Multi-domain Value Block Enhancement Algorithm Fused with Genetic Algorithm

Abstract

In order to solve the problems of the low-illuminance image with low contrast and blurred details, this paper design the multi-domain value block enhancement algorithm fused with genetic algorithm. The algorithm first searches for the optimal segmentation threshold of the brightness channel of the input image through genetic algorithm. Then, according to the obtained threshold, the brightness channel is divided into multiple subgraphs with different exposure levels. Next, all subgraphs are assessed by the multi-threshold block enhancement method, and the brightness of each subgraph is adjusted according to the assessment. After that the multi-scale fusion method is used to fuse the details of the input image into the brightness-enhanced image. Finally, the output image with normal brightness and rich details is reconstructed. This paper selects the enhancement algorithm proposed in the past three years for comparison experiments. The results show that the low-illuminance image is enhanced by the algorithm in this paper. Its entropy increases by 66.2%, enhancement by entropy increases by 94.7%, and average gradient increases by 97.2%. The increase of each index of the image enhanced by the algorithm in this paper is greater than that of other comparison methods, which proves that the algorithm in this paper has better performance.