Comparative study of speckle reduction approaches

The speckle reduction is an important problem in coherent imaging, such as synthetic aperture radar or ultrasound imagery. There are several well-known speckle filters for reducing speckle in coherent images. However, different speckle filters will have their own advantages and disadvantages. The performance of speckle filter depends strongly on the speckle and scene models that are used as the basis for filter development. In this thesis, we will compare four different approaches, Lee filter, Kuan filter, Frost filter and Improved Sigma filter for speckle reduction. This thesis will further describe the Improved Sigma filter since it is a new type of speckle filter for coherent images. We will apply these four speckle filters to our synthesized one-look amplitude synthetic aperture radar (SAR) image to reduce speckle noise. In order to compare these different speckle filters, the simulated SAR image data are used quantitatively to analyze these speckle filters. In addition, we will show their advantages and disadvantages in reducing speckle noise. Unlike the original Sigma filter, which has deficiencies in causing biased estimation and in depressing strong reflected targets, the Improved Sigma filter is more efficient when applied on simulated SAR image data