Direct robust non-negative matrix factorization and its application on image processing

Bin Shen, Zhanibek Datbayev, Olzhas Makhambetov

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

In real applications of image processing, we frequently face outliers, which cannot be simply treated as Gaussian noise. Nonnegative Matrix Factorization (NMF) is a popular method in image processing for its good performance and elegant theoretical interpretation, however, traditional NMF is not robust enough to outliers. To robustify NMF algorithm, here we present Direct Robust Nonnegative Matrix Factorization (DRNMF) for image denoising based on the assumptions that the ground truth data is of low rank and the outliers are sparse. This method explictly models the outliers in the data, and the sparsity of the outliers is controlled by L0 norm. The experiments show that DRNMF can accurately localize the outliers, and outperforms traditional NMF in image denoising.

Original languageEnglish
Title of host publication2012 6th International Conference on Application of Information and Communication Technologies, AICT 2012 - Proceedings
DOIs
Publication statusPublished - Dec 1 2012
Event2012 6th International Conference on Application of Information and Communication Technologies, AICT 2012 - Tbilisi, Georgia
Duration: Oct 17 2012Oct 19 2012

Publication series

Name2012 6th International Conference on Application of Information and Communication Technologies, AICT 2012 - Proceedings

Other

Other2012 6th International Conference on Application of Information and Communication Technologies, AICT 2012
CountryGeorgia
CityTbilisi
Period10/17/1210/19/12

Keywords

  • NMF
  • Outlier removal
  • image denoising
  • nonnegative representation
  • robust NMF
  • sparse error

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems

Fingerprint Dive into the research topics of 'Direct robust non-negative matrix factorization and its application on image processing'. Together they form a unique fingerprint.

Cite this