Area topology for road extraction and topographic data validation

Varley, David; Visvalingam, Maheswari

Abstract

First paragraph:
The Cartographic Information Systems Research Group (CISRG) of the University of Hull is researching the automatic recognition of spatial objects based on their spatial descriptions alone. The feasibility study , reported in this paper, formed a part of an SERC CASE project (January 1990 - January 1993); the collaborating body was the Ordnance Survey of Great Britain. The project involved the recognition of objects implied in large - scale topographic maps.
We define object recognition as the process of identification of objects from their forms alone; i . e. without recourse to semantic labels. Unlike recognition, extraction uses all available information, including semantic labels manually associated with line segments. This paper is concerned with the problem of road extraction. The principal aim of this feasibility study was to assess whether it would be possible to recognise roads given the structure and content of an experimental topographic database designed and created by the Ordnance Survey. The emphasis in the feasibility study was on acquiring a good understanding of large - scale topographic data and of the data
processing problems. Although some attention was paid to the efficiency of processing, this was not the primary concern at this stage. Efficient processing of geo-spatial data is a major topic of research which has to take into account a wider range of considerations (Frank, 1991) .

Citation

Varley, D., & Visvalingam, M. Area topology for road extraction and topographic data validation

• Working Paper Type: Discussion Paper
• Deposit Date: Mar 7, 2014
• Pages: 38
• Series Title: C.I.S.R.G. Discussion Papers (University of Hull Cartographic Information Systems Research Group)
• Series Number: 11
• Book Title: Area topology for road extraction and topographic data validation
• Keywords: Cartography -- Data processing, Computer science
• Public URL: https://hull-repository.worktribe.com/output/459280