Sidescan sonar image de-noising algorithm in multi-wavelet domain

Sidescan sonar image is widely used in seafloor mapping, geological survey, probing seafloor barrier and so on. For the complexity of underwater environment, the sonar is polluted by noise more seriously than that in optical image. So it is necessary to remove noise of sonar image. By transforming the noisy sonar image into multi-wavelet domain, at the same time using the knowledge of partial differential equations, and considering fractal dimension of every subband, a sonar image de-noising algorithm based on multi-wavelet threshold is proposed in the paper. The result of the experiment shows that the effect of the algorithm is better than that of scalar wavelets, and the original information is kept after de-noising using the algorithm.