DocumentCode :
998842
Title :
Building characterization using L-band polarimetric interferometric SAR data
Author :
Guillaso, Stéphane ; Ferro-Famil, Laurent ; Reigber, Andreas ; Pottier, Eric
Author_Institution :
Dept. of Comput. Vision & Remote Sensing, Berlin Univ. of Technol., Germany
Volume :
2
Issue :
3
fYear :
2005
fDate :
7/1/2005 12:00:00 AM
Firstpage :
347
Lastpage :
351
Abstract :
This letter proposes a building characterization technique for L-band polarimetric interferometric synthetic aperture radar (SAR) data. This characterization consists of building identification and height estimation. Initially, a polarimetric interferometric segmentation is performed to isolate buildings from their surroundings. This classification identifies three basic categories: single bounce, double bounce, and volume diffusion. In order to compensate for the misclassifications among the volume and the double-bounce classes, interferometric phases given by the high-resolution Estimation of Signal Parameters via Rotational Invariance Techniques (ESPRIT) method are analyzed. Once buildings are localized, a phase-to-height procedure is applied to retrieve building height information. The method is validated using E-SAR, German Aerospace Center (DLR) fully polarimetric SAR data, at L-band, repeat-pass mode, over the Oberpfaffenhofen, Germany, test site, with a spatial resolution of 1.5 m in range and azimuth. More than 80% of buildings are retrieved with acceptably accurate height estimates.
Keywords :
building; radar polarimetry; remote sensing by radar; synthetic aperture radar; terrain mapping; 0.23 m; ESPRIT method; Estimation of Signal Parameters via Rotational Invariance Techniques; Germany; L-band polarimetric interferometric SAR data; Oberpfaffenhofen test site; building characterization technique; building height estimation; building identification; misclassifications; polarimetric interferometric segmentation; single bounce; synthetic aperture radar; volume diffusion; Aerospace testing; Azimuth; Information retrieval; L-band; Parameter estimation; Phase estimation; Signal analysis; Spatial resolution; Synthetic aperture radar; Synthetic aperture radar interferometry; Building characterization; Estimation of Signal Parameters via Rotational Invariance Techniques (ESPRIT); classification; interferometry; polarimetry;
fLanguage :
English
Journal_Title :
Geoscience and Remote Sensing Letters, IEEE
Publisher :
ieee
ISSN :
1545-598X
Type :
jour
DOI :
10.1109/LGRS.2005.851543
Filename :
1468096
Link To Document :
بازگشت