期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2010
卷号：XXXVIII - Part 8
页码：219-224
出版社：Copernicus Publications
摘要：Urban extraction is one of the most expected applications using remote sensing, but the automatic extraction has been challenging. Especially in the field of SAR applications, the complex scattering in the urban area is sensitive to the building spatial arrangement, and it prevents from the automatic extraction. Spaceborne Polarimetric synthetic aperture radar (POLSAR), an advanced approach to synthetic aperture radar (SAR), has been operated since PALSAR (Phased Array type L-band Synthetic Aperture Radar) onboard ALOS (Advanced Land Observing Satellite) was launched in 2006. Several indicators derived from POLSAR data have been developed to classify landcovers, and some of them have been utilized to extract the geometric features of the target. One of such indicators is Polarization Orientation Angle (POA), which is reported to estimate slant angle of the target. In addition, the indicators related to the density of the target are also proposed. Therefore, in this research, we examined the effectiveness of such polarimetric indices as urban structural detectors through a laboratory experiment, and we developed a classifier composed of the indices, which can be applicable for the satellite data. First, we measured the backscattering from the concrete blocks arranged in different slant angles and distances in an anechoic radio wave chamber. In this experiment, it was found the interrelation between spatial arrangement and such indicators as POA and density-related indicators. As a result, it is demonstrated that POA is a good detector for slant angle of man-made structur es, and of all the indicators, entropy has the highest correlation with building density. Then, using satellite polarimetric data, we discriminated urban areas according to the classifier using POA and entropy. The comparison with aerial photo indicated that POA is an effective indicator to extract the slant buildings and that there are some areas where entropy distinguishes the difference of building density
