Optical Character Reader Techniques For Degraded Text

R. A. Gonsalves; P. S. Considine; A. A. Shea

doi:10.1117/12.955673

8 December 1977 Optical Character Reader Techniques For Degraded Text

R. A. Gonsalves, P. S. Considine, A. A. Shea

Proceedings Volume 0118, Optical Signal and Image Processing; (1977) https://doi.org/10.1117/12.955673
Event: 21st Annual Technical Symposium, 1977, San Diego, United States

Abstract

Commercially available OCR systems operate on high SNR text, use highly non-linear, serial processing techniques to discriminate between symbols, and achieve vanishingly small error rates. With degraded text, however, these discrimination techniques are no longer effective and an optimal OCR system must use classical signal detection techniques, namely, cross correlation of the text with stored symbols. Such a system using an optical correlator was first demonstrated in 1964 but has never been developed commercially, partially because no techniqe existed for the rapid generation and erasure of optical filters and because cross correlation yielded insufficient discrimination between symbols. The advent of real-time optical modulators such as the PROM may remove the first objection and we show in this paper that careful processing of the correlation data can remove the second. We review the theory (optimal detection with letter decoding based on language statistics) and present some examples.

Citation Download Citation

R. A. Gonsalves, P. S. Considine, and A. A. Shea "Optical Character Reader Techniques For Degraded Text", Proc. SPIE 0118, Optical Signal and Image Processing, (8 December 1977); https://doi.org/10.1117/12.955673

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available