Detection of rice-area using self-organizing feature map

A neural network of Self-Organizing feature Map (SOM) to classify remote sensing data including microwave and optical sensors for the estimation of rice-planted area is considered. The method is an unsupervised neural network and has capability of nonlinear discrimination and the classification function is determined by learning. The satellite data are observed before and after planting rice in 1999. Three RADARSAT and one SPOT/HRV data are used in Higashi-Hiroshima, Japan. RADARSAT image has only one band data and it is difficult to extract rice-planted area. However, SAR back-scattering intensity in rice-planted area decreases from April to May and increases from May to June. Thus, three RADARSAT images from April to June are used in this study. The SOM classification was applied the RADARSAT and SPOT to evaluate the rice-planted area estimation. It is shown that the SOM is useful for the classification of the satellite data.