Research On Underwater Image Restoration Based On Polarization Imaging

The increasing demand for the ocean has prompted mankind to continue to increase the exploration and development of the ocean.However,due to the unevenness of the movement of seawater and the irregularity of the movement of suspended particles in water,light is absorbed by suspended particles in seawater when it travels in water.And the occurrence of multiple reflections and strong scattering.Both light absorption and scattering phenomena can cause attenuation of light energy.The absorption of light leads to a low image contrast,and the strong underwater scattering environment will cause the light to deviate from the normal path,increase the background noise,make the image blurred,and increase the difficulty of our research.In order to improve the clarity of underwater images,this article will deeply study the polarization imaging system,based on the principle of polarized light imaging characteristics,study the backscattering noise characteristics of underwater imaging,and establish a complete set of polarization imaging underwater physics Model,and apply the model to the calculation of image enhancement algorithms and restoration algorithms.The image fusion algorithm is to integrate multiple original images to form a new image.The fused image can better express the image details.In this paper,the pyramid-based fusion enhancement algorithm is used to perform hierarchical decomposition of multi-scale image sequences from different angles.It is proposed to apply the weight fusion coefficient to the pyramid decomposition algorithm to obtain difference fusion images of different states,and apply the weight fusion coefficient to the Stock fusion map.Medium,the contrast of the polarized image is enhanced.In order to further improve the estimation accuracy of the image prior information.In the paper,the inclined edge method is used to perform Gaussian fitting calculation on the prior information of the image.The improved PSF value is used as prior knowledge into the total variational(TV)algorithm and the restoration algorithm based on the FOE model.The accuracy of the image reconstruction has a significant effect on the image restoration,reflecting the superiority of the algorithm.