Denoising of Underwater Acoustic Image

The high-resolution underwater acoustic image such as multireceiver synthetic aperture sonar (SAS) image often suffers from the speckle noise, which seriously affect the image performance. Besides, the subsequent tasks such as target detection and recognition are also seriously affected. In order to reduce the influence of speckle noise on the SAS image, the denoising method should be used to remove the speckle noise. In this paper, we propose to denoise the speckle noise based on the Sigma filter. With the presented method, the speckle noise is well removed. The metrics such as mean, variance, equivalent number of looks (ENL), and edge preserve index (EPI) are computed to evaluate the denoising performance, which demonstrates the effectiveness of proposed method.