DocumentCode
2139298
Title
Automatic extraction of vertical obstruction information from interferometric SAR elevation data
Author
Woods, Donald ; Folley, Christopher ; Kwan, Yit-Tsi ; Houshmand, Bijan
Author_Institution
Technol. Service Corp., Los Angeles, CA, USA
Volume
6
fYear
2004
fDate
20-24 Sept. 2004
Firstpage
3938
Abstract
Vertical obstructions (VOs) such as radio towers, power pylons, and windmills present a hazard to low-flying aircraft. We describe an automated VO detection scheme that processes digital surface models (DSMs) provided by interferometric SAR (InSAR) systems to produce heights and locations of VOs. By using only the height data from the InSAR, the resulting techniques can be applied to other DSM sources, such as lidar. Innovative morphological filtering algorithms are described, which optimize detection performance against various DSM resolutions over smooth or rough terrain. The height and position estimates of the detected VOs are computed, and they compare favorably to ground truth. Software that implements these algorithms has been packaged as an add-on module to ERDAS IMAGINE®.
Keywords
air safety; feature extraction; geophysical signal processing; height measurement; mathematical morphology; object detection; position measurement; radiowave interferometry; remote sensing by radar; synthetic aperture radar; terrain mapping; DSM resolutions; ERDAS IMAGINE; InSAR systems; add-on module; digital surface models; hazard; height data; height estimation; interferometric SAR elevation data; lidar; low-flying aircraft; morphological filtering algorithms; position estimation; power pylons; radio towers; rough terrain; smooth terrain; vertical obstruction information extraction; windmills; Aircraft; Data mining; Filtering algorithms; Hazards; Laser radar; Poles and towers; Power system modeling; Radio interferometry; Surface morphology; Synthetic aperture radar interferometry;
fLanguage
English
Publisher
ieee
Conference_Titel
Geoscience and Remote Sensing Symposium, 2004. IGARSS '04. Proceedings. 2004 IEEE International
Print_ISBN
0-7803-8742-2
Type
conf
DOI
10.1109/IGARSS.2004.1369988
Filename
1369988
Link To Document