DocumentCode
3445017
Title
DEM-based slope unit derivation
Author
Chenchao Xiao ; Yuan Tian ; Ting Li ; Zhifang Gao
Author_Institution
China Aero Geophys. Survey & Remote Sensing Center for Land & Resources, Beijing, China
fYear
2013
fDate
20-22 June 2013
Firstpage
1
Lastpage
5
Abstract
This paper focuses on slope unit derivation from DEM by object-based image analysis. A landform classification system consisting of ridge, valley, slope unit, and flat terrain classes is firstly constructed considering the detailed characteristics of slope unit. Based on the classification system, a procedure for slope unit derivation from DEM, which takes into consideration the topographic factors including slope gradient, slope aspect and curvature, is designed within multi-scale OBIA framework. The method to determine the optimum segmentation scale by COLV is also introduced. A case study of Shenzhen is carried out to verify the feasibility and accuracy of the designed procedure. The results of the case study shows that when the scale parameter is 20, the derived slope units remarkably denote the detailed landform structure and the bounds of the derived units are largely consistent to the concave and convex terrain bounds.
Keywords
digital elevation models; geophysical image processing; image classification; COLV; DEM-based slope unit derivation; concave terrain bounds; convex terrain bounds; digital elevation model; flat terrain classes; landform classification system; multiscale OBIA framework; object-based image analysis; optimum segmentation scale; slope aspect; slope gradient; slope units; topographic factors; Geographic information systems; Image analysis; Image segmentation; Remote sensing; Surface topography; Surface treatment; Terrain factors; DEM; GIS; object-based image analysis; slope unit;
fLanguage
English
Publisher
ieee
Conference_Titel
Geoinformatics (GEOINFORMATICS), 2013 21st International Conference on
Conference_Location
Kaifeng
ISSN
2161-024X
Type
conf
DOI
10.1109/Geoinformatics.2013.6626065
Filename
6626065
Link To Document