An entropy-preserving histogram modification algorithm for image contrast enhancement

Abstract

Contrast enhancement is one of the most crucial image processing steps for image quality control. Histogram equalization (HE) is a commonly used contrast-enhancement algorithm that stretches the image intensity histogram to enhance contrast indirectly. Because of its simplicity and efficiency, numerous contrast-enhancement methods based on HE have been proposed in recent years. However, although these methods usually focus on improving contrast, other features of the original image are not well-preserved. In this paper, we propose a new HE-based algorithm that enhances image contrast based on an assumption of maximum entropy to maintain other features of image quality. The experimental results verify that our proposed algorithm is optimal for generating enhanced images, according to both quantitative estimation and qualitative human visual inspection.