Title:	CLASSIFICATION SPACE BASED ON THE TARGET ORIENTATION ANGLE USING POLARIMETRIC SAR DATA
Author(s):	Boularbah Souissi, Mounira Ouarzeddine , Aichouche Belhadj-Aissa
ISBN:	978-972-8939-22-9
Editors:	Yingcai Xiao, Tomaz Amon and Piet Kommers
Year:	2010
Edition:	Single
Keywords:	Polarimetry, Radar SAR, decomposition, classification
Type:	Full Paper
First Page:	117
Last Page:	123
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	In this paper, we propose a space model for land cover classification using polarimetric synthetic aperture radar (SAR) data, based on the orientation angle and the HH backscattering polarization. We demonstrate how these two combined parameters are used in order to decouple mixed young coniferous and deciduous forests from dense coniferous forests, and to identify both of surface scattering represented by bare soil and some agriculture fields, and urban area represented by a set of buildings. In order to assess the accuracy of the image classification results, a confusion matrix was calculated for each class. As a quality parameters, kappa coefficient, producer and user accuracy were used. Results found for land cover that are different in structure were similar to the Entropy/Alpha/Wishart classifier. The potential of the proposed space model algorithm is investigated using fully polarimetric airborne data acquired in the P band.
	Go Back