Abstract: Due to the absorption and scattering of light when propagating underwater,underwater images often exhibit color shift,low contrast,blurry and uneven illumination.According to the imaging model of underwater image,many images from the seabed are often degraded severely,and these low-quality images cannot fully provide the ocean scene information,which is difficult to meet the practical application requirements.Therefore,this paper proposes an underwater image enhancement method based on color correction and deblurring.The proposed method effectively combines the two stages of both color correction and deblurring.During the color correction stage,the contrast of each original image is firstly stretched.After the contrast stretching is completed,some images may be overstretched or understretched.According to the gray world prior,after the contrast stretching,the gamma correction is further used to optimize and adjust the contrast and color of these images,so that the sum of the gray values from the R,G,and B channels of each image tends to be equal.Then,in the deblurring stage,it utilizes the dark channel prior to deblur the color-corrected image for the final enhanced image.Experimental results show that the proposed method has a good overall recovery effect,it can effectively restore image information,and obtain good enhancement performance in both subjective and objective evaluation.In addition,the proposed method can be used as a pre-processing step for computer vision tasks such as underwater image classification,and it can improve the classification accuracy of underwater image set by about 16% in our experiment.

Key words: Color correction, Contrast stretching, Deblurring, Gray world, Image classification, Underwater image