Systems and methods for identifying a document with forged information

• US 7,706,580 B2
• Filed: 12/19/2003
• Issued: 04/27/2010
• Est. Priority Date: 03/15/2002
• Status: Active Grant

• Current Assignee: Computer Sciences Corporation (DXC Technology Company)
• Original Assignee: Computer Sciences Corporation (DXC Technology Company)
• Inventors: Mason, James G., Blinova, Katerina, Berkhuysen, Johan Willem Piere, Shridhar, Malayappan, Nugmanov, Babur, Bakker, Ronny, Houle, Gilles
• Primary Examiner(s): Bali; Vikkram
• Assistant Examiner(s): Zeilberger; Daniel
• Application Number:
  US10/741,157
• Publication Number:
  US 20040237047A1
• Time in Patent Office: 2,321 Days
• Field of Search: 382/119, 382/120, 382/121, 382/122, 382/123, 348/161
• US Class Current: 382/119
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/133 Evaluation of quality of the acquired characters
• CPC Class Codes: G06V 30/274 Syntactic or semantic context, e.g. balancing
• CPC Class Codes: G06V 40/30 Writer recognition; Reading and verifying signatures

Abstract

Methods and systems are provided for analyzing and assessing documents using a writing profile for documents, such as a payment instrument. A method may include providing a document to a computer system. In an embodiment, the method may further include comparing writing in information fields of the document to at least one forger writing profile representation. In some embodiments, at least one forger writing profile representation may be obtained from at least one information field of at least one document that includes forged information. In an embodiment, the document may be identified as a document including forged information from an approximate match of at least one forger writing profile representation with writing in the document.

Claims

1. A method of identifying a document comprising forged information using a computer system, comprising:

providing the document to the computer system, wherein the document provided to the computer system comprises one or more signature information fields and one or more non-signature information fields;

the computer system comparing handwriting in at least two of the information fields of the document provided to the computer system to one or more forger writing profile representations from one or more signature information fields of at least one document comprising forged information associated with a known forger and from one or more non-signature information fields of the at least one document comprising forged information associated with the known forger, wherein comparing handwriting in the at least two information fields comprises;

the computer system comparing handwriting in at least one of the signature information field of the document provided to the computer system to handwriting from at least one signature information field of the document comprising forged information associated with the known forger in at least one forger writing profile representation of the forger writing profile representations; and

the computer system comparing handwriting in at least one of the one or more non-signature information fields of the document provided to the computer system to handwriting in a corresponding non-signature information field from the at least one document comprising forged information associated with the known forger in the at least one forger writing profile representation, wherein comparing handwriting in at least one of the one or more non-signature information fields of the document to handwriting in a corresponding non-signature information field from the at least one document comprising forged information comprises;

the computer system comparing handwriting in an amount field of the document provided to the computer system to handwriting in a corresponding amount field from the at least one document comprising forged information associated with the known forger in the at least one forger writing profile representation;

the computer system determining one or more variances in handwriting within the amount field of the document provided to the computer system, wherein the one or more variances in handwriting within the amount field include written characters of the same type being written by at least two different individuals;

identifying the document provided to the computer system as a document comprising forged information from;

an approximate match of the handwriting associated with the known forger in the at least one forger writing profile representation with the handwriting in the document provided to the computer system; and

the determined one or more variances within the amount field including the written characters of the same type being written by at least two different individuals; and

assessing fraud in the document provided to the computer system using at least one of the comparisons, wherein evidence of fraud comprises the identification of the document provided to the computer system as a document comprising forged information based on the approximate match of the handwriting associated with the known forger with the handwriting in the document provided to the computer system and the determined one more variances within the amount field of the document provided to the computer system.

2. The method of claim 1, further comprising comparing writing in at least two of the information fields of the document provided to the computer system to at least two forger writing profile representations from at least one information field of at least one document comprising forged information.

3. The method of claim 1, further comprising comparing writing in at least two of the information fields of the document provided to the computer system to at least two forger writing profile representations from at least two information fields of at least one document comprising forged information.

