Underwater image restoration based on diffraction bounded optimization algorithm with dark channel prior

The restoration of underwater images is challenging due to the turbid nature of water. The effects of diffraction due to the varying refractive index of water medium, absorption and scattering properties of water and the distance between the image spot to the camera, can lead to geometric distortions and blur. Many algorithms exist so far to restore the quality of underwater images. The effect of turbulence in underwater scenarios areas are challenging for most restoration algorithms. Although the metric gauge for the restoration of the underwater scene is still under evaluation, a new algorithm based on the Diffraction Bounded Image Formation Model (DB-IFM) is proposed. The diffraction correction restores the image based on the diffraction-bounded restoration. The dark channel of the image assessed with various transmission maps based on Dark Channel Prior (DCP) algorithm. For a broad set of data with varying luminance, the outcomes are looked at both subjectively and quantitatively by enhanced restoration.