Method and apparatus for extracting information from forms

• US 5,010,580 A
• Filed: 08/25/1989
• Issued: 04/23/1991
• Est. Priority Date: 08/25/1989
• Status: Expired due to Term

• Current Assignee: Hewlett-Packard Company (HP Inc.)
• Original Assignee: Hewlett-Packard Company (HP Inc.)
• Inventors: Vincent, Kent D., Jamp, Rueiming
• Primary Examiner(s): Moore, David K.
• Assistant Examiner(s): Santos, Daniel
• Application Number:
  US07/398,677
• Time in Patent Office: 606 Days
• Field of Search: 382/17, 382/61, 250/226, 358/76, 358/78
• US Class Current: 382/163
• CPC Class Codes: G06V 10/143 Sensing or illuminating at different wavelengths

Abstract

A system for extracting handwritten or typed information from forms that have been printed in colors other than the color of the handwritten or typed information. The information extraction system includes a detector for detecting color values for scanned pixel locations on a printed form; a comparator for comparing the color values with reference color values; an identifier for identifying ones of the scanned pixel locations that have color values that correspond to the reference color values; and an optical character recognition engine for receiving data regarding the identified locations.

Claims

1. A system for extracting handwritten and typed information of a predetermined color from a form printed in a different color, comprising:

scanner means for scanning pixel locations on a printed form;

detector means for detecting at least two color values for each scanned pixel location;

comparator means for comparing the detected color values with reference color values;

identifier means for identifying and distinguishing ones of the pixel locations on the printed form from adjacent pixel locations by comparison of their color values to the reference color values, which identifier means provides a binary output for each scanned pixel location such that one state of the binary output represents correspondence between the detected color values of the scanned pixel location and at least one of the reference color values and, thereby, indicates whether handwritten and typed information on the printed form is to be extracted from the form when the color of the printing on the form is substantially different from the color of the handwritten and typed information at adjacent locations on the printed form; and

an optical character recognition engine for receiving data regarding identified ones of the pixel locations.

2. A system according to claim 1, wherein scanned pixels having detected color components that equal or exceed said reference color components represent information which is to be extracted from said form.

3. A system according to claim 1, wherein scanned pixels having detected color components that are less than or equal to said reference color components represent information which is to be extracted from said form.

4. A system according to claim 1 further including means to identify form identity codes that have been printed on forms in a standard location.

5. A system according to claim 1 further including a threshold color map and means that automatically identify a form and its color scheme such that threshold values appropriate for the form can be automatically loaded into the threshold color map.

6. A system for extracting handwritten or typed information from forms that have been printed in one or more colors that are different than the color of the handwritten or typed information, comprising:

scanner means for scanning selected pixel locations on a printed form;

color convertor means comprising a dichroic filter means for detecting at least two color values for each scanned pixel location;

comparator means for comparing the detected color values with reference color values;

identifier means for providing a binary output for each scanned pixel location which distinguishes certain ones of the pixel locations on the printed form from adjacent pixel locations on the printed form, with one state of the binary output representing correspondence between the detected color values of the scanned pixel location and at least one of the reference color values to, thereby, identify whether handwritten and typed information should be extracted from the form by comparison of color values at the scanned pixel locations with the reference color values;

an optical character recognition engine for receiving data regarding identified ones of the pixel locations; and

threshold color map means that automatically identify a printed form and its color scheme.

7. A system according to claim 6 wherein scanned pixels having detected color components that equal or exceed said reference color components represent information which is to be extracted from said form.