Discrimination metric for Bi-Histogram equalization of contrast enhancement

A simple discrimination metric (DMBHE) is developed to avoid the drawbacks of conventional histogram equalization for gray scale images. The proposed technique uses both global and local information to remap the intensity levels that helps to improve the image contrast. The original histogram is divided into sub-histograms with respect to the mean value. Discrimination metrics are used, so that high contrast per pixel between original image and enhanced image is obtained. The simulation results show that the proposed method performed well for microstructure images. Parameters like structural similarity index and contrast per pixel are used to analyze the image quality.