4. The method of claim 1, wherein at least one forger writing profile representation is a member of a forger writing profile associated with a known forger, and further comprising identifying the forger of the document provided to the computer system from the forger writing profile if the document provided to the computer system is identified as forged.

5. The method of claim 1, wherein at least one forger writing profile representation comprises a forger identity and writing of the forger, and further comprising identifying a forger of the document provided to the computer system from the forger writing profile.

6. The method of claim 1, wherein providing the document to the computer system comprises providing images of the document to the computer system.

7. The method of claim 1, further comprising assessing fraud in the document provided to the computer system using at least two of the comparisons.

8. The method of claim 1, wherein comparing writing in the at least two information fields of the document provided to the computer system comprises comparing at least one characteristic of the writing.

9. The method of claim 1, wherein evidence of fraud comprises at least a portion of the handwriting in at least two of the information fields of the document provided to the computer system to approximately match at least one forger writing profile representation.

10. The method of claim 1, wherein assessing fraud in the document provided to the computer system comprises assessing fraud using at least two of the comparisons, wherein evidence of fraud comprises at least a portion of the writing in at least one of the information fields of the document provided to the computer system to approximately match at least one forger writing profile representation.

11. The method of claim 1, further comprising comparing writing in at least two of the information fields of the document provided to the computer system to at least one forger writing profile representation from at least two information fields of at least one forged document.

12. The method of claim 1, further comprising comparing writing in at least two of the information fields of the document provided to the computer system to at least two forger writing profile representations from at least one information field of at least one forged document.

13. The method of claim 1, further comprising comparing handwriting in at least two of the information fields of the document provided to the computer system to at least two forger writing profile representations from at least two information fields of at least one forged document.

14. The method of claim 1, wherein the document provided to the computer system is a payment instrument.

15. The method of claim 1, wherein providing the document to the computer system comprises obtaining images of writing of at least one information field.

16. The method of claim 1, further comprising creating a mathematical representation of the writing in at least one information field.

17. The method of claim 1, wherein the writing in the at least two information fields of the document provided to the computer system comprises at least one image.

18. The method of claim 1, wherein the writing in the at least two information fields of the document provided to the computer system comprises at least one type of written information.

19. The method of claim 1, wherein the writing in the at least two information fields of the document provided to the computer system comprises at least one type of written information, and wherein at least one type of written information comprises a word type.

20. The method of claim 1, wherein the writing in the at least two information fields of the document provided to the computer system comprises at least one type of written information, and wherein at least one type of written information comprises a character type.

21. The method of claim 1, wherein the writing in the at least two information fields of the document provided to the computer system comprises at least one global feature of the writing.

22. The method of claim 1, wherein the writing in the at least two information fields of the document provided to the computer system comprises at least one local feature of the writing.

23. The method of claim 1, wherein the writing in the at least two information fields of the document provided to the computer system comprises at least one syntax pattern.

24. The method of claim 1, wherein the writing in the at least two information fields of the document provided to the computer system comprises at least one lexicon name for at least one information field.

25. The method of claim 1, wherein at least one forger writing profile representation comprises at least one mathematical representation.

26. The method of claim 1, wherein at least one forger writing profile representation comprises at least one image.

27. The method of claim 1, wherein at least one forger writing profile representation comprises at least one writing variant of an example of at least one type of written information, and wherein at least one type of written information comprises a word type.

28. The method of claim 1, wherein at least one forger writing profile representation comprises at least one writing variant of an example of at least one type of written information, and wherein at least one type of written information comprises a character type.

29. The method of claim 1, wherein at least one forger writing profile representation comprises at least one writing variant of an example of at least one type of written information, and further comprising determining at least one of the variants with a cluster algorithm.

30. The method of claim 1, wherein at least one forger writing profile representation comprises at least one global characteristic of the writing.

31. The method of claim 1, wherein at least one forger writing profile representation comprises at least one local characteristic of the writing.

32. The method of claim 1, wherein at least one forger writing profile representation comprises at least one variant of a syntax pattern.

33. The method of claim 1, wherein at least one forger writing profile representation comprises at least one lexicon name for at least one information field.

