding has been proposed in this paper. Firstly, a new dictionary atom initialization is proposed in which data samples with weak correlation are selected as the initial dictionary atoms in order to effectively reduce the correlation among them. Then, in the process of dictionary learning, the correlation between atoms has been measured by correlation coefficient, and strong correlation atoms have been eliminated and replaced by weak correlation atoms in order to improve the representation capacity of the dictionary. An image classification scheme has been achieved by applying the weak correlation dictionary construction method proposed in this paper. Experimental results show that, the proposed method averagely improves image classification accuracy by more than 2%, compared to sparse coding spatial pyramid matching (ScSPM) and other existing methods for image classification on the datasets of Caltech-101, Scene-15, etc.

关键词: image classification, sparse coding, correlation coefficient, dictionary initialized

Abstract: For sparse coding, the weaker the correlation of dictionary atoms is, the better the representation capacity of dictionary will be. A weak correlation dictionary construction method for sparse coding has been proposed in this paper. Firstly, a new dictionary atom initialization is proposed in which data samples with weak correlation are selected as the initial dictionary atoms in order to effectively reduce the correlation among them. Then, in the process of dictionary learning, the correlation between atoms has been measured by correlation coefficient, and strong correlation atoms have been eliminated and replaced by weak correlation atoms in order to improve the representation capacity of the dictionary. An image classification scheme has been achieved by applying the weak correlation dictionary construction method proposed in this paper. Experimental results show that, the proposed method averagely improves image classification accuracy by more than 2%, compared to sparse coding spatial pyramid matching (ScSPM) and other existing methods for image classification on the datasets of Caltech-101, Scene-15, etc.

Key words: image classification, sparse coding, correlation coefficient, dictionary initialized

中图分类号: