Title :
Polarimetric SAR image classification using Multiple-Component Scattering Model and Support Vector Machine
Author :
Zhang, Lamei ; Zou, Bin ; Jia, Qingchao ; Zhang, Ye
Author_Institution :
Dept. of Inf. Eng., Harbin Inst. of Technol., Harbin, China
Abstract :
The classification of polarimetric SAR image based on Multiple-Component Scattering Model (MCSM) and Support Vector Machine (SVM) is presented in this paper. MCSM is a potential decomposition method for a general condition. SVM is a popular tool for machine learning tasks involving classification, recognition or detection. The scattering powers of single-bounce, double-bounce, volume, helix, and wire scattering components are extracted from full polarimetric SAR images. Combined with the scattering power and the texture feature, SVM is used for the polarimetric classification. We generate a validity test for the method using EMISAR L-band full polarized data of Foulum Area (DK), Denmark. The preliminary result indicates that this method can classify most of the areas correctly.
Keywords :
image classification; support vector machines; synthetic aperture radar; Denmark; EMISAR L-band full polarized data; Foulum Area; SVM; double-bounce components; helix components; machine learning tasks; multiple-component scattering model; polarimetric SAR image classification; potential decomposition method; scattering power; single-bounce components; support vector machine; wire scattering components; Data mining; Image classification; L-band; Machine learning; Polarization; Scattering; Support vector machine classification; Support vector machines; Testing; Wire; Classification; Multiple-Component Scattering Model (MCSM); Polarimetric SAR (PolSAR); Support Vector Machine (SVM);
Conference_Titel :
Synthetic Aperture Radar, 2009. APSAR 2009. 2nd Asian-Pacific Conference on
Conference_Location :
Xian, Shanxi
Print_ISBN :
978-1-4244-2731-4
Electronic_ISBN :
978-1-4244-2732-1
DOI :
10.1109/APSAR.2009.5374178
