PROJECT TITLE :

AgentGeo: Multi-Agent System of Satellite Images Mining

ABSTRACT:

The continuing growth, availability and demand for remote sensing knowledge combined with human impossibility in manual and individual analysis of these information leads to a would like to investigate new tools and techniques for automated analysis (or semi-automatic) and intelligent remote sensing databases. Mining images is an alternative for explore massive database of remote sensing pictures, which are strategies to discover patterns within the image repository. The Multi-Agent Systems have features that, properly implemented, will bring vital enhancements in analysis of satellite pictures. During this context, the aim of this paper is to present the AgentGeo, a multi-agent system of satellite image mining capable of performing the classification method with multiple images at a time. The AgentGeo implements functions such as creation, edition and choice of agents, choice and creation of the environment and use of agents mining to mine the satellite imagery. This method was developed in Java and uses features from TerraView and GeoDMA. The results of the classification method were satisfactory once the developed agents had an accuracy higher than 90% when put next to Google Maps pictures.