Research On The Method And Key Technology Of Underwater Optical Image Definition

With the gradual declining of land resources,the exploitation and utilization of marine resources,along with marine environmental protection,have been elevated onto the pride of place in this field,and underwater detection technology has become an important supporting technology and research field.By contrast,the underwater optical image has become the most effective method for underwater close-range detection due to its rapid imaging,high resolu-tion,rich texture and detailed information and color information.However,the difference between underwater environment and land environment makes the underwater optical image vulnerable to the following factors in the imaging process:First,the random noise caused by the irregular motion of electronic components and photons in the imaging process;Second,the image blurring and atomizing effect caused by the absorption and scattering of irradiated light by water body;Third,the water body possessing spectral regions different absorption of light in different spectral regions,which leads to the color distortion of underwater images.These factors usually make the underwater optical image blurred,low contrast,obvious noise,color distortion and other problems Therefore,it is of great necessity to improve the quality of underwater optical images through definition process.The purpose of underwater optical image definition is designed to enhance the detail,contrast and visualization of the images.The definition of underwater images needs to focus on the following three issues:How to effectively suppress the signal noise caused by the ir-regular motion of suspended particles and photons in water;How to effectively recover the optical signal attenuation caused by different light wavelengths and keep the image color un-distorted;How to further enhance the visual contrast of underwater optical images and im-prove the image quality of underwater optical images?The above three questions are the key and difficult problems in underwater optical image definition.To this end,this paper conducts theoretical and experimental research on the definition method of underwater optical images by focusing on the three key links of underwater optical image denoising,restoration and en-hancement.The main work includes the following aspects:1.Aiming at the problem of underwater optical image vision blur and low contrast caused by noise,a denoising method of least squares estimation based on homomorphic sub-block is proposed.According to the characteristics of underwater image,an underwater image degradation model is constructed by using the Beer-Lambert law.The degraded image is converted into brightness image and its saturation.The original brightness image is obtained by least square estimation according to the brightness image and its homomorphic sub-block noise variance.The underwater color restoration image with noise elimination can be obtained by the model of constant saturation.2.To solve the problem of low definition and color distortion of underwater optical im-ages caused by inconsistent attenuation of light signals of different wavelengths in water,a method of brightness restoration and color correction of underwater image based on homo-morphic filtering is proposed.The standard residual energy ratio is used to replace the trans-mission ratio to construct a de-noised underwater degraded image model,then a color con-stancy model is established by first order McLaughlin series expansion.he degraded image is converted into brightness image and its background brightness image,then the degraded im-age model is used to get the restored brightness image through homomorphic filtering,and then the restored color image is obtained according to the color constancy model.The de-graded image is converted into brightness image and its background brightness image,then the degraded image model is used to get the restored brightness image through homomorphic filtering,and then the restored color image is obtained according to the color constancy mod-el.3.An underwater optical image enhancement method based on iterative histogram is proposed to solve the problem of low contrast and fuzzy detail caused by underwater light at-tenuation and scattering.In HSV space,Retinex model is adopted to extract the light layer image and detail layer image from the restoration image V channel.The image of detail layer was enhanced by S function transformation,and the image of light layer was enhanced by it-erative histogram equalization.The enhanced images of detail layer and light layer were merged and converted to RGB space to obtain the enhanced color image.For the sharpening method of underwater images in this paper,mathematical expression is given through theoretical analysis,algorithm structure is constructed,algorithm is designed and implemented,and verification experiment is completed.Simulation results of denoising,restoration,enhancement and integrated sharpening of underwater degraded images verify the feasibility and effectiveness of the proposed algorithm in this paper.