Recognizing buildings using a mobile system and a reference city model

  • Authors:
  • Wanji Mai;Chris Tweed;Gordon Dodds

  • Affiliations:
  • Virtual Engineering Centre, Queen's University Belfast, Belfast, Northern Ireland, UK;School of Architecture, Queen's University Belfast, Belfast, Northern Ireland, UK;Virtual Engineering Centre, Queen's University Belfast, Belfast, Northern Ireland, UK

  • Venue:
  • Digital Cities'03 Proceedings of the Third international conference on Information Technologies for Social Capital: cross-Cultural Perspectives
  • Year:
  • 2003

Quantified Score

Hi-index 0.01

Visualization

Abstract

Recent advances in the development of personal digital assistants (PDAs) and wireless communication networks enable a new generation of sophisticated mobile applications. PDAs can now support a range of add-on devices, such as digital cameras, and communicate using a variety of networking protocols, such as GPS, WiFi, and Bluetooth. This paper reports on research into and development of portable hardware that will enable users in the field to send images, and associated positional data from a PDA to a server for further image processing. The central aim is to provide navigational and informational services to an urban mobile user based on building recognition. The paper begins by describing the hardware system before presenting research into server-side building recognition methods that operate by comparing user-supplied images of similar images generated by an existing 3D digital model. This method uses a combination with segmentation and Hough Transform. Results of image capture and matching are given.