Andrew P. Berman and Linda G. Shapiro,
Efficient Content-Based Retrieval: Experimental Results,
In IEEE Workshop on Content-based Access of Image and Video
Libraries (CBAIVL'99),
Fort Collins, Colorado, USA, pp. 55-61, June 22 1999.
Andrew P. Berman and Linda G. Shapiro,
Efficient Content-Based Retrieval: Experimental Results,
In IEEE Workshop on Content-based Access of Image and Video
Libraries (CBAIVL'99),
Fort Collins, Colorado, USA, pp. 55-61, June 22 1999.The goal of our research has been to create technology useful in a generalized system for content-based image retrieval. Such a system should search and retrieve images quickly, and do so over a wide range of queries. As user definitions of similarity may change from session to session, we believe that flexibility in query formulation is an important quality for content-based retrieval systems. This flexibility can be achieved by providing the user with a large set of distance measures that determine the similarity between a user query and a database image and a small set of methods for combining several of them for a trial query. Distance measure calculation requires accessing either the images being compared, or pre-computed associated data. An image database can consist of millions of images, each one taking many megabytes of storage space. Distance measure calculations can be individually expensive as well. Thus, a system that must calculate the distance from the query image to each image in a large database may exhibit unsatisfactory performance.