Decision criteria for automated form population

• US 7,849,398 B2
• Filed: 04/26/2007
• Issued: 12/07/2010
• Est. Priority Date: 04/26/2007
• Status: Expired due to Fees

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Dabet, Sebastien, Poirier, Hervé, Bressan, Marco
• Primary Examiner(s): Hong; Stephen S
• Assistant Examiner(s): YANG, I CHAN
• Application Number:
  US11/789,823
• Publication Number:
  US 20080267505A1
• Time in Patent Office: 1,321 Days
• Field of Search: 715/221, 715/224, 715/226, 382/309, 382/310
• US Class Current: 715/226
• CPC Class Codes: G06F 40/174 Form filling; Merging
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/1444 Selective acquisition, locating or processing of specific regions, e.g. highlighted text, fiducial marks or predetermined fields
• CPC Class Codes: G06V 30/412 Layout analysis of documents structured with printed lines or input boxes, e.g. business forms or tables

Abstract

A method is provided for selecting fields of an electronic form for automatic population with candidate text segments. The candidate text segments can be obtained by capturing an image of a document, applying optical character recognition to the captured image to identify textual content, and tagging candidate text segments in the textual content for fields of the form. The method includes, for each of a plurality of fields of the form, computing a field exclusion function based on at least one parameter selected from a text length parameter, an optical character recognition error rate, a tagging error rate, and a field relevance parameter; and determining whether to select the field for automatic population based on the computed field exclusion function.

Claims

1. A method for selecting fields of an electronic form for automatic population with candidate text segments that are obtained by capturing an image of a document, applying optical character recognition to the captured image to identify textual content, and tagging candidate text segments in the textual content for fields of the form, the method comprising:

for each of a plurality of fields of an electronic form, estimating a manual entry time and a manual correction time for the field, wherein the estimated manual entry time is an estimated time period for a user to enter a text segment into the field without automatic population and the estimated manual correction time is an estimated time period for a user to correct a text segment in the field after automatic population;

computing a field exclusion function based on the estimated manual entry time and the estimated manual correction time, the computation of the field exclusion function further based on at least one parameter selected from;

a text length parameter,

an optical character recognition error rate,

a computed tagging error rate, and

a field relevance parameter which has been assigned to a respective field; and

determining whether to select the field for automatic population based on the computed field exclusion function.

2. The method of claim 1, wherein the field exclusion function is also based on the field relevance parameter.

3. The method of claim 1, further comprising:

for each of a plurality of fields of the form, assigning a relevance parameter to the field based on whether filling of the field is mandatory or not and wherein the field exclusion function is based on the assigned relevance parameter of the field.

4. The method of claim 1, wherein the optical character recognition error rate is estimated from information provided by an optical character recognition device.

5. The method of claim 1, further comprising for each of a plurality of the fields of the form, assigning an image capture parameter which accounts for image quality attributes of a device used in capturing the image, and wherein the field exclusion function is based on the image capture parameter.

6. The method of claim 1, further comprising for each of a plurality of the fields of the form, assigning a text attributes parameter based on text attributes and wherein the field exclusion function is based on the text attributes parameter.

7. The method of claim 1, further comprising for each of a plurality of the fields of the form, assigning an interface attributes parameter based on manual insertion and correction times for a selected user interface, and wherein the field exclusion function is based on the interface attributes parameter.

8. The method of claim 1, wherein the text length parameter for the field is based on an average length of text segments for that field obtained from a representative sample of documents.

9. The method of claim 1, wherein at least a first of the fields of the form is assigned a relevance parameter which differs from a relevance parameter of a second field of the form.

10. The method of claim 1, wherein the computing of the field exclusion function is performed prior to the tagging of candidate text segments in the textual content for fields of the form.

