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

Thesis (M.Eng.) - Asian Institute of Technology, 1987

The study has conducted to and reliability of remotely identification of land use/land Basin of Thailand. assess the applicability, feasibility sensed data for evaluation and cover categories in Central Rayong The result of Principal Component Analysis shows that band 1 and 2 are similar to PCl and band 3 is similar to PC2 yielded more or less similar results. However, from PC3 it was not possible to extract out any valuable information. The ratio of band 3/band 2 after edge enhancement was found successful classification procedure. The major problem of the imagery was the similarity between wet land and bare land. This problem was overcome by the black and white imagery of band 2 and band 3. The imagery was classified into different land use/land cover types using both visual and digital analysis, and the visual analysis provided a simple and inexpensive means for classification. The computer analysis yielded better result considering the detail inventory of land use/land cover types, inspire of heterogeneous and mixed categories of the study area. The reliability and mapping accuracy obtained from digital analysis and visual analysis are 85%, 73.9% and 73%, 57.48% respectively which can be accepted for necessary land use/land cover inventories.