AUTOMATED GENERATION OF FORM DEFINITIONS FROM HARD-COPY FORMS

摘要

A computer-implemented method for form generation includes capturing an image of a hard- copy form, and automatically processing the image to identify form fields in the image and text characters associated with each of the form fields. Geometrical coordinates of the form fields that define respective filling areas for entry of information into the fields are determined. Optical character recognition (OCR) is applied to the text characters in order to ident ify form field names. Associations are determined between the form field names and object names of corresponding data objects. The geometrical coordinates of the filling areas of the form fields are combined with the object names of the data objects corresponding to the form fields to generate a form definition.