Research On Visibility Enhancement Of Underwater Images

Underwater image is one of the important means of access to marine information,due to the complexity of the underwater environment,light is strongly absorbed and scattered which causes low contrast?color distortion and image detail fuzzy.In regard to these problems,this paper has proposed two different underwater image visibility enhancement schemes.Underwater images restored by dark channel prior method often lead to the overrated transmission and underestimated background light,whereas the process of underwater image formation wasn't taken into account in the image enhancement method based on Retinex theory.Based on the above issues,an effective visibility enhancement scheme for underwater image is proposed.Firstly,a color correction algorithm based on statistical method is adopted to remove the color cast.Then,the adaptive Wiener filter is used to optimize the initial transmission map with the boundary constraints aiming at the shortcomings of the traditional dark channel prior to transmission.Lastly,a weighted L₁-norm regularization model based on the Retinex theory is proposed to enhance the luminance layer.Experimental results demonstrate the effectiveness of the proposed method in restoring the original color of the scene and enhancing image contrast and visibility.Aiming at the shortage of underwater image restoration algorithm based on new optical model,this paper proposes an improved method including visibility restoration and contrast enhancement.Firstly,based on new optical model with no influence of red channel information,this paper manage to put forward an algorithm by setting the threshold value to judge whether to add red channel information to dark channel prior.Then,as for the artificial light,we add saturation prior to dark channel prior which weakens the effect of artificial light.Based on the assumption that blue channel and green channel links than consistent inadequacy,this paper uses the attenuation coefficient estimating the transmission.Finally,the contrast of the restored image is enhanced and the magnified noise is reduced by non-local means denoising.Experimental results show that the improved algorithm has better visual effect than the traditional underwater image restoration algorithm and other stated-of-the-art underwater algorithm.