Geometric Accuracy Testing, Evaluation and Applicability of Space Imagery to the Small Scale Topographic Mapping of the Sudan

Elniweiri, Abdelazim Elniweiri Hassan (1988) Geometric Accuracy Testing, Evaluation and Applicability of Space Imagery to the Small Scale Topographic Mapping of the Sudan. PhD thesis, University of Glasgow.

Abstract

The geometric accuracy, interpretabilty and the applicability of using space imagery for the production of small-scale topographic maps of the Sudan have been assessed. Two test areas have been selected. The first test area was selected in the central Sudan including the area between the Blue Nile and the White Nile and extending to Atbara in the Nile Province. The second test area was selected in the Red Sea Hills area which has modern 1:100,000 scale topographic map coverage and has been covered by six types of images, Landsat MSS TM and RBV; MOMS; Metric Camera (MC); and Large format Camera (LFC). Geometric accuracy testing has been carried out using a test field of well-defined control points whose terrain coordinates have been obtained from the existing maps. The same points were measured on each of the images in a Zeiss Jena Stereocomparator (Stecometer C II) and transformed into the terrain coordinate system using polynomial transformations in the case of the scanner and RBV images; and space resection/intersection, relative/absolute orientation and bundle adjustment in the case of the MC and LFC photographs. The two sets of coordinates were then compared. The planimetric accuracies (root mean square errors) obtained for the scanner and RBV images were: Landsat MSS +/-80 m; TM +/-45 m; REV +/-40 m; and MOMS +/-28 m. The accuracies of the 3-dimensional coordinates obtained from the photographs were: MC:-X=+/-16 m, Y=+/-16 m, Z=+/-30 m; and LFC:- X=+/-14 m, Y=+/-14 m, and Z=+/-20 m. The planimetric accuracy figures are compatible with the specifications for topographic maps at scales of 1:250,000 in the case of MSS; 1:125,000 scale in the case of TM and RBV; and 1:100,000 scale in the case of MOMS. The planimetric accuracies (vector =+/-20 m) achieved with the two space cameras are compatible with topographic mapping at 1:60,000 to 1:70,000 scale. However, the spot height accuracies of +/-20 to +/-30 m - equivalent to a contour interval of 50 to 60 m - fall short of the required heighting accuracies for 1:60,000 to 1:100,000 scale mapping. The interpretation tests carried out on the MSS, TM, and RBV images showed that, while the main terrain features (hills, ridges, wadis, etc.) can be mapped reasonably well, there was an almost complete failure to pick up the cultural features - towns, villages, roads, railways, etc. - present in the test areas. The high resolution MOMS images and the space photographs were much more satisfactory in this respect though still the cultural features are difficult to pick up due to the buildings and roads being built out of local material and exhibiting little contrast on the images.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: Remote sensing, Geographic information science and geodesy
• Date of Award: 1988
• Depositing User: Enlighten Team
• Unique ID: glathesis:1988-77748
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 14 Jan 2020 11:53
• Last Modified: 14 Jan 2020 11:53
• URI: https://theses.gla.ac.uk/id/eprint/77748