The study of realistic image rendition algorithms based on human visual system

Lily Lv*, Kun Gao, Guoqiang Ni, Manjun Xiao

*Corresponding author for this work

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

1 Citation (Scopus)

Abstract

Realistic image rendition is to reproduce the human perception of natural scenes, where the "realistic" means providing machine vision with the ideal images according to human visual system. The main characteristics of realistic image rendition involve color constancy, lightness constancy, image enhancement and high dynamic range compression. In this paper, the developments on realistic image rendition are investigated, human visual system structure and characteristics are described in detail and some excellent algorithms are studied and discussed, especially the classical Retinex series algorithms and tone mapping algorithms. In the end, the performances of these algorithms are simulated and shown.

Original languageEnglish
Title of host publication2010 Symposium on Photonics and Optoelectronic, SOPO 2010 - Proceedings
DOIs
Publication statusPublished - 2010
EventInternational Symposium on Photonics and Optoelectronics, SOPO 2010 - Chengdu, China
Duration: 19 Jun 201021 Jun 2010

Publication series

Name2010 Symposium on Photonics and Optoelectronic, SOPO 2010 - Proceedings

Conference

ConferenceInternational Symposium on Photonics and Optoelectronics, SOPO 2010
Country/TerritoryChina
CityChengdu
Period19/06/1021/06/10

Keywords

  • Color constancy
  • Dynamic range compression
  • Human vision system
  • Realistic image rendition
  • Retinex

Fingerprint

Dive into the research topics of 'The study of realistic image rendition algorithms based on human visual system'. Together they form a unique fingerprint.

Cite this