11. A method for populating a form comprising:

capturing an image of a physical document;

applying optical character recognition to the captured image to identify textual content;

tagging candidate text segments in the textual content for fields of the form;

for each of a plurality of the fields of the form, estimating a manual entry time and a manual correction time for the field, wherein the estimated manual entry time is an estimated time period for a user to enter a text segment into the field without automatic population and the estimated manual correction time is an estimated time period for a user to correct a text segment in the field after automatic population; and

automatically populating a field of the form with the tagged candidate text segment if the field is designated as an automatically populated field, otherwise leaving the field blank, the designation of the field depending on the estimated manual entry time and manual correction time and at least one of;

a text length parameter,

a predetermined tagging error rate,

an optical character recognition error rate, and

a field relevance parameter which has been previously assigned to that field.

12. The method of claim 11, further comprising:

for each of a plurality of fields of the form;

computing a field exclusion function which is based on the estimated manual entry time and manual correction time, the text length parameter, tagging error rate, optical character recognition error rate, and, optionally, the field relevance parameter, and

determining whether to automatically populate the field of the form based on the field exclusion function.

13. The method of claim 12, further comprising:

for each of a plurality of fields of the form, assigning a relevance parameter to the field and wherein the field exclusion function is based on the assigned relevance parameter of the field.

14. The method of claim 12, further comprising, establishing a tagging error rate and wherein the field exclusion function is a function of the tagging error rate.

15. The method of claim 12, further comprising, establishing an optical character recognition error rate and wherein the field exclusion function is a function of the optical character recognition error rate.

16. The method of claim 12, further comprising, at least one of:

computing an image capture attributes parameter based on image quality attributes of the captured image, and wherein the field exclusion function is a function of the image capture attributes parameter;

computing a text attributes parameter based on text attributes of the captured image, and wherein the field exclusion function is a function of the text attributes parameter;

computing an interface attributes parameter based on manual insertion and correction times for a selected user interface, and wherein the field exclusion function is a function of the interface attributes parameter.

17. The method of claim 11, further comprising identifying fields to the user for manual population, which are determined not to be automatically populated.

18. The method of claim 11, wherein the physical document comprises a business card.

19. An apparatus for populating an electronic form including a processor which performs the method of claim 11 and a graphical user interface which displays a form to be populated, the user interface displaying automatically populated fields for which a determination has been made by the processor to automatically populate the field.

20. A computer program product comprising a tangible computer-readable recording medium encoding instructions, which when executed on a computer causes the computer to perform the method of claim 11.

21. An apparatus for populating a form comprising:

memory which holds instructions for;

an optical character recognition engine configured to identify textual content in a captured image of a hardcopy document;

a tagging module which tags candidate text segments in the textual content for population of fields of a form to be populated; and

a field exclusion module which for each of a plurality of fields of the form, designates the field as a manual entry field or an automatically populated field, based on a field exclusion parameter, and automatically populates the fields designated as automatically populated fields, leaving the manually populated fields blank, the field exclusion parameter being a function of a manual entry time and a manual correction time for a respective one of the plurality of fields, and at least one of a predetermined tagging error rate, an optical character recognition error rate, and a field relevance parameter which has been assigned to the field based on whether the field is mandatory or not, wherein the estimated manual entry time is an estimated time period for a user to enter a text segment into the field without automatic population and the estimated manual correction time is an estimated time period for a user to correct a text segment in the field after automatic population; and

a processor which executes the instructions.

22. The apparatus of claim 21, further comprising at least one of:

a tagging error rate module which determines a tagging error for fields of the form; and

an optical character recognition error rate module which determines an optical character recognition error rate for fields of the form.

23. The apparatus of claim 21, further comprising a graphical user interface configured for displaying the form with those fields automatically populated which have been designated by the field exclusion module as automatically populated fields.

24. The apparatus of claim 21, further comprising an image capture device which generates the captured image.

25. A graphical user interface configured to display a form in which fields are automatically designated as manually populated or automatically populated based on:

a determination of manual correction and manual entry times for the field, wherein the estimated manual entry time is an estimated time period for a user to enter a text segment into the field without automatic population and the estimated manual correction time is an estimated time period for a user to correct a text segment in the field after automatic population,

average length of content of the field, and

a relevance of the field;

the graphical user interface configured for populating those fields designated as automatically populated with candidate text segments derived from a captured image of a document and leaving fields blank that are designated as manually populated.