A Multi-Temporal Supervised Binary-Tree Classification Scheme for Polarimetric SAR with Maximum Difference of Polarization Signature
Conference: EUSAR 2016 - 11th European Conference on Synthetic Aperture Radar
06/06/2016 - 06/09/2016 at Hamburg, Germany
Proceedings: EUSAR 2016
Pages: 4Language: englishTyp: PDF
Personal VDE Members are entitled to a 10% discount on this title
Authors:
Huang, Xiaodong (Department of Geography, University of Western Ontario, Canada)
Wang, Jinfei (Department of Geography, University of Western Ontario, London, Ontario, N6A 5C2, Canada)
Shang, Jiali (Remote Sensing Applications Development Research Branch, Agriculture and Agri-Food (AAFC), Canada)
Abstract:
In this paper, a multi-temporal supervised binary-tree classification scheme (MTSBTCS-MDPS) is developed integrating the time dimension and the maximum difference of polarization signature (MDPS), in which each pair of targets is maximally distinguished through choosing the optimum orientation and ellipticity angles on an optimum data acquisition date. Seven C-band fully polarimetric RADARSAT-2 data with FQ21 mode acquired in 2012 over South-western Ontario, Canada are used for validation. Compared its classification result with that obtained from the single image by MDPS alone, the MTSBTCS-MDPS has much higher accuracy with its κ coefficient approximately 0.8 and overall accuracy of 84%.