Elimination of Marine Snow effect from underwater image - An adaptive probabilistic approach

Along with color loss, another severe problem of underwater optical imaging is Marine Snow effect which occurs because of back scattering from suspended organic detritus, solid particles or bubbles. Their appearance like tiny sparkling dots often reduces the scene perception and sometimes leads to spurious features on segmentation. This paper is concerned with removal of the marine snow effect from underwater images by a probabilistic approach considering the local statistics of luminance properties after a RGB to YCbCr transform.