Tone mapping for high dynamic range image using a probabilistic model

Abstract

In this paper, a probabilistic model is proposed for high dynamic image's tone reproduction. This novel method learns a distribution for local pixel energy of the tone. With the constraint of the gradient variation on the HDR image, an energy distribution is set up based on the similarity between the gradient variation on the HDR and the LDR image. The probabilistic framework for the tone mapping operation is formulated into an energy minimization process by a Maximum A posteriori (MAP) deduction. It turns out that, the proposed method generates LDR image with more visual information than the previous ones. Experimental results show that this approach is convincible and competitive, which can be applied in areas like advanced image editing, displayer development, etc. © by Institute of Software, the Chinese Academy of Sciences. All rights reserved.