Japanese / English

Detail of Publication

Text Language Japanese
Authors Tomohiro Nakai, Koichi Kise, and Masakazu Iwamura
Title Fast Camera-Based Document Image Retrieval Using Local Arrangements of Feature Points
Journal Trans. IEICE
Vol. J89-D
No. 9
Pages pp.2045-2054
Reviewed or not Reviewed
Month & Year September 2006
Abstract Document image retrieval is a task of searching document images from the database based on users' queries. In this paper, we propose a method of document image retrieval which employs document images captured by digital cameras as queries. Unlike document image retrieval using scanners, it is necessary to deal with the problem of perspective distortion. In the proposed method, document images are characterized using local arrangements of feature points, which allows us fast retrieval of O(N) where N is the number of feature points. For the description of arrangements, perspective invariants are employed for the retrieval under perspective distortion.
URL http://search.ieice.org/bin/summary.php?id=j89-d_9_2045&category=D&year=2006&lang=J&abst=
Back to list