A Grayscale Reader for Camera Images of Xerox DataGlyphs.
In: Rosin, PL and Marshall, D, (eds.)
BMVC 2002: Proceedings of the 13th British Machine Vision Conference: 2nd-5th September 2002, Cardiff University.
(pp. 698 - 707).
BMVA: Malvern, UK.
Xerox DataGlyph technology is a 2-D matrix barcode encoding which com- bines relatively high data density with an error correction scheme and an unobtrusive format. Some of the more exciting applications proposed for this technology involve the decoding of unconstrained camera images of DataGlyphs. This paper details a set of grayscale methods including steered gradient filters and the Radon transform which can decode many uncon- strained camera images. This method has been successfully tested on nearly camera images with very good results. Flowport™ and is an active area of development 1. This paper describes robust methods of decoding DataGlyphs from images taken by a hand-heldcamera or webcam, extending the range of applications for which DataGlyphs can be used. Thisintroductionexplainsindetail the DataGlyphformatand theuniquechallengesof decoding camera images of DataGlyphs. The paper then describes the proposed decoder and the methods used in it. Results of the new decoder are compared with a binarization and correlation-based method, followed by the paper's conclusions
|Title:||A Grayscale Reader for Camera Images of Xerox DataGlyphs|
|UCL classification:||UCL > School of BEAMS > Faculty of Engineering Science > Computer Science|
Archive Staff Only