An Exploratory Study of Generalisation of GIS Data Layers using Object-Oriented Modelling

MSc Dissertation, University College London

Chao Li

Abstract

Generalisation remains a significant problem in GIS. It is necessary for deriving multiple representations from a single data set for data integration, data analysis and the cartographic presentation of information. Traditional approaches to generalisation have focused on the simplification of the graphic component. However, changes in scale and resolution should be linked to changes in the underlying geographical meaning, that is, model generalisation. This dissertation explores the suitability of using an object-oriented approach to achieve model generalisation. This is carried out in the context of Hong Kong where 1:5,000 scale and 1:10,000 scale map series need to be derived from digital 1:1,000 scale Arc/Info data sets. The characteristics of the object-oriented approach are reviewed and an object-oriented model has been created for a sample data set. The model is implemented through ArcView Avenue and C++ programming. In achieving multiple representations from a single data set, this dissertation provides proof-of-concept for the use of an object-oriented approach to model generalisation in GIS.