Agent approach to Feature Based Image Retrieval in a Network

Abstract The developments in the field of internet allow users in almost all the professional areas for exploiting the opportunities offered by the ability to access and manipulate remotelystored images. The large multimedia database has to be processed within a small fraction of seconds for many of the real time applications. This demand of using the technique of content based image retrieval (CBIR) as a scheme for searching large database for image retrieval has addressed some of the issues that need to be solved for having an efficient system. The project focused on the issues of image retrieval and also suggests a method to get an accurate result by using a hybrid search methodology. It works with genetic algorithm to get a local optimal result and neural network to get a global optimal result. The overall system architecture is based on a multi-agent paradigm where agents autonomously search for images. Results show that this system is capable of significantly reducing the time and management effort associated with large amounts of image data.