On the Convex Model of Speckle Reduction

Faming Fang, Yingying Fang, Tieyong Zeng*

*Corresponding author for this work

Research output: Chapter in book/report/conference proceedingConference proceedingpeer-review

Abstract

Speckle reduction is an important issue in image processing realm. In this paper, we propose a novel model for restoring degraded images with multiplicative noise which follows a Nakagami distribution. A general penalty term based on the statistical property of the speckle noise is used to guarantee the convexity of the denoising model. Moreover, to deal with the minimizing problem, a generalized Bermudez-Moreno algorithm is adopted and its convergence is analysed. The experimental results on some images subject to multiplicative noise as well as comparisons to other state-of-the-art methods are also presented. The results can verify that the new model is reasonable.

Original languageEnglish
Title of host publicationImaging, Vision and Learning Based on Optimization and PDEs
Subtitle of host publicationIVLOPDE, Bergen, Norway, August 29 – September 2, 2016
EditorsXue-Cheng Tai, Egil Bae, Marius Lysaker
PublisherSpringer Cham
Pages121-141
Number of pages21
Edition1st
ISBN (Electronic)9783319912745
ISBN (Print)9783319912738
DOIs
Publication statusPublished - 19 Nov 2018
EventInternational conference on Imaging, Vision and Learning Based on Optimization and PDEs, IVLOPDE 2016 - Bergen, Norway
Duration: 29 Aug 20162 Sept 2016

Publication series

NameMathematics and Visualization
ISSN (Print)1612-3786
ISSN (Electronic)2197-666X
NameIVLOPDE: International Conference on Imaging, Vision and Learning based on Optimization and PDEs

Conference

ConferenceInternational conference on Imaging, Vision and Learning Based on Optimization and PDEs, IVLOPDE 2016
Country/TerritoryNorway
CityBergen
Period29/08/162/09/16

Scopus Subject Areas

  • Modelling and Simulation
  • Geometry and Topology
  • Computer Graphics and Computer-Aided Design
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'On the Convex Model of Speckle Reduction'. Together they form a unique fingerprint.

Cite this