Using Color Correlation To Improve Restoration of Color Images

Daniel Keren, Anna Gotlib

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


This chapter addresses the problem of restoration of images that have a few channels of information. The method that is suggested by the chapter is to use a probabilistic scheme which is useful for image restoration, and to incorporate an additional term into it, which results in a better correlation among the three color bands in the restored image. Initial results are good; typically, there is a reduction of 30% in the root mean square (RMS) error as compared to the standard restoration carried out separately on each color band.
Original languageEnglish
Title of host publicationInternational Workshop on Image and Signal Processing
EditorsB.G. Mertzios , P. Liatsis
StatePublished - 1996


Dive into the research topics of 'Using Color Correlation To Improve Restoration of Color Images'. Together they form a unique fingerprint.

Cite this