Image Contrast Enhancement by Stepwise Histogram Specification

We propose an algorithm for enhancing the contrast of grayscale images by stepwise histogram specification, which produces a contrast-enhanced image of which the histogram is stepwise piecewise constant, and has no gap between neighboring intensities, that is, no 0-valued bin exists in the histogram. As a result, false contouring effect caused by the intensity gap can be avoided successfully. In order to evaluate the performance of contrast enhancement methods, we also propose an image quality measure, which incorporates two conventional image quality measures into a single one. Experimental results show that the proposed algorithm outperforms two conventional histogram equalization methods in terms of the image quality measure.