Work

An overview of speckle noise filtering in SAR images

Year: 1997
Type: article
Abstract: Synthetic Aperture Radar Images are corrupted by a signal-dependent noise called speckle, that decreases the potentiality of these images for human or automatic interpretation. This degradation is due to the coherent nature of the radiation that is used and its interaction with the roughness of the terrain. In this paper, a survey of the techniques that have been proposed for reducing speckle noise in a SAR image is presented. First, a statistical model for the speckle noise is presented. This model, multiplicative in its nature, leads to the majority of the algorithms for speckle noise reduction. We describe the Lee, Kuan, Frost and adaptive versions of the last three filters, which are the most well known algorithms, as well as the MAP filter and some heuristic filters. We also briefly present new algorithms based on the theories of Markov Random Fields and Robust Estimation. Finally, we describe techniques for the evaluation of speckle reduction filters.
Language: English
Open Access status: green
OpenAlex ID: W100718534
FWCI: 0.361
Cites: 36
Cited by: 25
Related to: 10