Parametric indices of fuzziness for automated image enhancement

This paper presents an automated fuzziness-driven algorithm for image enhancement. A class of parametric indices of fuzziness is introduced, which serves as the optimization criterion of the contrast-enhancement procedure. We show that the parametric class of indices constitutes a one-parameter generalization of the linear index of fuzziness of a set. The modification of the membership values of image pixels in the fuzzy plane is performed by finding the optimal S-function, which maximizes the parametric index of fuzziness (PIF). The first-order fuzzy moment of the image is used for tuning the PIF. Experimental results demonstrate the efficiency of the proposed framework in enhancing even highly low-contrasted images and also its ability to improve existing contrast-enhancing algorithms.