Back to articles
Volume: 21 | Article ID: art00043
On the Information Content along Edges in Trichromatic Images
  DOI :  10.2352/CIC.2013.21.1.art00043  Published OnlineJanuary 2013

We introduce a theoretical framework for measuring the information content in the edges extracted from a color image. The main difficulty in estimating the amount of information (differential entropy [1, 2]) in an image signal is to fit an appropriate probability mass function to the trichromatic image data. To estimate the amount of information in the edges extracted from a color image, we first convolve the image with a derivative filter. By fitting a Kotz-Type probability distribution to the convolved image, we then estimate the differential entropy of the edge coefficients as a measure of the uncertainty involved in the edge content of a postreceptoral chromatic image. The proposed estimation of differential entropy provides an efficient means of processing the edge content information under a variety of natural illuminations, which might be further used as a quantitative measure for evaluating color constant image retrieval.

Subject Areas :
Views 8
Downloads 0
 articleview.views 8
 articleview.downloads 0
  Cite this article 

Shahram Peyvandi, Juan L Nieves, Alan Gilchrist, "On the Information Content along Edges in Trichromatic Imagesin Proc. IS&T 21st Color and Imaging Conf.,  2013,  pp 236 - 239,

 Copy citation
  Copyright statement 
Copyright © Society for Imaging Science and Technology 2013
Color and Imaging Conference
color imaging conf
Society of Imaging Science and Technology
7003 Kilworth Lane, Springfield, VA 22151, USA