Transformed representation for fingerprint data with high recognition accuracy
The invention provides a computing device for processing fingerprint template data, wherein the computing device comprises modules configured for generating transformed fingerprint template data from original fingerprint minutiae template data, in which minutiae are represented at least by Cartesian or polar coordinates and an inherent minutiae direction, the computing device being configured to generate for each minutia of the original fingerprint minutiae data corresponding mega-vicinity data. The computing device may further comprise a comparator module being configured to compare two transformed fingerprint template data that comprises, for each minutia, mega-vicinity data of a first fingerprint and a second fingerprint. The computing device may further comprise a biometrics-secret binding module for protecting.
1. A system for transforming and binding biometric data of an individual, the system comprising:
a distinguishing feature extractor module that is configured for generating a distinguishing feature vector (f 1 ) from a biometric feature (b) of an individual, said distinguishing feature vector (f 1 ) being configured for differentiating individuals, and said distinguishing feature vector (f 1 ) being derived from minutia coordinate locations from within original minutiae data;
a hash function module that is configured for generating a cryptographic key (k) from first auxiliary data (ad 1 ), said first auxiliary data (ad 1 ) being based at least upon a secret of the individual;
a cryptographic module that is configured for inputting second auxiliary data (ad 2 ), and configured to perform a cryptographic function keyed by said cryptographic key (k), and configured for outputting results (cr); and
a modulo-P addition operation module that is configured for performing an exclusive-OR operation upon a combination of said results (cr) and said distinguishing feature vector (f 1 ) and outputting a result of said exclusive-OR operation as fused data (fd).
2. The system of claim 1 further including a block cipher module for encryption of said minutia coordinates locations.
3. The system of claim 2 wherein said block cipher module employs an AES-128 encryption algorithm.
4. The system of claim 3 wherein said minutia coordinate locations are derived from an ISO format minutiae triplet.
5. The system of claim 1 wherein said minutia coordinate locations are represented as polar or as Cartesian coordinates.
6. The system of claim 1 further including a quantization module for quantizing said minutia coordinate locations.
7. The system of claim 6 further including a robust feature extractor module that is configured for generating a robust feature vector (f 2 ) that is based upon said biometric feature (b) of the individual.
8. The system of claim 7 wherein said robust feature vector (f 2 ) is derived from counter information that is output from said quantization module.
9. The system of claim 8 wherein said counter information is input into a block cipher module, the block cipher module provided for encryption of said minutia coordinate locations.
10. A method for transforming and binding biometric data of an individual, the method comprising:
providing a distinguishing feature extractor module that is configured for generating a distinguishing feature vector (f 1 ) from a biometric feature (b) of an individual, said distinguishing feature vector (f 1 ) being configured for differentiating individuals and said distinguishing feature vector (f 1 ) being derived from minutia coordinate locations from within original minutiae data;
providing a hash function module that is configured for generating a cryptographic key (k) from first auxiliary data (ad 1 ), said first auxiliary data (ad 1 ) being based at least upon a secret of the individual;
providing a cryptographic module that is configured for inputting second auxiliary data (ad 2 ), and configured to perform a cryptographic function keyed by said cryptographic key (k), and configured for outputting results (cr); and
providing a modulo-P addition operation module that is configured for performing an exclusive-OR operation upon a combination of said results (cr) and said distinguishing feature vector (f 1 ) and outputting a result of said exclusive-OR operation as fused data (fd).
11. A computing device comprising at least one processing unit and a computer readable medium storing computer program instructions that when executed by the at least one processing unit implement a method for transforming and binding biometric data of an individual, the method comprising:
generating a distinguishing feature vector (f 1 ) from a biometric feature (b) of an individual, said distinguishing feature vector (f 1 ) being configured for differentiating individuals, and said distinguishing feature vector (f 1 ) being derived from minutia coordinate locations from within original minutiae data;
generating a cryptographic key (k) from first auxiliary data (ad 1 ), said first auxiliary data (ad 1 ) being based at least upon a secret of the individual;
inputting second auxiliary data (ad 2 ), performing a cryptographic function keyed by said cryptographic key (k), and outputting results (cr); and
performing an exclusive-OR operation upon a combination of said results (cr) and said distinguishing feature vector (f 1 ) and outputting a result of said exclusive-OR operation as fused data (fd).
12. The apparatus computing device of claim 11 wherein the method implemented further includes encrypting said minutia coordinate locations.
13. The computing device of claim 12 wherein an AES-128 encryption algorithm is employed for encrypting said minutia coordinate locations.
14. The computing device of claim 13 wherein said minutia coordinate locations are derived from an ISO format minutiae triplet.
15. The computing device of claim 11 wherein said minutia coordinates locations are represented as polar or as Cartesian coordinates.
16. The computing device of claim 11 wherein the method implemented further includes quantizing said minutia coordinate locations.
17. The computing device of claim 16 wherein the method implemented further includes generating a robust feature vector (f 2 ) that is based upon said biometric feature (b) of the individual.
18. The computing device of claim 17 wherein said robust feature vector (f 2 ) is derived from counter information that is output from said quantization module.
19. The computing device of claim 18 wherein said counter information is used as an input for encrypting said coordinates of minutia locations.