Underwater Image Enhancement Method Combining Detail Enhancement and Exposure Fusion
Jie Yang, Xiaokui Ren · 2022 IEEE 5th International Conference on Information Systems and Computer Aided Education (ICISCAE) · 2022
Aiming at the problems of color distortion, detail loss and low contrast of underwater images, an underwater image enhancement method combining detail enhancement and exposure fusion is proposed. Firstly, the color balance algorithm of RGB three-channel histogram quantization is used to balance image color and improve contrast. Secondly, images with different exposure degrees are obtained by adjusting the gamma value. In order to solve the problem of detail loss caused by noise amplification in the exposure process, homomorphic filtering is performed on the images after contrast-limited adaptive histogram equalization (CLAHE). Finally, the final result image is constructed using multi-scale fusion technology and combined with the weights defined by contrast, exposure and saturation. Experimental results show that the proposed method can effectively correct blue-green bias, improve image contrast and enrich texture details. The UCIQE is 16.94% higher than the average value of the comparison method.