A comprehensive evaluation of filters for radar speckle suppression
Yongwei Sheng, Zong-Guo Xia · 2005
This paper presents the result of a comprehensive evaluation of filters for radar speckle suppression. Seven filters in the Radar Module of Erdas/IMAGINE(R) were evaluated, including the mean filter, the median filter, the Lee-sigma filter, the local region filter, the Lee filter, the Frost filter, and the MAP (maximum a posteriori) filter. The performance of these filters was measured in terms of five criteria: speckle suppression index, edge enhancing index, feature preserving index (for both linear features and point features), image detail preserving coefficient, and speckle image analysis. Visual effect of filtered image and its filter theoretical basis were discussed. The relationship between filter performance and speckle patterns was also examined.