Multi-agent system for detecting and analysing changes on satellite image sequence

The analysis of satellite images representing natural scenes introduced a very important volume information and requires a substantial elaboration at all levels: pre-processing, segmentation, recognition and interpretation. The consideration of all these phases influences unmistakably the quality of the tasks of processing and time execution. Quality and time are often the only criteria retained by these applications. In this paper, we propose a multi-agent approach which consists in developing a multi-agent system based on parallel processing concept for satellite image analysis changes, we describe the basics of our multi agent architecture concept and show how it can be used as basis of an automatic task parallel processing to speed up image processing. We further illustrate the design and implementation of the multi agent system that uses image analysis agents for generating and processing parallel programs by calling the available method. We validated our approach on satellite images of SPOT4 representing a north Tunisian region for different dates. The results obtained consists in classifying images and detecting the change.