Research On The Quality Perception And Restoration Of Underwater Turbulence Image

Compared with the situations in the air,underwater imaging poses significant challenge at the extended range.Even in the most clear marine waters,the best visibility range is only tens of meters,while under optimal conditions,the visibility in the air is in the order of miles.The significant reduction in range is caused by the absorption and scattering of the medium,which is the water itself and its internal components,such as particles of various sources,including algal cells,debris,deposits,plankton,and even bubbles near the surface.In addition to the full study of the influence of particle backscatter,due to changes in temperature and salinity in the small scale,underwater optical transmission may be affected by the changes of refractive index.Ocean turbulence leads to the refractive index change can result in underwater optical target fuzzy,especially in the high frequency part decreases,thus reduces the details.Firstly,the influence of underwater turbulence on the optical image is analyzed.In view of the current situation of underwater turbulence fuzzy research,based on the atmospheric turbulence model,the Kolmogorov power spectrum is improved.Considering the influence of temperature and salinity on the wave structure function and the refractive power spectrum,the underwater turbulence model is established.The underwater turbulence fuzzy image database is established by simulating the turbulence fuzzy simulation of underwater clear image.Secondly,based on the established underwater subjective turbulence fuzzy database,an objective underwater fuzzy image distortion evaluation method based on support vector machine is proposed.A number of image statistical features are extracted to reflect the degree of image blur.Using the learning ability of support vector machine,a series of features are extracted and the quality of the image is predicted.Finally,a restoration algorithm based on Wiener filter and a restoration algorithm based on image fusion are proposed for the phenomenon of underwater image blur caused by turbulence.The validity of the algorithm is proved by the experimental results.