A method of speckle noise reduction for SAR data

Synthetic Aperture Radar (SAR) is a high resolution active microwave sensor, but its processed data have nuisance noise called speckle noise which degrades image quality. This noise is conspicuous when the data are processed in small number of looks. In this paper, a new method for speckle reduction, the CST filter, is proposed. Using a characteristic that a distribution of observed SAR data for homogeneous targets can be estimated by theoretical probability density function (PDF), this method determines whether the target pixel is a member of a homogeneous target or not, then applies noise reduction only to the pixel, which is a member of homogenous targets. The proposed method, as well as three major existing methods, are applied to the simulated SAR data and the JERS-1/SAR data. By comparing all filtered images, the proposed method has better abilities of reproduction for a homogeneous target as well as preserving structured targets, such as, edges and point targets.