Identifying roads and houses from scanned aerial photography and combining them with TIGER and digital terrain data of suburban areas

Author

Gervin, Janette C. ; Ragan, Robert M.

Author_Institution

NASA Goddard Space Flight Center, Greenbelt, MD, USA

fYear

1993

fDate

18-21 Aug 1993

Firstpage

1550

Abstract

A method for deriving the location of houses and roads from scanned NHAP and NAPP aerial photography using spectral, spatial (spatial smoothing and filtering and road segment size), and contextual (proximity to road) techniques was demonstrated. Despite differences in house size and construction age, analysis of both test and verification sites achieved acceptable accuracies, with more than 92 percent of the houses correctly identified. This house and road image derived from scanned USGS NHAP aerial photography, demographic data from the Bureau of the Census TIGER/Line files, and digital terrain data provided by the USGS Digital Elevation Model (DEM) files were then combined using analysis techniques developed to utilize imagery from spaceborne multispectral instruments

Keywords

cartography; geophysical techniques; geophysics computing; image recognition; remote sensing; NAPP; TIGER; cartography; context; contextual; digital terrain; geophysical measurement technique; house; image recognition; imagery; land surface terrain mapping; location; pattern recognition; remote sensing; road; scanned aerial photography; spatial data; suburban; urban area; Demography; Digital elevation models; Filtering; Image analysis; Image segmentation; Instruments; Photography; Roads; Smoothing methods; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Geoscience and Remote Sensing Symposium, 1993. IGARSS '93. Better Understanding of Earth Environment., International

Conference_Location

Tokyo

Print_ISBN

0-7803-1240-6

Type

conf

DOI

10.1109/IGARSS.1993.322766

Filename

322766