Evaluation of GIMP Retinex Filtering of Images by Means of the Shen++ Max Shannon Entropy Finder

Retinex filters are used for processing and enhancing images, obtaining very good results when the images have a degraded contrast, such as those recording foggy or nighttime scenes. Among the several existing Retinex filters, one freely available on the web is that provided by GIMP, the GNU Image Manipulation Program. This software has some parameters that can be adjusted to optimize the output image. Here we investigate a criterion for the optimization of filtering parameters, based on the maximization of image entropy, evaluated using the Shannon formalism. For this task a software, the Shen++ Max Shannon Entropy Finder, is proposed. The software is working on a bulk set of images generated after processing the original image with several different values of the filtering parameters, giving the filtered images having the largest and the smallest Shannon entropy.