Research On Adaptive Color Correction And Sharpening For Underwater Images

In order to develop Marine strategy and explore underwater resources,underwater image plays a very important role as the carrier of Marine information.Due to the complexity of the underwater imaging environment and the bad illumination conditions,the image taken by the current image acquisition equipment in the underwater environment has some problems,such as reduced quality,blurred scene feature information,low contrast,gray concentration,color distortion,etc.,which greatly limits their application in the actual scene.Aiming at these problems,this paper proposes two adaptive color correction and sharpening methods.The specific research contents are as follows:1.Aiming at the problem of deviation and concentration of histogram distribution of underwater image,this paper proposes an adaptive color correction method for underwater image.Firstly,the compensation and attenuation of R,G and B channels of underwater image are carried out to solve the deviation problem.Then,the histogram stretching based on Gaussian model is carried out to solve the problem of concentration.Experiments show that this method can effectively solve the color distortion problem of underwater images.2.Aiming at the inaccuracy of estimation of underwater ambient light in traditional image restoration methods,an underwater image restoration method based on physical model is proposed.The method estimates the ambient light according to the gradient change of the image and avoids the influence of some light areas.Then,the calculation formula of the transmission image is improved to improve the accuracy of estimation of the transmission image of low illumination underwater,and the transmission image is linear stretched by piecewise affine transformation.The experiment shows that this method can achieve good image restoration effect,and the sharpness of the restored image is improved obviously.3.In addition to the underwater image restoration method based on physical model,this paper also proposes an underwater image enhancement method based on multi-path fusion,which includes two multi-scale fusion paths.The first multi-scale fusion path uses the multi-exposure fusion method.This method can show more detail in the deep areas in the middle and distant areas,but at the same time makes the close areas too dark.To solve this problem,a second multi-scale fusion path is introduced,which uses a series of CLAHE enhancement results with different clipping values for multi-scale fusion to enhance the global details.Finally,the results of the two multi-scale fusion paths were linearly fused to obtain the final fusion result.Experimental results show that this method has a good effect on underwater image enhancement.4.The UIEB data set was used for the experiment,and the method in this paper is evaluated from subjective and objective perspectives,and compared with five mainstream underwater image processing methods.In addition,feature point matching,target recognition and saliency detection are also carried out to illustrate the superiority and practicability of the proposed method.