Image contrast enhancement with brightness preservation using an optimal gamma and logarithmic approach

Abstract

In this study, a new enhancement framework is proposed for low contrast and dark images where traditional histogram equalisation (HE), gamma and logarithmic transformation are incorporated to achieve a visually pleasing image. Before the operation of HE on the input image, gamma and logarithmic transformation are performed in order to preserve the fine details of the image. A new gamma value of the proposed algorithm helps to restrain histogram spikes to avoid over-enhancement and noise artefacts effect. After that, a novel logarithmic transformation is used to map a narrow range of low-intensity values in the input image to a wider range of output levels. Thus, the dark input values are spread out into the higher intensity values, which improve the overall contrast and brightness of the image. The proposed method is compared with various state-of-the-art techniques. The large dataset has been used to check the feasibility of the technique. The subjective and objective analysis shows that the proposed algorithm outperforms most of the existing contrast-enhancement algorithms and the results are natural-looking, good contrast images with almost no artefacts.