Noise Reduction in Sonar Images of Seabed Targets Based on a Variational Method

Abstract

Sonar imaging is widely applied in the intelligent perception of underwater targets. Sonar images usually have severe noise. The noise reduction in sonar images is an important and difficult problem in sonar image processing. There have been reports of variational methods used in image noise reduction. A method for the sonar image noise reduction based on a variational method is given in this paper, and it is compared with other commonly used methods for the noise reduction. It can be seen from the experimental results that the effects of the noise reduction and the edge preservation for the method in this paper are both better. The method can better balance the effects of the noise reduction and the edge preservation. However, the method requires iterative operations and requires a large amount of calculation.