An effective technique to identify river's stage through satellite images by means of RBFNN

Abstract

Today, a significant role is played by satellite image processing in the research improvement of various subject of analysis such as Astronomy, Remote Sensing, GIS, Agriculture Monitoring and Disaster Management. Forecasting natural disasters so that necessary safety measures can be taken to safeguard the surroundings is the objective behind the utilization of remote sensing images in most of the researches. In this paper, the stage of a river is predicted utilizing satellite images of the river. Initially, in the preprocessing phase, the image is filtered and then converted to the LAB color space for acute analysis. Subsequently, the segmentation process is carried out using the designed Radial Basis Function Neural Network (RBFNN) and then morphological operation are performed on the image. After that in the testing phase, the segmented image is analyzed and the stage of the river is identified as either normal or flood or draught using the designed RBFNN. © 2011 Springer-Verlag.