Method and apparatus for converting documents between paper medium and electronic media using a user profile

• US 5,832,100 A
• Filed: 05/03/1993
• Issued: 11/03/1998
• Est. Priority Date: 08/30/1991
• Status: Expired due to Fees

• Current Assignee: Computer Sciences Corporation (DXC Technology Company)
• Original Assignee: TRW Limited (Zeppelin - Stiftung der Stadt Friedrichshafen)
• Inventors: Lawton, Stephen R., Warfel, George H. Jr.
• Primary Examiner(s): Johns, Andrew W.
• Application Number:
  US08/057,206
• Time in Patent Office: 2,010 Days
• Field of Search: 382/1, 382/7, 382/61, 382/100, 382/137, 382/140, 382/317, 364/401, 364/705.05, 364/705.06, 358/403, 395/925, 395/934, 705/1
• US Class Current: 382/100
• CPC Class Codes: G06F 18/00 Pattern recognition
• 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

Abstract

An apparatus and method for providing access to computerized information services for users unequipped with computers, particularly information services related to commercial transactions. Facsimile images of hardcopy forms are used as means for conveying transactional information.

Each facsimile image contains character and data fields with information for a particular trade transaction. The system captures the facsimile image in bitmap form and converts the character and data fields into computer readable form. The system enhances each image and performs several levels of feature extraction, including character recognition techniques. To efficiently and automatically convert great document volumes, the system optimizes database lookup and human operator manual keying to assist character recognition techniques in developing accurate coded text from the features of the facsimile image. The system must produce accurate coded text strings as the system subsequently converts the coded text string into files of coded text or EDI messages for transmission over telecommunications networks to independent computer systems. Unrecognized or inaccurate characters in the coded text are impermissible in order to form valid EDI messages or coded text files. Thus, validation of the coded text is an important part of the present invention. Additionally, user services provide additional processing of data received from independent computer systems into management information of additional business value to particular system users.

Claims

1. A method for processing a commercial transaction form to convert it from an electronic image of a form into a valid coded text string that can initiate automated data processing procedures, comprising the steps of:

receiving an encoded rasterized image of the form from a particular user, said rasterized image including a plurality of character and field data for conversion into the valid coded text string;

decoding said encoded image to produce an electronic representation of the form in a computer memory;

enhancing said electronic representation, said enhancing step comprising the steps of;

removing noise from said electronic representation by identifying and suppressing random noise bits;

deskewing said electronic representation; and

compensating for any velocity distortion in image formation that exceeds a predetermined threshold;

accessing a user profile associated with said particular user, said user profile including user/forms lexicons, user transaction contents and forms inventories, to extract and validate a first set of said plurality of character and field data from said electronic representation;

constructing a coded text string, said text string constructing step including the steps of;

keying manually a second set of said plurality of character and field data not extracted and validated as part of said first set of said plurality of character and field data; and

accessing said user profile, responsive to said first and second sets of said plurality of character and field data, to assemble the coded text from said image of form to produce an assembled text; and

validating, using said user profile, said assembled text to produce the valid coded text string.

2. The processing method of claim 1 further comprising the step of displaying said validated coded text string to said particular user for approval.

3. The processing method of claim 1 further comprising the step of:

transmitting said validated coded message to an independent computer system for initiation processing of a particular transaction identified in the form image.

4. The processing step of claim 3 further comprising the steps of:

receiving an inbound message from said independent computer system indicating the status or results of said processing; and

notifying said particular user of said status or results.

5. A method for processing a form to convert it from an electronic image of a form into a valid coded text string that automatically initiates data processing procedures, comprising the steps of:

capturing an image of the form received from a particular user;

extracting a plurality of character and field data candidate strings from said image; and

validating said character and field data candidate strings using a user profile associated with said particular user to produce the valid coded text string in a computer-readable format.

6. The processing method of claim 5 wherein said text validating step includes the step of producing a valid EDI message from the valid coded text string.

7. The processing method of claim 6 further comprising the step of:

transmitting said valid EDI message to an independent computer system for initiation of automated data processing of a particular transaction identified in the form image.

8. The processing method of claim 5 further comprising the step of transmitting the valid coded text string to said particular user for approval.

9. The form processing method of claim 8 further including the step of transmitting a facsimile message to said particular user including said validated coded text string.

10. The processing method of claim 5 further comprising the step of:

transmitting said validated coded text to an independent computer system for initiation of automated data processing of one or more particular transactions identified in the form image.

11. The processing method of claim 10 further comprising the steps of:

receiving an inbound EDI message from said independent computer system responsive to said automated data processing;

formatting said inbound EDI message to produce a message understandable by said particular user; and

transmitting said understandable message to said particular user.

12. The processing method of claim 11 further comprising the step of presenting information services to a user responsive to a plurality of said inbound EDI messages.

13. The processing method of claim 10 further comprising the steps of:

receiving an inbound coded text file from said independent computer system responsive to said automated data processing;

formatting said inbound coded text file to produce a message understandable by said particular user; and

transmitting said understandable message to said particular user.

14. The processing method of claim 13 further comprising the step of presenting information services to a user responsive to a plurality of said validated coded text files.

15. The form processing method of claim 5 wherein said form is a standardized document having blank areas and said particular user enters data onto said standardized document.

16. The form processing method of claim 5 wherein said step of capturing an image includes transmitting said form via a facsimile machine.

