CLASSIFICATION OF RADARSAT-2 POLARIMETRIC DATA FOR DIFFERENT LAND FEATURES USING FUZZY AND ANN CLASSIFIER
SAR data analysis for a range of applications from compact and fully polarimetric SAR like RADARSAT-2 is becoming popular every day on the fact that they offer features like higher resolution imaging, wide swath, reduced PRFs. Data from this family of SAR are very useful in several applications. The pixel percentage belonging to the user-defined area that is assigned to cluster in a confusion matrix for RADARSAT-2 over the Vancouver area has been analyzed for classification. In this study, Fuzzy and ANN classifiers over RADARSAT-2 (RS2) is computed and compared. First the input image(fully polarimetric SAR like RADARSET-2) noise is removed by applying lee filter. Then applying feature extraction process to get texture feature from the image. These features are applied to ANN and Fuzzy classifier to find the classified image i.e. water, urban and vegetation. In comparison with a conventional single channel or dual channel polarization, RADARSAT-2 is fully polarimetric, making it offer better land feature contrast for classification operation.
Reference Paper: Supervised Classification of RADARSAT-2 Polarimetric Data for Different Land Features
Author Name: Abhishek Maity
Source: Computer Vision and Pattern Recognition-Cornell University
Year:2016
Request source code for academic purpose, fill REQUEST FORM or contact +91 7904568456 by WhatsApp or info@verilogcourseteam.com, fee applicable.
SIMULATION VIDEO DEMO
