A geometrical data classification using Self-Organizing map with fixed possible matching units

Self-Organizing map (SOM) is a type of artificial neural network (ANN). It is the most well-known in unsupervised cluster or data classification. Consequently, this paper aims firstly to develop SOM algorithm with fixed possible matching units in order to apply with remote sensing of unsupervised classification. Secondly, it is to classify remote sensing for analyzing status of rice paddy by using SOM algorithm. The sample area is in Ayutthaya province. The research uses 3 variables which consist of band combinations, Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI) to classify remote sensing data for analyzing status of rice paddy by using SOM. It is concluded that the result of this research help to develop unsupervised classification of remote sensing to obtain more convenient and accurate utilization.