34. The method of claim 1, wherein the document provided to the computer system is a payment instrument, wherein comparing handwriting in at least one of the one or more other non-signature information fields of the document provided to the computer system to handwriting in a corresponding non-signature information field from the at least one document comprising forged information associated with the known forger in the at least one forger writing profile representation comprises comparing handwriting in a payee field of the payment instrument to handwriting in a corresponding payee field from at least one payment instrument comprising forged information associated with the known forger in the at least one forger writing profile representation.

35. The method of claim 1, wherein the document provided to the computer system is a payment instrument, wherein comparing handwriting in at least one of the one or more other non-signature information fields of the document provided to the computer system to handwriting in a corresponding non-signature information field from the at least one document comprising forged information associated with the known forger in the at least one forger writing profile representation comprises comparing handwriting in a date field of the payment instrument to handwriting in a corresponding date field from at least one payment instrument comprising forged information associated with the known forger in the at least one forger writing profile representation.

36. The method of claim 1, wherein the document provided to the computer system is a payment instrument, wherein one of the information fields for which handwriting is compared is a signature field of the payment instrument, wherein at least one of the one or more non-signature information fields of the document provided to the computer system for which writing is compared is a non-signature field of the payment instrument.

37. The method of claim 1, further comprising:

analyzing a correlation of information between an amount field in the document provided to the computer system and a payee field in the document provided to the computer system, wherein analyzing a correlation of information between the amount field in the document provided to the computer system and a payee field in the document provided to the computer system comprising analyzing the amount field of the document provided to the computer system based on what information is in the payee field of the document provided to the computer system.

38. The method of claim 1, wherein the compared amount field is a courtesy amount field, wherein a portion of the written characters in the courtesy amount field are assessed to be written by the known forger based on a comparison between handwriting in the courtesy amount field of the document provided to the computer system to handwriting in the corresponding amount field from the at least one document comprising forged information.

39. The method of claim 1, wherein the compared amount field is a courtesy amount field, wherein determining variances in handwriting within the amount field of the document provided to the computer system comprises determining that, in the courtesy amount field of the document provided to the computer system, at least one of the written characters of the same type is a number added by alteration by a forger.

40. The method of claim 1, further comprising comparing pre-printed information in the document provided to the computer system against at least one stock profile representation, and assessing potential fraud based on the comparison of the pre-printed information against the stock profile representation.

41. The method of claim 1, wherein assessing fraud in the document provided to the computer system using at least one of the comparisons comprises performing two or more different fraud tests based on the comparison of one of the information fields, and assigning a different fraud weight to at least one of the fraud tests than to at least one other of the fraud tests.

42. The method of claim 41, wherein one of the fraud tests involves a courtesy amount field, wherein the fraud test relating to the courtesy amount field is weighted higher than at least one other fraud test.

43. The method of claim 41, wherein one of the fraud tests involves a legal amount field, wherein the fraud test relating to the legal amount field is weighted higher than at least one other fraud test.

44. The method of claim 1, wherein at least one of the non-signature field information fields of the document provided to the computer system is a payee field, the method further comprising determining whether a name in the payee field of the document matches at least one payee name from a lexicon of suspicious payee names, wherein a suspicious payee name is a payee name frequently involved in transactions with high fraud risk.

45. The method of claim 1, further comprising analyzing a correlation of information between an amount field in the document provided to the computer system and a payee field in the document provided to the computer system wherein analyzing the correlation of information between at least two of the non-signature information fields comprises reading information from at least one cross-correlation table.

46. A system, comprising:

a CPU;

a data memory coupled to the CPU; and

a system memory coupled to the CPU, wherein the system memory is configured to store one or more computer programs executable by the CPU, and wherein the computer programs are executable to implement a method for identifying a document comprising forged information, the method comprising;

providing the document to the computer system, wherein the document provided to the computer system comprises one or more signature information fields and one or more non-signature information fields;

comparing handwriting in at least two of the information fields of the document provided to the computer system to one or more forger writing profile representations from one or more information fields of at least one document comprising forged information associated with a known forger and from one or more non-signature information fields of the at least one document comprising forged information associated with the known forger, wherein comparing writing in the at least two information fields comprises;

