Please use this identifier to cite or link to this item:
Title: Automated classification of basic-level terrain features in digital elevation models
Authors: Graff, Linda H.
Keywords: Terrain analysis
Automated terrain classification
Terrain reasoning
Digital elevation data
Publisher: U.S. Army Topographic Engineering Center.
Engineer Research and Development Center (U.S.)
Series/Report no.: TEC ; 0013.
Description: Technical report
Abstract: This study addresses a two-class problem in automatic terrain classification. The basic-level terrain feature selected for initial delineation is a "mount" - an aggregation of elevated features including hills, mountains, and ranges. All remaining areas are classified collectively as "non-mount." A method was developed to partition digital elevation models (DEMs) into mount and non-mount areas automatically. Then, the developed method is compared to those results obtained by a manual classification of synthetic stereo images produced from the same digital elevation data. The results of this work suggest that it may be possible to replicate the manual identification of mounts in certain physiographic regions. However, the general utility of the mount/non-mount classification approach appears to be limited by the nature of the regional terrain and by the quality of available digital data.
Rights: Approved for public release; distribution is unlimited.
Appears in Collections:Technical Report

Files in This Item:
File Description SizeFormat 
TEC-0013.pdf5.58 MBAdobe PDFThumbnail