Underwater image enhancement based on DCP and depth transmission map

摘要

Seeing that the light in the water is affected by absorption and scattering, underwater image will suffer degradation including low contrast, low visibility and color deviation. Therefore, the key issue of underwater image enhancement is to improve the visibility and the contrast of underwater images. In this paper, we proposed an underwater image dehazing algorithm combining three main steps of homomorphic filtering, double transmission map and dual-image wavelet fusion. First at all, we removed the color deviation in the underwater image by homomorphic filtering. Then, we obtained the enhanced image by depth map which calculate the difference between the light and dark channels. Finally, the dual-image wavelet fusion technique is used to combine the enhanced image obtained by the depth map with the enhanced image obtained by the dark channel. In addition, we obtained the contrast enhanced image which use Contrast-Limited Adaptive Histogram Equalization (CLAHE) method. Through simulation experiments, the proposed method has better visual effects and better effect on entropy, average gradient and underwater color image quality evaluation (UCIQE) compared with other popular methods.