Speckle Noise Suppression of SAR Color Image using Hybrid Mean Median Filter

Aperture Rader (SAR) color images are inherently affected by multiplicative speckle noise, which is a granular noise that degrades the quality of SAR information content. Image interpretation, recognition and terrain classification processes become very difficult due to the presence of speckle noise. In this paper, an algorithm using Hybrid order statistics filter- HMM (Hybrid Mean Median) is proposed, which is a combination of mean and median filters to suppress speckle noise in SAR color images. The filter is tested with color SAR color image by two different mask sizes of 7*7 and 5*5. The performance of the proposed filter is tested against the standard filters such as Frost, Kuan, Lee and Median for reduction of speckle noise. The performance metrics shows that the proposed filter performs better in terms of PSNR and MSE.