Images captured by synthetic aperture radar (SAR) are generally corrupted with noises. SAR images are commonly affected by speckle noise and impulse noise. Noise filtering techniques must remove noises and simultaneously preserve valuable information present in the images. This article presents a noise filtering techniques based on soft computing. The proposed filters are implemented in two phases: In the first phase, the presence of noise in a particular pixel is detected by using fuzzy logic, and in the second phase, noisy pixels are filtered by using Self-Organizing Map (SOM). From our experiment, it is found that the proposed SOM filter reduces both speckle and impulse noise and also preserves the information present in the image. The proposed SOM filter is comparatively evaluated with various filters based on peak signal-to-noise ratio and edge-preserving factor.
Medhi, K., Amitab, K., Kandar, D., & Paul, B. S. (2018). Noise Reduction in Synthetic Aperture Radar Images Using Fuzzy and Self-Organizing Map. In Lecture Notes in Electrical Engineering (Vol. 462, pp. 295–301). Springer Verlag. https://doi.org/10.1007/978-981-10-7901-6_32