Improving image retrieval with a thesaurus for shapes. The VORTEX prototyp.

Abstract

Successful retrieval of relevant images from large-scale image collections is one of the current problems in the field of data management. In this paper, I propose a system which combines techniques from Computer Vision, such as feature extraction, with a thesaurus for objects / shapes. A brief presentation of the problem area is given, along with a presentation and test results from a prototype system, VORTEX. The VORTEX system has been compared to a retrieval system based on low-level image features alone, and is able to achieve a significantly higher degree of image recall.