HR: 0800h
AN: IN31A-03 [Abstracts]
TI: Automated building extraction from LiDAR data
AU: * Dong, P
EM: pdong@unt.edu
AF: University of North Texas, 1155 Union Circle, #305279, Denton, TX 76203, United States
AB:
This paper presents some new methods for efficient building extraction using Light Detection and Ranging
(LiDAR) data.
One-meter resolution Digital Surface Models (DSM) and Digital Elevation Models (DEM) will be created using
spatial interpolation, and a Normalized Digital Surface Model (nDSM) will be created from DSM and DEM. As an
application example, the nDSM will be used for automated separation of residential areas from commercial
and industrial areas in the City of Denton, TX. The residential areas will be further used for population
estimation at the census tract, census block group, and census block levels. For accuracy assessment,
building areas derived from LiDAR will be compared with the 2000 parcel data in GIS, and population
estimates derived from residential areas will be compared with the 2000 census data. The impact of the
research includes: (1) facilitate the application of LiDAR data for more accurate land-use mapping; (2) provide
an efficient and cost-effective method for population estimation in inter-census years; and (3) provide a new
way of redistributing population data at finer grids to support other researches such as epidemic modeling and
emergency management and planning.
DE: 0758 Remote sensing
DE: 1819 Geographic Information Systems (GIS)
SC: Earth and Space Science Informatics [IN]
MN: 2009 Joint Assembly