Underwater Image Restoration Algorithm Based on Least Squares Estimation

In order to improve the underwater image quality which is degraded by noise and scattering,this paper proposes an underwater image restoration algorithm based on least squares estimation.This work establishes an underwater image degradation model by Beer-Lambert law.The variance of additive white Gaussian noise is estimated by making an analysis on local statistical characteristics of homogeneous patch.This work derives an image filter method based on least squares estimation which is used to restore the original image.The enhanced underwater image is obtained by applying Gamma correction to stretch the light layer component which is decomposed by Retinex model.The proposed algorithm is validated in both aspects of visual perception and objective assessment.Experimental results show that the proposed algorithm is capable to restrain the blurred effect.The quality of underwater restored image is obviously enhanced in terms of color,contrast,detail and clarity.