Skip to main navigation Skip to search Skip to main content

Multiple degree total variation (MDTV) regularization for image restoration

  • School of Electronics and Information Engineering, Harbin Institute of Technology
  • University of Iowa

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We introduce a novel image regularization termed as multiple degree total variation (MDTV). This type of regularization combines the first and second degree directional derivatives, thus providing a good balance between preservation of edges and region smoothness. In order to solve the resulting optimization problem, we propose an iteratively reweighted majorize minimize algorithm. We demonstrate the utility of the MDTV regularization in the context of image denoising and compressed sensing image reconstruction. We compare the proposed method with the standard TV, and the state of the art higher degree methods, including higher degree total variation (HDTV) and total generalized variation (TGV) based schemes. Numerical results indicate that the MDTV penalty provides improved image recovery performance.

Original languageEnglish
Title of host publication2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings
PublisherIEEE Computer Society
Pages1958-1962
Number of pages5
ISBN (Electronic)9781467399616
DOIs
StatePublished - 3 Aug 2016
Externally publishedYes
Event23rd IEEE International Conference on Image Processing, ICIP 2016 - Phoenix, United States
Duration: 25 Sep 201628 Sep 2016

Publication series

NameProceedings - International Conference on Image Processing, ICIP
Volume2016-August
ISSN (Print)1522-4880

Conference

Conference23rd IEEE International Conference on Image Processing, ICIP 2016
Country/TerritoryUnited States
CityPhoenix
Period25/09/1628/09/16

Keywords

  • Compressed sensing
  • Majorize minimize
  • Multiple degree total variation (MDTV)

Fingerprint

Dive into the research topics of 'Multiple degree total variation (MDTV) regularization for image restoration'. Together they form a unique fingerprint.

Cite this