Share Email Print

Proceedings Paper

Optimizing OCR accuracy for bi-tonal, noisy scans of degraded Arabic documents
Author(s): Paul Herceg; Benjamin Huyck; Christopher Johnson; Linda Van Guilder; Amlan Kundu
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Acquiring foreign language from degraded hardcopy documents is of interest to military and border control applications. Bi-tonal image scans are desirable because file size is small. However, the nature of hardcopy degradations and the scanner or image enhancement software capabilities used directly affect the quality of the captured image and the extent of language acquisition. We applied a collection of manual treatments to hardcopy Arabic documents to develop a corpus of bi-tonal images. We then used this corpus in an exploratory study to derive conclusions about how bi-tonal images could be enhanced. This paper discusses the manually degraded Arabic document corpus, the image enhancement study, and the significant optical character recognition (OCR) improvements obtained with simple scanner driver adjustments.

Paper Details

Date Published: 25 May 2005
PDF: 9 pages
Proc. SPIE 5817, Visual Information Processing XIV, (25 May 2005); doi: 10.1117/12.606447
Show Author Affiliations
Paul Herceg, The MITRE Corp. (United States)
Benjamin Huyck, The MITRE Corp. (United States)
Christopher Johnson, The MITRE Corp. (United States)
Linda Van Guilder, The MITRE Corp. (United States)
Amlan Kundu, The MITRE Corp. (United States)

Published in SPIE Proceedings Vol. 5817:
Visual Information Processing XIV
Zia-ur Rahman; Robert A. Schowengerdt; Stephen E. Reichenbach, Editor(s)

© SPIE. Terms of Use
Back to Top