A Constrained Algorithm Based NMFα for Image Representation

Nonnegative matrix factorization (NMF) is a useful tool in learning a basic representation of image data. However, its performance and applicability in real scenarios are limited because of the lack of image information. In this paper, we propose a constrained matrix decomposition algorithm for imag...

Full description

Saved in:
Bibliographic Details
Main Authors: Chenxue Yang, Tao Li, Mao Ye, Zijian Liu, Jiao Bao
Format: Article
Language:English
Published: Wiley 2014-01-01
Series:Discrete Dynamics in Nature and Society
Online Access:http://dx.doi.org/10.1155/2014/179129
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850165672179400704
author Chenxue Yang
Tao Li
Mao Ye
Zijian Liu
Jiao Bao
author_facet Chenxue Yang
Tao Li
Mao Ye
Zijian Liu
Jiao Bao
author_sort Chenxue Yang
collection DOAJ
description Nonnegative matrix factorization (NMF) is a useful tool in learning a basic representation of image data. However, its performance and applicability in real scenarios are limited because of the lack of image information. In this paper, we propose a constrained matrix decomposition algorithm for image representation which contains parameters associated with the characteristics of image data sets. Particularly, we impose label information as additional hard constraints to the α-divergence-NMF unsupervised learning algorithm. The resulted algorithm is derived by using Karush-Kuhn-Tucker (KKT) conditions as well as the projected gradient and its monotonic local convergence is proved by using auxiliary functions. In addition, we provide a method to select the parameters to our semisupervised matrix decomposition algorithm in the experiment. Compared with the state-of-the-art approaches, our method with the parameters has the best classification accuracy on three image data sets.
format Article
id doaj-art-ee808fb17d664c2887c2ebba7c1c62a1
institution OA Journals
issn 1026-0226
1607-887X
language English
publishDate 2014-01-01
publisher Wiley
record_format Article
series Discrete Dynamics in Nature and Society
spelling doaj-art-ee808fb17d664c2887c2ebba7c1c62a12025-08-20T02:21:41ZengWileyDiscrete Dynamics in Nature and Society1026-02261607-887X2014-01-01201410.1155/2014/179129179129A Constrained Algorithm Based NMFα for Image RepresentationChenxue Yang0Tao Li1Mao Ye2Zijian Liu3Jiao Bao4School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Science, Chongqing Jiaotong University, Chongqing 400074, ChinaSchool of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaNonnegative matrix factorization (NMF) is a useful tool in learning a basic representation of image data. However, its performance and applicability in real scenarios are limited because of the lack of image information. In this paper, we propose a constrained matrix decomposition algorithm for image representation which contains parameters associated with the characteristics of image data sets. Particularly, we impose label information as additional hard constraints to the α-divergence-NMF unsupervised learning algorithm. The resulted algorithm is derived by using Karush-Kuhn-Tucker (KKT) conditions as well as the projected gradient and its monotonic local convergence is proved by using auxiliary functions. In addition, we provide a method to select the parameters to our semisupervised matrix decomposition algorithm in the experiment. Compared with the state-of-the-art approaches, our method with the parameters has the best classification accuracy on three image data sets.http://dx.doi.org/10.1155/2014/179129
spellingShingle Chenxue Yang
Tao Li
Mao Ye
Zijian Liu
Jiao Bao
A Constrained Algorithm Based NMFα for Image Representation
Discrete Dynamics in Nature and Society
title A Constrained Algorithm Based NMFα for Image Representation
title_full A Constrained Algorithm Based NMFα for Image Representation
title_fullStr A Constrained Algorithm Based NMFα for Image Representation
title_full_unstemmed A Constrained Algorithm Based NMFα for Image Representation
title_short A Constrained Algorithm Based NMFα for Image Representation
title_sort constrained algorithm based nmfα for image representation
url http://dx.doi.org/10.1155/2014/179129
work_keys_str_mv AT chenxueyang aconstrainedalgorithmbasednmfaforimagerepresentation
AT taoli aconstrainedalgorithmbasednmfaforimagerepresentation
AT maoye aconstrainedalgorithmbasednmfaforimagerepresentation
AT zijianliu aconstrainedalgorithmbasednmfaforimagerepresentation
AT jiaobao aconstrainedalgorithmbasednmfaforimagerepresentation
AT chenxueyang constrainedalgorithmbasednmfaforimagerepresentation
AT taoli constrainedalgorithmbasednmfaforimagerepresentation
AT maoye constrainedalgorithmbasednmfaforimagerepresentation
AT zijianliu constrainedalgorithmbasednmfaforimagerepresentation
AT jiaobao constrainedalgorithmbasednmfaforimagerepresentation