Chapter

Advances in Multimedia Information Systems

Volume 1508 of the series Lecture Notes in Computer Science pp 110-120

Date:

Retrieving Images by Content: The Surfimage System

  • Chahab NastarAffiliated withINRIA Rocquencourt
  • , Matthias MitschkeAffiliated withINRIA RocquencourtSiemens AG
  • , Nozha BoujemaaAffiliated withINRIA Rocquencourt
  • , Christophe MeilhacAffiliated withINRIA Rocquencourt
  • , Héléne BernardAffiliated withAlcatel Corporate Research Center
  • , Marc MautrefAffiliated withAlcatel Corporate Research Center

* Final gross prices may vary according to local VAT.

Get Access

Abstract

Surfimage is a versatile content-based image retrieval system allowing both efficiency and flexibility, depending on the application. Surfimage uses the query-by-example approach for retrieving images and integrates advanced features such as image signature combination, multiple queries, query refinement, and partial queries. The classic and advanced features of Surfimage are detailed hereafter. Surfimage has been extensively tested on dozens of databases, demonstrating performance and robustness. Several experimental results are presented in the paper.