Document authentication using extracted digital fingerprints
First Claim
1. A method comprising:
- acquiring digital image data of an image of at least a selected authentication region of a document, without printing or adding anything on to the document, and wherein the document has been previously forged or altered at least in the authentication region from its original state;
extracting at least one native feature from the acquired digital image data from the authentication region, wherein the at least one native feature describes physical characteristics of the authentication region without recognizing content such as alpha-numeric characters that may appear in the authentication region;
forming a feature vector to represent the at least one native feature in a compact form;
storing the feature vector in a database record as part of a digital fingerprint of the document in a forgery database for use in detecting forged or altered documents among a class of documents to which the forged or altered document belongs;
repeating the foregoing steps over additional documents that have been forged or altered to add, respectively, additional feature vectors in additional database records as part of additional digital fingerprints of the additional documents to the forgery database, the additional feature vectors representing corresponding native features in the additional documents;
acquiring second digital image data of a portion of a suspect document, wherein the portion of the suspect document corresponds to the authentication region used to build the forgery database;
extracting second features from the second digital image data;
forming a second feature vector to represent the second features in the compact form;
querying the forgery database using the second feature vector to obtain a result based on zero or more stored feature vectors that match the second feature vector, each stored feature vector that matches the second feature vector defining a fraud indicator match;
counting a number of the fraud indicator matches;
comparing the number of the fraud indicator matches to a predetermined threshold value to determine a confidence level that the suspect document is forged or altered;
generating a report of the determined confidence level based on the comparison; and
transmitting the report to a user interface.
3 Assignments
0 Petitions
Accused Products
Abstract
Methods and systems for detecting changes, alterations and forgeries in documents utilize digital fingerprinting. A set of documents of a given type that are known to be altered or forgeries are scanned and “fingerprinted” to build a forgery database. Suspect documents are scanned and the resulting fingerprint data is used to query the forgery database to look for matching records, although a match need not be exact. Matches may be counted and compared to a threshold number to assess a confidence that a document is forged or not. The fingerprinting may be directed to at least one authentication region of a document, and one or more features in a region may be incorporated into a fingerprint of the document.
195 Citations
34 Claims
-
1. A method comprising:
-
acquiring digital image data of an image of at least a selected authentication region of a document, without printing or adding anything on to the document, and wherein the document has been previously forged or altered at least in the authentication region from its original state; extracting at least one native feature from the acquired digital image data from the authentication region, wherein the at least one native feature describes physical characteristics of the authentication region without recognizing content such as alpha-numeric characters that may appear in the authentication region; forming a feature vector to represent the at least one native feature in a compact form; storing the feature vector in a database record as part of a digital fingerprint of the document in a forgery database for use in detecting forged or altered documents among a class of documents to which the forged or altered document belongs; repeating the foregoing steps over additional documents that have been forged or altered to add, respectively, additional feature vectors in additional database records as part of additional digital fingerprints of the additional documents to the forgery database, the additional feature vectors representing corresponding native features in the additional documents; acquiring second digital image data of a portion of a suspect document, wherein the portion of the suspect document corresponds to the authentication region used to build the forgery database; extracting second features from the second digital image data; forming a second feature vector to represent the second features in the compact form; querying the forgery database using the second feature vector to obtain a result based on zero or more stored feature vectors that match the second feature vector, each stored feature vector that matches the second feature vector defining a fraud indicator match; counting a number of the fraud indicator matches; comparing the number of the fraud indicator matches to a predetermined threshold value to determine a confidence level that the suspect document is forged or altered; generating a report of the determined confidence level based on the comparison; and transmitting the report to a user interface. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12)
-
-
13. A system comprising:
-
an image capture device to acquire digital image data of an image of at least a selected authentication region of a document, without printing or adding anything on to the document, and wherein the document has been previously forged or altered at least in the authentication region from its original state; and a processor arranged to access the digital image data, the processor coupled to a non-volatile memory storing machine-readable code executable by the processor, wherein the machine-readable code is arranged to cause the processor, upon execution, to; extract at least one native feature from the acquired digital image data from the authentication region, wherein the at least one native feature describes physical characteristics of the authentication region without recognizing content such as alpha-numeric characters that may appear in the authentication region; form a feature vector to represent the at least one native feature in a compact form; store the feature vector in a database record as part of a digital fingerprint of the document in a forgery database for use in detecting forged or altered documents among a class of documents to which the forged or altered document belongs; repeat the foregoing steps over additional documents that have been forged or altered to add, respectively, additional feature vectors in additional database records as part of additional digital fingerprints of the additional documents to the forgery database, the additional feature vectors representing corresponding native features in the additional documents; acquire second digital image data of a portion of a suspect document, wherein the portion of the suspect document corresponds to the authentication region used to build the forgery database; extract second features from the second digital image data; form a second feature vector to represent the second features in the compact form; query the forgery database using the second feature vector to obtain a result based on zero or more stored feature vectors that match the second feature vector, each stored feature vector that matches the second feature vector defining a fraud indicator match; count a number of the fraud indicator matches; compare the number of the fraud indicator matches to a predetermined threshold value to determine a confidence level that the suspect document is forged or altered; generate a report of the determined confidence level based on the comparison; and transmit the report to a user interface. - View Dependent Claims (14, 15, 16, 17, 18, 19, 20, 21, 22, 23)
-
-
24. A manufacture comprising a non-volatile, computer-readable memory, the memory storing instruction to cause a processor to:
-
access digital image data, the digital image data acquired from at least a selected authentication region of a document; extract at least one native feature from the acquired image data from the authentication region, wherein the native feature describes physical characteristics of the authentication region without recognizing content such as alpha-numeric characters that may appear in the authentication region; form a feature vector to represent the at least one native feature in a compact form; store the feature vector in a database record as part of a digital fingerprint of the document in a forgery database for use in detecting forged or altered documents among a class of documents to which the forged document belongs; repeat the foregoing steps over additional documents that have been forged or altered to add additional feature vectors to the forgery database, the feature vectors representing corresponding native features in the additional documents; acquire second digital image data of a portion of a suspect document, wherein the portion of the suspect document corresponds to the authentication region used to build the forgery database; extract second features from the second image data; form a second feature vector to represent the second features in a compact form; querying the forgery database using the second feature vector to obtain a result based on zero or more stored feature vectors that match the second feature vector, each stored feature vector that matches the second feature vector defining a defining a fraud indicator match; count a number of the fraud indicator matches; compare the number of the fraud indicator matches to a predetermined threshold value to determine a confidence level that the suspect document is forged or altered; generate a report of the determined confidence level based on the comparison; and transmit the report to a user interface. - View Dependent Claims (25, 26, 27, 28, 29, 30, 31, 32, 33, 34)
-
Specification