A thesis submitted in partial fulfillment of t he requirements for the degree of Master of Engineering, School of Environment, Resources and Development

Thesis (M.Eng.) - Asian Institute of Technology, 1994

In this study, ERS - 1 Synthetic Aperture Radar {SAR) and SPOT/1 HRV xs, LANDSAT Thematic Mapper {TM) data were evaluated to determine their utility to discriminate landuse\ cover and monitor landus e change of urban environment in the northern fringe area of Metropolitan Bangkok . The primary emphasis of the study was landuse \cover and landuse \cover changes discrimination performance of SPOT, TM, SAR and SAR combined with VIR. Among them, the study focuses on developing techniques which utilize synergistically both SAR and VIR sensing data with integration of GIS . To meet this objective , a key requirement was the development of a procedure to combine different period, different sensor data. Two data sets were merged into a single synergistic data set which contains information from both sets was generated. The result indicates that the combined us e of the different period SAR and VIR data sets enabled to detect change classes and provides a more complete characterization in urban environment than conventional methods using VIR a lone. Some special c l asses, such as new house and slum/low income house, have been observed indicating the strengths of such a combining application. Major difference between optical and radar systems are noted and their consequences were evaluated for synergistic c lassification. The advantage and relative limitations of the combining application are discussed. In an attempt to improvement the level of classification and refine the effective of monitoring landuse \cover changes. A GIS approach was employed, through establishing various thematic database, topographic data , landuse data, highway data and other ancillary data can be co-registered respectively with the different period classified images. The operation of a spatial co-occurrence data yielded improved discrimination of classes. The final classification provided better identification than remote sensing data alone. The operation of comparison using ERDAS and ARC/INFO software for classified images integration with various GIS database would be better monitor different classes change in different period.