Genetic algorithm based adaptive histogram equalization (GAAHE) technique for medical image enhancement

Abstract

In Magnetic Resonance Imaging (MRI), the poor quality images may not provide the sufficient information for the visual interpretation of the affected locations of human body. So, to improve the image visions and to provide computational support, a novel adaptive image enhancement technique has been proposed in this paper, named as genetic algorithm based adaptive histogram equalization (GAAHE) technique. The proposed framework includes genetic algorithm, histogram sub-division and modified probability density function (PDF). A novel approach of subdivision is applied to the histogram using the exposure threshold and optimal threshold for preserving the brightness and reducing the information loss. To make the proposed technique more adaptive, the threshold parameters are optimized by utilizing the concept of genetic algorithm, guided by the proposed multi-objective fitness function. Then, the PDF of each sub-histogram is modified to enhance the image quality. The experimental results show that, the proposed GAAHE technique performs superior over other existing enhancement techniques.