Unsupervised Classification of Polarimetric Circular SAR Images Based on H/α/Multi-aperture Polarimetric Entropy

Conference: EUSAR 2018 - 12th European Conference on Synthetic Aperture Radar
06/04/2018 - 06/07/2018 at Aachen, Germany

Proceedings: EUSAR 2018

Pages: 4Language: englishTyp: PDF

Personal VDE Members are entitled to a 10% discount on this title

Authors:
Hong, Wen (University of Chinese Academy of Sciences, Beijing, China & Institute of Electronics, Chinese Academy of Sciences, Beijing, China & Key Laboratory of Technology in Geospatial Information Processing and Application System, Beijing, China)
Xue, Feiteng; Chen, Shiqiang; Shen, Wenjie (Institute Of Electronics Chinese Academy Of Sciences, P.R. China)
Lin, Yun (Institute of Electronics, Chinese Academy of Sciences, P.R. China)

Abstract:
In conventional polarimetric synthetic aperture radar(PolSAR), targets are usually assumed isotropic and potential polarimetric variations across azimuth are unconsidered. As to circular SAR (CSAR), the azimuthal view angle is much larger. Polarimetric variations across azimuth are no longer negligible. H/α unsupervised classification method classifies data by polarimetric properties. H/α method can not distinguish anisotropic and isotropic targets which have same polarimetric properties when applied to polarimetric CSAR(Pol-CSAR) data. Recently multi-aperture polarimetric entropy(MAPE) was proposed in Ref. [10]. MAPE can be used to identify anisotropic and isotropic targets. In this paper, an unsupervised classification method based on H/ /MAPE is proposed. The proposed classification method classifies data with H/ /MAPE space. Anisotropic and isotropic targets which have same polarimetric properties are well classified into two classes by H/α/MAPE method. The effectiveness and advantages of H/α/MAPE method are demonstrated on Pol-CSAR data, acquired from the Institute of Electronics airborne CSAR system at P-band.