Nature image feature extraction using several sparse variants of non-negative matrix factorization algorithm

Authors:
Li Shang;Yan Zhou;Jie Chen;Wen-jun Huai
Affiliations:
Department of Electronic Information Engineering, Suzhou Vocational University, Suzhou, Jiangsu, China,Department of Automation, University of Science and Technology of China, Hefei, Anhui, China;Department of Electronic Information Engineering, Suzhou Vocational University, Suzhou, Jiangsu, China;Department of Electronic Information Engineering, Suzhou Vocational University, Suzhou, Jiangsu, China;Department of Electronic Information Engineering, Suzhou Vocational University, Suzhou, Jiangsu, China
Venue:
ISNN'12 Proceedings of the 9th international conference on Advances in Neural Networks - Volume Part II
Year:
2012

Citing 3
Cited 0

Non-negative Matrix Factorization with Sparseness Constraints

The Journal of Machine Learning Research
Learning Sparse Representations by Non-Negative Matrix Factorization and Sequential Cone Programming

The Journal of Machine Learning Research
Sparse nonnegative matrix factorization applied to microarray data sets

ICA'06 Proceedings of the 6th international conference on Independent Component Analysis and Blind Signal Separation

Quantified Score

Hi-index	0.00

Visualization

Abstract

Non-negative matrix factorization (NMF) is an efficient local feature extraction algorithm of natural images. To extract well features of natural images, some sparse variants of NMF, such as sparse NMF (SNMF), local NMF (LNMF), and NMF with sparseness constraints (NMFSC), have been explored. Here, used face images and palmprint images as test images, and considered different number of feature basis dimension, the validity of feature extraction using SNMF, LNMF and NMFSC is testified. Experimental results demonstrate that the level of feature extraction of LNMF is the best, and that of NMFSC is the worse, which also provides some guidance to use different NMF based algorithm in image processing task, and our task in this paper behave certain theory research meaning and application in practice.