Still images retrieval from a remote database: The system Imagine

Abstract This paper describes the main features of a prototype image retrieval system, nicknamed Imagine . Response time and scale-ability , that is the ability of maintaining the service in a wide range of workstation performances and network digital rates, constituted the focus of the investigation. The assumption that the database is located in a site remote from the user workstation, and that the network connecting them is relatively poor in bandwidth, forced the design of rather sophisticated navigation procedures beyond the adoption of the JPEG image coding scheme. In order to have an acceptable response time, the user must in fact be able to easily identify the desired images without having to transfer a large number of eventually useless ones. Reported objective performance evaluations have been carried out on a functionally complete laboratory system.