Visual knowledge annotation and management by using qualitative spatial information

  • Authors:

  • Pedro José Vivancos-Vicente;Jesualdo Tomás Fernández-Breis;Rodrigo Martínez-Béjar;Rafael Valencia-García

  • Affiliations:

  • Tecnologías del Conocimiento y Modelado Cognitivo (TECNOMOD) Group, Facultad de Informática, Murcia, Spain;Tecnologías del Conocimiento y Modelado Cognitivo (TECNOMOD) Group, Facultad de Informática, Murcia, Spain;Tecnologías del Conocimiento y Modelado Cognitivo (TECNOMOD) Group, Facultad de Informática, Murcia, Spain;Tecnologías del Conocimiento y Modelado Cognitivo (TECNOMOD) Group, Facultad de Informática, Murcia, Spain

  • Venue:
  • PKAW'06 Proceedings of the 9th Pacific Rim Knowledge Acquisition international conference on Advances in Knowledge Acquisition and Management
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

The wide use of the Internet and the increasingly improvement of communication technologies have led users to need to manage multimedia information. In particular, there is an ample consensus about the necessity of new computational systems capable of processing images and “understand” what they contain. Such systems would ideally allow to retrieve multimedia content, to improve the way of storing it or to process the images to get some information interesting for the user. This paper presents a methodology for semi-automatically extracting knowledge from 2D still visual multimedia content, that is, images. The knowledge is acquired through the combination of several approaches: computer vision (to get and to analyse low level features), qualitative spatial analysis (to obtain high level information from low level features), ontologies (to represent knowledge), and MPEG-7 (to describe the information in a standard-way and make the system capable of performing queries and retrieve multimedia content).