Skip to main navigation Skip to search Skip to main content

A PDE formalization of retinex theory

  • Jean Michel MOREL*
  • , Ana Belén PETRO
  • , Catalina SBERT
  • *Corresponding author for this work

Research output: Journal PublicationsJournal Article (refereed)peer-review

Abstract

In 1964, Edwin H. Land formulated the Retinex theory, the first attempt to simulate and explain how the human visual system perceives color. His theory and an extension, the "reset Retinex" were further formalized by Land and McCann. Several Retinex algorithms have been developed ever since. These color constancy algorithms modify the RGB values at each pixel to give an estimate of the color sensation without a priori information on the illumination. Unfortunately, the Retinex LandMcCann original algorithm is both complex and not fully specified. Indeed, this algorithm computes at each pixel an average of a very large set of paths on the image. For this reason, Retinex has received several interpretations and implementations which, among other aims, attempt to tune down its excessive complexity. In this paper, it is proved that if the paths are assumed to be symmetric random walks, the Retinex solutions satisfy a discrete Poisson equation. This formalization yields an exact and fast implementation using only two FFTs. Several experiments on color images illustrate the effectiveness of the Retinex original theory. © 2006 IEEE.
Original languageEnglish
Article number5458027
Pages (from-to)2825-2837
Number of pages13
JournalIEEE Transactions on Image Processing
Volume19
Issue number11
Early online date3 May 2010
DOIs
Publication statusPublished - Nov 2010
Externally publishedYes

Bibliographical note

This work was supported in part by the Centre National d'Etudes Spatiales (MISS project) and by the Office of Naval Research under Grant N00014-97-1-0839.

Keywords

  • Color perception
  • FFT
  • PDE
  • Retinex theory
  • stochastic integral

Fingerprint

Dive into the research topics of 'A PDE formalization of retinex theory'. Together they form a unique fingerprint.

Cite this