Form Design for High Accuracy Optical Character Recognition

Form Design for High Accuracy Optical Character Recognition Michael D. Garris, [email protected] Darrin L. Dimmick, [email protected] National Institute of Standards and Technology, Building 225, Room A216 Gaithersburg, Maryland 20899 Phone: (301)975-2928, FAX: (301)840-1357 Published in IEEE Transactions PAMI, June 1996. ABSTRACT Financial institutions, insurance companies, and government agencies are all aggressively pursuing the integration of automated forms processing into their everyday work flows. To use existing optical character recognition (OCR) technology, the forms that are currently hand-keyed will probably need to be redesigned. This paper presents some of the quantitative results generated by a comprehensive study of three versions of a redesigned tax form. Analyses show that using separately spaced bounding character boxes to represent fields provides superior machine readability over fields without character boxes, fields containing vertical ticks (combs), and fields with adjoining character boxes. It is also shown that character boxes containing two vertically stacked ovals cause writers much more difficulty to complete than do empty character boxes. The analyses also provide quantitative proof that writer idiosyncratic responses on forms are the major source of errors within the recognition system. These idiosyncracies (such as writers crossing out previously printed characters or writing over them) must be effectively handled in order improve recognition performance. This paper demonstrates how form design can help, and it provides empirical data to support some of the rules-of-thumb by measuring the impact specific changes to a form have on machine readability and on the writer.