The Sonar Image Denosing Method of Surfacelet Based on Sample Matrix

A new method for sonar image denoising which colligated the strongpoint of Surfacelet transform and sample matrix was presented. Surfacelet transform is good at multidimensional signal processing, but Surfacelet transform is not adapted to process 2-D image directly. In this paper, a noise images based on sample matrix is cycled spinning at a direction and compressed to image sequence, which is implemented Surfacelet transform and employed the hard thresholding denoising to coefficient respectively and then linear average in airspace. The experimental results indicate that image de-noised has not Gibbs-like phenomena of Wavelet and nick effect. The method not only can get obviously better visual effect and high PSNR value, but also hold detail and texture of image.