Autonomous Terrain Classification using Satellite Imagery

Abstract

Geo graphics Information System also commonly called as G IS is an innovative technological field that fuses geographical highlights with tabular geographically and statistically ascribed data and information, keeping in mind the end goal to delineate, map and survey genuine issues of earth . A G IS is a structure for collecting, overseeing, and examining information. Established in the study of geography, G IS incorporates numerous kinds of data.Hundreds of thousands of associations in virtually all intents and purposes are utilizing G IS to make maps that communicate, perform examination and analysis on earth, share information, and take care of complex issues around the world. This is changing the way the world works. Terrain segmentation and classification is a vital step of any G IS based application. It can be used to determine the various land forms in a region.Keeping in view its importance, objective of this project is to develop a desktop application for terrain segmentation and classification in satellite images. Classification of four generic land forms ( Urban settlement, Barren lands, Vegetation and Water Bodies) are the target of this system. By applying both Color based and texture based techniques, class ac curacies comparable with state of art are achieved