Speckle reduction in remote sensing images

摘要

Speckle noise contaminates image content and thus detracts from image interpretation. Speckle noise is the grainy salt-and-pepper pattern present in radar imagery caused by the interaction of out-of-phase waves with a target. Speckle noise reduction is usually employed prior to further image analysis. The primary goal of speckle filtering is to reduce speckle noise without sacrificing information content. The ideal speckle filter should adaptively smooth speckle noise, retain edges and features, and also preserve subtle but distinguishable details, such as thin linear features and point targets. Various speckle filters have been devised due to their different purposes and different capacities. In the present paper we are analyzing median, Lee and Wiener2 filters for the sampled SAR images from ENVISAT, RADARSAT2, US Library Congress and comparing their performances statistically by analyzing quality parameters like data mean, standard deviation, kurtosis and skewness. The paper is concluded with visual quality comparison of the input images and de-speckled images.