comparing handwriting in at least one of the signature information fields of the document provided to the computer system to handwriting from at least one signature information field of the document comprising forged information associated with the known forger in at least one forger writing profile representation of the forger writing profile representations; and

comparing handwriting in at least one of the one or more non-signature information fields of the document provided to the computer system to handwriting in a corresponding non-signature information field from the at least one document comprising forged information associated with the known forger in the at least one forger writing profile representation, wherein comparing handwriting in at least one of the one or more non-signature information fields of the document to handwriting in a corresponding non-signature information field from the at least one document comprising forged information comprises;

comparing handwriting in an amount field of the document provided to the computer system to handwriting in a corresponding amount field from the at least one document comprising forged information associated with the known forger in the at least one forger writing profile representation;

determining one or more variances in handwriting within the amount field of the document provided to the computer system, wherein the one or more variances in handwriting within the amount field include written characters of the same type being written by at least two different individuals;

identifying the document provided to the computer system as a document comprising forged information from;

an approximate match of the handwriting associated with the known forger in the at least one forger writing profile representation with the handwriting in the document provided to the computer system; and

the determined one or more variances within the amount field including the written characters of the same type being written by at least two different individuals; and

assessing fraud in the document provided to the computer system using at least one of the comparisons, wherein evidence of fraud comprises the identification of the document provided to the computer system as a document comprising forged information based on the approximate match of the an approximate match of the handwriting associated with the known forger in the at least one forger writing profile representation with the handwriting in the document provided to the computer system; and

the determined one or more variances within the amount field including the written characters of the same type being written by at least two different individuals; and

assessing fraud in the document provided to the computer system using at least one of the comparisons, wherein evidence of fraud comprises the identification of the document provided to the computer system as a document comprising forged information based on the approximate match of the handwriting associated with the known forger with the handwriting in the document provided to the computer system and the determined one more variances within the amount field of the document provided to the computer system.

47. The system of claim 46, wherein at least one forger writing profile representation is a member of a forger writing profile associated with a known forger, and further comprising identifying the forger of the document from the forger writing profile if the document is identified as forged.

48. A tangible, computer readable storage medium comprising program instructions stored thereon, wherein the program instructions are computer-executable to implement a method for identifying a document comprising forged information, the method comprising:

providing the document to the computer system, wherein the document provided to the computer system comprises one or more signature information fields and one or more non-signature information fields;

comparing handwriting in at least two of the information fields of the document provided to the computer system to one or more forger writing profile representations from one or more signature information field of at least one document comprising forged information associated with a known forger and from one or more non-signature information fields of the at least one document comprising forged information associated with the known forger, wherein comparing writing in the at least two information fields comprises;

comparing handwriting in at least one of the signature information fields of the document provided to the computer system to handwriting from at least one signature information field of the document comprising forged information associated with the known forger in at least one forger writing profile representation of the forger writing profile representations; and

comparing handwriting in at least one of the one or more non-signature information fields of the document provided to the computer system to handwriting in a corresponding non-signature information field from the at least one document comprising forged information associated with the known forger in the at least one forger writing profile representation, wherein comparing handwriting in at least one of the one or more non-signature information fields of the document to handwriting in a corresponding non-signature information field from the at least one document comprising forged information comprises;

comparing handwriting in an amount field of the document provided to the computer system to handwriting in a corresponding amount field from the at least one document comprising forged information associated with the known forger in the at least one forger writing profile representation;

determining one or more variances in handwriting within the amount field of the document provided to the computer system, wherein the one or more variances in handwriting within the amount field include written characters of the same type being written by at least two different individuals;

identifying the document as a document comprising forged information from;

an approximate match of the handwriting associated with the known forger in the at least one forger writing profile representation with the handwriting in the document provided to the computer system; and

the determined one or more variances within the amount field including the written characters of the same type being written by at least two different individuals; and

assessing fraud in the document provided to the computer system using at least one of the comparisons, wherein evidence of fraud comprises the identification of the document provided to the computer system as a document comprising forged information based on the approximate match of the handwriting associated with the known forger with the handwriting in the document provided to the computer system and the determined one more variances within the amount field of the document provided to the computer system.