Image contrast enhancement by constrained variational histogram equalization

Abstract

Histogram equalization is a simple and effective method for image contrast enhancement as it can automatically define the enhancement transformation function based on the image content. However, it tends to change the global appearance of the processed image by shifting its mean brightness toward the middle of the graylevel range, thus making it unsuitable for some applications, such as consumer electronics. In this paper we propose a new method referred as Constrained Variational Histogram Equalization, which builds on the variational formulation of histogram equalization to automatically define an optimal transformation function that would compromise between contrast enhancement and the preservation of the image mean brightness. Experimental evaluation proved this new method to outperform the classical histogram equalization method.