Modelling agents in C++CL for content-based image retrieval

We describe a hybrid system for dealing with content-based retrieval from a library of images. An agent architecture is used in conjunction with a neural network to match and retrieve images. We explain how we define and use our image indices which allows us to treat the problem in a distributed fashion. We combine the indices with a multiple agent approach to help reduce our search space and to retrieve an image as a parallel action. The resulting system is one which is flexible but controlled. A neural network is used to generate a feature vector from the images which is used by the agents. The agents are based on the theories behind Lilies. Lilies (Localisation and InterLeaving strategies) was developed to deal with multiple agents for a planning environment. Lilies was used to prototype the initial representation of the agents used for searching the database. Although an implementation of the search-agents has been developed in Lilies, we also describe in this paper the design to convert the agent model to C++CL platform. C++CL is an object oriented framework layered on top of PVM which allows planned and unplanned reconfiguration of distributed systems. We introduce two issues of distributed systems: a neural net, for image compression; and the design of an agent searching architecture using C++CL.