Infinite Number of Looks Prediction in SAR Filtering by Linear Regression
Mohamed Yahia, Tej-Albaha Hamrouni, Riadh Abdelfattah · IEEE Geoscience and Remote Sensing Letters · 2017
Speckle filtering in synthetic aperture radar (SAR) images is essential for the extraction of significant information for homogeneous extended targets. In this letter, we exploited the minimum mean square error to establish a linear rule between the values of the filtered pixels and their variances. Then, the filtered pixel for infinite number of looks (INL) was predicted by linear regression of means and variances for various window sizes. The results show that the proposed INL-prediction filter improved the filtering performances of the original filters. Simulated and real SAR data were used for validation.