Incorporating Metric Access Methods for Similarity Searching on Oracle Database

Daniel S. KasterPedro H. BugattiAgma J. M. TrainaCaetano Traina Jr.

The volume of multimedia and complex data (images, videos, audio, time series, DNA sequences, and others) has been growing at a very fast pace. Thus, it is necessary to store in databases many types of data which are not naturally handled by Database Management Systems (DBMSs). Complex data are suited to be queried by similarity. Many works addressed techniques for similarity searching, but the majority of them are not conceived to be integrated into a database engine. However, including similarity search into the database core would allow one to take advantage of the resources provided by the DBMS to perform queries integrating complex and conventional data. Recently, Oracle Corp. developed the Oracle interMedia module to support multimedia data in its database manager, providing several operations to manipulate them. It allows performing content-based image search through proprietary functions to extract intrinsic features from images and to compute their similarity. In this paper we describe a module for similarity search developed using the Oracle's Extensible Architecture Framework. Our approach allows including user-defined feature extraction methods and distance functions into the database core, providing a wider flexibility. We also present experiments that show that employing our module to query images by content outperformed the Oracle interMedia module, both in the precision of the results and in the performance of executing queries.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: