A rapid scene depth estimation model based on underwater light attenuation prior for underwater image restoration

Abstract

Underwater images present blur and color cast, caused by light absorption and scattering in water medium. To restore underwater images through image formation model (IFM), the scene depth map is very important for the estimation of the transmission map and background light intensity. In this paper, we propose a rapid and effective scene depth estimation model based on underwater light attenuation prior (ULAP) for underwater images and train the model coefficients with learning-based supervised linear regression. With the correct depth map, the background light (BL) and transmission maps (TMs) for R-G-B light are easily estimated to recover the true scene radiance under the water. In order to evaluate the superiority of underwater image restoration using our estimated depth map, three assessment metrics demonstrate that our proposed method can enhance perceptual effect with less running time, compared to four state-of-the-art image restoration methods.