Cao, Z; (2016) Improving the accuracy and the efficiency of geo-processing through a combinative geo-computation approach. Doctoral thesis , UCL (University College London).

Preview

Text Cao_PhD Thesis_Zhiwei_Cao_UCL.pdf Download (7MB) | Preview

Abstract

Geographical Information Systems (GIS) have become widely used for applications ranging from web mapping services to environmental modelling, as they provide a rich set of functions to solve different types of spatial problems. In the meantime, implementing GIS functions in an accurate and efficient manner has received attention, throughout the development of GIS technologies. This thesis describes the development and implementation of a novel geo-processing approach, namely Combinative Geoprocessing (CG), which is used to address data processing problems in GIS. The main purpose of the CG approach is to improve the data quality and efficiency of processing complex geo-processing models. Inspired by the concept of Map Calculus (Haklay, 2004), in the CG approach GIS layers are stored as functions and new layers are created through a combination of existing functions. The functional programming environment (Scheme programming language) is used in this research to implement the function-based layers in the CG approach. Furthermore, a set of computation rules is introduced in the new approach to enhance the performance of the function-based layers, such as the CG computation priority, which provides a way to improve the overall computation time of geo-processing. Three case studies, which involve different sizes of spatial data and different types of functions are investigated in this research in order to develop and implement the CG approach. The first case study compares Map Algebra and our approach for manipulating two different raster layers. The second case study focuses on the investigation of a combinative function through the implementation of the IDW and Slope functions. The final case is a study of computational efficiency using a complex chain processing model. Through designing the conceptual model of the CG approach and implementing the CG approach in the number of case studies, it was shown that the new approach provides many advantages for improving the data quality of geo-processing. Furthermore, the overall computation time of geo-processing could be reduced by using the CG approach as it provides a way to use computer resources efficiently and avoid redundant computations. Last but not least, this thesis identifies a new research direction for GIS computations and GIS software development, such as how a robust geo-processing tool with higher performance (i.e. data quality and efficiency) could be created using the CG approach.

Download activity - last month

Export as XMLJSONCSV

Download activity - last 12 months

Export as CSVJSONXML

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item