Bidimensional empirical mode decomposition for noise reduction in sonar images

In recent years, high resolution sonar images have been used for mapping the seafloor for a wide variety of purposes, such as the detection, localization and eventually classification of objects lying on sea beds. However, sonar images are often highly corrupted by noise. The purpose of this paper is to apply a BEMD-based filter to remove or suppress unwanted noise. Compared with classical means on simulated data and real sonar images, the filter used here can enhance the signal-to-noise ratio as well as preserve the edge information.