An underwater image enhancement method based on brightness channel replacement and double-scale fusion

Abstract

The application provides an underwater image enhancement method based on brightness channel replacement and double-scale fusion, and belongs to the technical field of underwater optical imaging and digital image processing; the application firstly acquires underwater scene information through multi-band imaging, and uses a pre-trained contrast advantage band search network to recommend an optimal detail band; then, a preliminary color image is synthesized based on the three primary color principle, and a U-Shape Transformer deep learning network is used for color restoration; then, the color restoration image is converted to a Lab color space, and a gray image of the optimal detail band is used to replace the brightness channel, so that the color and the detail are fused; finally, a high-quality underwater enhanced image is output through a dark channel prior-based de-scattering enhancement processing. Compared with the existing single-feature underwater image enhancement method, the application algorithm has better effectiveness and robustness, and can significantly improve the visual quality and usability of the underwater image.

Images