Phenomenological marine snow model for optical underwater image simulation: Applications to color restoration

Optical imaging plays an important role in oceanic science and engineering. However, the design of optical systems and image processing techniques for subsea environment are challenging tasks due to water turbidity. Marine snow is notably a major source of image degradation as it creates white bright spots that may strongly impact the performance of image processing methods. In this context, it is necessary to have a tool to foresee the behavior of these methods in marine conditions. This paper presents a phenomenological model of marine snow for image simulation. In order to highlight the interest of such a modeling for image processing characterization, the impact of marine snow perturbation on a color restoration technique is analyzed and a solution to improve the robustness of the algorithm is finally proposed.