Research on the Method of Color Compensation and Underwater Image Restoration Based on Polarization Characteristics
kaiqiang gong, Dou Hua · 2022 3rd International Conference on Computer Vision, Image and Deep Learning & International Conference on Computer Engineering and Applications (CVIDL & ICCEA) · 2022
The attenuation degree of light in different bands is different when transmitting underwater, which leads to the problems of color distortion and blurred target details in the turbid underwater image. Combined with Monte Carlo simulation and measured experiments, this paper studied the polarization characteristics of light transmission in different wavelength bands underwater. The results show that the longer the wavelength of light, the stronger the ability of polarization retention underwater. Based on this feature, a new method of color compensation of underwater images is proposed. By fusing the polarization image with the intensity image, the exposure of the low-brightness areas in the image is improved, so as to compensate for the areas with serious color distortion in the underwater image. And then the image is deblurred and enhanced by using the dark channel prior principle. Comparative experimental results indicate that the method can effectively compensate for the color distortion of underwater images and significantly improves the visibility of underwater targets.