17. A system for producing valid coded text from data extracted from an image of a hardcopy of a commercial transaction form, the image including a plurality of characters represented by an electronic raster having a matrix of binary picture elements making up the plurality of characters wherein the plurality of characters includes information identifying a particular user and at least one type of transaction from a plurality of users and a plurality of transactions, comprising:

(a) image enhancement means for;

(i) deskewing the image when angular distortion of the image exceeds a predefined value; and

(ii) removing an axial distortion from the image introduced during conversion of the hardcopy to the electronic raster to produce an enhanced document image;

(b) character extraction means, coupled to said image enhancement means, for operating on said enhanced document image to identify and produce a first set of machine readable representations of the plurality of characters, said character extraction means including a database having the plurality of user identity and transaction data information stored and responsive to said first set of said plurality of characters to produce a second set of machine readable representations of the plurality of characters, said character extraction means further including means for keying manually a third set of the plurality of characters not in said first set and second set;

(c) assembly means, coupled to said character extraction means, for receiving said first, second and third sets of the plurality of characters to construct and validate coded text making a transaction;

(d) communication means, coupled to said assembly means, for transmitting to a host computer said valid coded text for subsequent automated data processing; and

(e) means, coupled to communication means, for delivering an output of said automated data processing to the particular user.

18. An apparatus for processing a form to convert it from an electronic image of a form into a valid coded text string that initiates automated data processing procedures, comprising:

means for receiving an encoded rasterized image of the form sent by a particular user, said rasterized image including a plurality of character and field data for conversion into the validated coded text;

means, coupled to said means for receiving, for decoding said encoded image to produce an electronic representation of the form in a memory of a computer;

means, coupled to said means for decoding, for enhancing said electronic representation, said means for enhancing comprising;

means for removing noise from said electronic representation by identifying and suppressing random noise bits;

means, coupled to said means for removing noise, for deskewing said electronic representation; and

means, coupled to said means for deskewing, for compensating for any velocity distortion in image formation that exceeds a predetermined threshold;

means, coupled to said means for enhancing, for accessing a user profile associated with said particular user, said user profile including user/forms lexicons, user transaction contents and forms inventories, to extract and validate a first set of said plurality of character and field data from said electronic representation;

means, coupled to said means for enhancing, for constructing a coded text string, said means for constructing comprising;

means for keying manually a second set of said plurality of character and field data not extracted and validated as part of said first set of said plurality of character and field data; and

means, coupled to said means for keying, for accessing said user profile, responsive to said first and second sets of said plurality of character and field data, to assemble said coded text string; and

means, coupled to said means for accessing, for validating said coded text string to produce the valid coded text string.

19. The form processing apparatus of claim 18 wherein said means for validating said coded text includes means for producing validated coded text.

20. The form processing apparatus of claim 19 further comprising:

means, coupled to said means for producing validated coded text, for transmitting said validated coded text to a computer for processing into output text.

21. The form processing apparatus of claim 20 further comprising:

means, coupled to said transmitting means, for receiving said output text from said computer; and

means, coupled to said means for receiving, for delivering said output text to said particular user.

22. The form processing apparatus of claim 21 further comprising:

means, coupled to said means for receiving, for processing said output text from said computer; and

means, coupled to means for receiving, for delivering a result of said processing to a user.

23. The form processing apparatus of claim 18 further comprising means, coupled to said means for validating said coded text, for displaying the valid coded text string to said particular user for approval.

24. An apparatus for processing a form to convert it from an electronic image of a form into a valid coded text string that can initiate a user service identified by the form, comprising:

means for capturing an image of the form received from a particular user;

means, coupled to said means for capturing, for extracting a plurality of character and field data candidate strings from said image;

means, coupled to said means for extracting, for validating, using a user profile associated with said particular user, said character and field data candidate text to produce the validated coded text in computer-readable form as output text;

means, coupled to said means for validating, for receiving and processing output text to produce a validated message; and

means, coupled to said means for receiving and processing output text, for providing user services responsive to said validated message.

25. The form processing apparatus of claim 24 wherein said means for validating includes means for producing a validated coded EDI message.

26. The form processing apparatus of claim 24 further comprising means, coupled to said means for validating, for transmitting said validated coded text string to said particular user for approval.

27. The form processing apparatus of claim 24 further comprising:

means, coupled to said means for validating, for transmitting said validated coded text to a computer for initiation of the processing identified in the form image.

28. The form processing apparatus of claim 27 further comprising:

means, coupled to said means for transmitting, for receiving processing results from said computer; and

means, coupled to said means for receiving, for transmitting to said particular user such results.

29. An apparatus for processing a form to convert it from an electronic image of a form into a valid coded text string that initiates automated data processing, comprising:

a telecommunications server including an image receiver and a computer gateway, for receiving an electronic image of a form from a user and for transmitting coded text strings;

an image and data server, coupled to said telecommunications server, for storing said electronic image and coded text;

an image processing server, coupled to said image and data server, for extracting character and field data candidate text from said image through use of a dynamic database containing transaction context information and user profile information associated with said user, wherein said dynamic database is updated responsive to transactions initiated by forms submitted from said user; and

a data processing server, coupled to said image processing server and to said telecommunications server, for;

confirming a correspondence between said character and field data candidate text and information contained within said image to produce a coded text string in computer readable form;

constructing a valid data file from said coded text string; and

passing said valid data file to said telecommunications server for transmission to an independent computer system to initiate processing identified within said image.

30. A method for initiating a commercial transaction, comprising the steps of:

entering a plurality of data into a form that identifies a particular commercial transaction;

thereafter

converting said form into a binary image;

thereafter

extracting a plurality of candidate strings from said binary image, said plurality of candidate strings including a first identification string identifying a particular user;

thereafter

accessing a user profile associated with said identification string, wherein said user profile is dynamically assembled from a prior use by said particular user;

thereafter

validating said plurality of candidate strings using said user profile to produce a valid computer-readable message identifying said particular user and said particular commercial transaction; and

thereafter

initiating said particular commercial transaction using said valid computer-readable message.

31. The commercial transaction initiating method of claim 30 further comprising the step of updating said user profile responsive to said valid computer-readable message.