Histogram equalization (HE) is a widely used contrast enhancement (CE) method in image processing applications. The algorithm can be easily implemented; however, it tends to transform the average brightness of an image toward the middle of the gray scale. In addition, unpleasant artifacts often appear in the enhanced images. In order to overcome these drawbacks, various HE-based methods which aim at specific issues were proposed. Some of them might overlook the problems inherent in the implementations of histogram equalization and histogram specification (HS). This paper presents a simple histogram modification scheme to solve those problems according to the characteristic of implementation. Two boundary values of the support of histogram are found and set to corresponding values, respectively. The probability density function of an image is then recomputed and the updated mapping function is used to perform histogram equalization. Experimental results show that the proposed approach can effectively improve the quality of images enhanced by histogram equalization and specification methods, and even histogram redistribution methods such as gray-level grouping (GLG).
