A no-reference perceptual image sharpness metric based on a cumulative probability of blur detection

Niranjan D. Narvekar, Lina Karam

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

151 Scopus citations

Abstract

In this paper, a no-reference objective sharpness metric based on a cumulative probability of blur detection is proposed. The metric is evaluated by taking into account the Human Visual System (HVS) response to blur distortions. The perceptual significance of the metric is validated through subjective experiments. It is shown that the proposed metric results in a very good correlation with subjective scores especially for images with varying foreground and background perceived blur qualities. This is accomplished with a significantly lower computational complexity as compared to existing methods that take into account the visual attention information.

Original languageEnglish (US)
Title of host publication2009 International Workshop on Quality of Multimedia Experience, QoMEx 2009
Pages87-91
Number of pages5
DOIs
StatePublished - Nov 11 2009
Event2009 International Workshop on Quality of Multimedia Experience, QoMEx 2009 - San Diego, CA, United States
Duration: Jul 29 2009Jul 31 2009

Publication series

Name2009 International Workshop on Quality of Multimedia Experience, QoMEx 2009

Other

Other2009 International Workshop on Quality of Multimedia Experience, QoMEx 2009
Country/TerritoryUnited States
CitySan Diego, CA
Period7/29/097/31/09

Keywords

  • Human visual system
  • Image quality
  • Just noticeable blur
  • No-reference
  • Objective
  • Perceptual
  • Sharpness metric
  • Visual attention

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Science Applications
  • Software

Fingerprint

Dive into the research topics of 'A no-reference perceptual image sharpness metric based on a cumulative probability of blur detection'. Together they form a unique fingerprint.

Cite this