Rainfall Estimation from GMS Imagery Data Using Neural Networks

Abstract

The forecasting approaches based on Remote-Sensing techniques have been developed. But there are difficulties in these approaches because of computational performance which is calculated in time and capacity of computer. This work is the method of forecasting the variation of clouds with Neural Network. Neural Network is defined as that application of the model of brain and neuron systems to engineering and is useful to analyze the infrared and visible images of GMS. Many characteristics of GMS images such as ground data, temperature and wind are stored in a Neural Network for forecasting rainfall. The range of imagery data was reveled through the comparison between infrared and visible imagery data to estimate rainfall.