Blind inverse gamma correction with maximized differential entropy

Abstract

Unwanted nonlinear gamma distortion frequently occurs in a great diversity of images during the procedures of image acquisition, processing, or display. Blind inverse gamma correction, which automatically determines a single proper restoration gamma value for a given image, is of paramount importance to attenuate the distortion. In this paper, a gamma correction method (GCME) is proposed directly from a maximized differential entropy model. Incorporating the change-of-variables rule, GCME adopts a novel computing pipeline to address the entropy decrease barrier of maximum entropy assumption. Surprisingly, the optimization of GCME has a mathematical concise closed-form solution, resulting in efficient implementation and accurate gamma restoration. Tested on variable datasets, GCME could obtain an accurate estimation of a large range of gamma distortion (0.1 to 3.0), outperforming the state-of-the-art methods. Besides, the proposed GCME was applied to three typical applications, including automatic gamma adjustment, natural/medical image contrast enhancement, and fringe projection profilometry image restoration. Furthermore, the GCME is general and can be seamlessly extended to the masked image, multi-channel (colour or spectrum) image or multi-frame video, and free of arbitrary tuning parameter.