IMAGE ENHANCEMENT USING EVOLUTIONARY ALGORITHM

Abstract

A new approach is presented for the enhancement of color images using the fuzzy logic technique. Here we propose ‘Fork point’ a measure of exposure by which we can divide the gray levels of the image in over and under exposed region. Image of different kind can have both underexposed and overexposed region. The enhancement of underexposed and overexposed regions is the more attentive part of this paper. The new transformation operator is proposed to modify the fuzzy membership values of underexposed region of an image. For overexposed region, a rectangular hyperbolic function is used while for underexposed region a non linear function is applied in which order of non linearity is depended upon membership function. The shape and range of these functions are controlled by the parameters involved and these parameters are optimized using the bacterial foraging optimization algorithm. In terms of image enhancement we are considering only saturation and intensity parts. We are excluding to hue parts because hue contains the original colour composition, and we should preserve it for proper colour information about image. This approach is applicable to a degraded image of mixed type. On comparison, the proposed transforms yield better results than existing transforms, used in [12] for overexposed region.