A Channel Modified Underwater Image Enhancement Algorithm
Guixin Sun, Wenbin Yu · 2022 IEEE International Conference on Unmanned Systems (ICUS) · 2022
The influence of the medium and particles in the water leads to the attenuation and scattering of light wave propagation. The underwater image will appear blur and color deviation in the imaging process, which makes the underwater target recognition, target tracking, feature extraction and other applications difficult. In view of the above problems, this paper proposes a channel modified underwater image enhancement algorithm. First, the underwater image with color deviation is modified and equalized, a mapping is made by using the intensity center of the histogram. The mapped three channel information is fused into the adaptive histogram equalization with limited contrast, which improves the color deviation and insufficient contrast of the image. Second, the image is deblurred by a priori algorithm of dark channel. Experiments on underwater image data sets show that the proposed method is more effective than existing algorithms in dealing with the degradation of underwater image imaging, and achieves better image quality indicators.