University of Khartoum

Quality Assessment of Digital Elevation Models Derived From Different Data Sources

Quality Assessment of Digital Elevation Models Derived From Different Data Sources

Show full item record

Title: Quality Assessment of Digital Elevation Models Derived From Different Data Sources
Author: Abdalmagid, Awadelgeed Mohamed Awadelgeed
Abstract: This thesis describes research undertaken to assess the quality of Digital Elevation Models (DEMs) derived from variant sources of elevation data sets, in particular, elevation data captured using remote sensing devices such as Shuttle Radar Topography Mission (SRTM), Advanced Space-borne Thermal Emission and Reflection Radiometer (ASTER) and Environmental-Satellite (ENVISAT); data sets that extracted from existing topographic maps, Real Time Kinematic- Global Positioning System (RTK-GPS); electronic tachometry and elevation data sets acquired by digital photogrammetric means. Also this thesis describes the cartographic quality of the contour lines drawn from the constructed DEMs and their validity for topographic mapping at variant scales. Despite increasing concern for understanding and working with the uncertainty within DEMs, knowledge about DEM quality is still at a primitive stage especially in the developing countries. The research presented here comprises a holistic approach to DEM quality assessment. This holistic approach combines statistical and visual quality assessment techniques. The statistical quality assessment technique was based on root mean square error, average absolute error, and minimum/ maximum error, while a visual quality assessment technique was based on carrying out comparisons between corresponding contour maps, error profiles, and error frequency histograms. The quality of a number of DEMs, produced using varying data sources, sampling patterns and interpolation methods, is assessed using these techniques. It is shown that these two assessment techniques are required to give a comprehensive assessment of all aspects of DEM quality. These two techniques are based on the hypothesis that the distribution and magnitude of elevation error within a DEM are at least partly related to the heterogeneity of the terrain, interpolation method, elevation data density and distribution. The results obtained in this study for different sources of elevation data set could be summarised as follows: The overall root mean square error of the digital elevation model derived from elevation data set acquired by electronic tachometry = ± 0.035m; the overall root mean square error of the DEM derived from elevation data set acquired by Real Time Kinematic- Global Positioning System = ± 0.043m; the overall root mean square error of the DEM derived from elevation data set acquired by digital photogrammetric means = ± 0.657m; the overall root mean square error of the DEM derived from 1:100000 scale topographic maps = ± 6.308m; the overall root mean square error of ENVISAT-DEM = ± 7.094m; the overall root mean square error of SRTM-DEM = ± 12.154m, and the overall root mean square error of ASTER-DEM = ± 15.607m. Statistical and visual results revealed that the errors found in all DEMs tested in this study were randomly distributed; hence, they could not be removed or remedied. A key outcome of this research is a defined procedure for producing a comprehensive DEM quality report. The research presented in this thesis provides the knowledge and the means for incorporating consideration of DEM quality. Users of DEMs were advised to be aware of the uncertainty of the DEMs provided by commercial agencies and firms. Additionally, generation of DEMs from topographic maps, with reasonable accuracy, can be executed using appropriate and unsophisticated interpolation method such as natural neighbour (NN). A general yard stick for the assessment of the quality of the TOPOMAP-DEM generated contours was suggested based on the general topographic mapping standards (equation 7-4). Also, a refined version of equation 7-4, was proposed by relating the elevation ranges to the RMSE values for different study areas (equation -7-5).
Description: 195 Pages
URI: http://khartoumspace.uofk.edu/123456789/16859
Date: 2015-11-03


Files in this item

Files Size Format View

This item appears in the following Collection(s)

Show full item record

Share

Search DSpace


Browse

My Account