Image contrast enhancement via entropy production

Abstract

In this paper a novel approach for image contrast enhancement is proposed. Employing a thermodynamical model to define local information content, the method derives a measure of local contrast which takes into account spatial structure across multiple scales. From this measure a local contrast map is computed, which is then applied to the image, giving rise to a selective enhancement of the original image. Here some applications to medical images will be presented as well as a comparison with other methods.