Fingerprint recognition method and apparatus

1. A fingerprint recognition method comprising:

generating an enrollment modified image by modifying a fingerprint image corresponding to a fingerprint to be enrolled;

extracting enrollment property information from the fingerprint image;

generating mapping information that maps the enrollment modified image to the enrollment property information;

storing the enrollment modified image and the enrollment property information;

matching an input modified image and the enrollment modified image;

dividing an overlapping region of the input modified image and the enrollment modified image into sub-regions;

calculating, for each of the sub-regions, histogram information of the input modified image relative to the enrollment modified image; and

determining a score corresponding to the matching based on the histogram information,

the modifying of the fingerprint image including converting a type of the fingerprint image to a modified type, the modified type being a local binary pattern (LBP) type, a histogram of oriented gradients (HOG) type, a Haar feature type, a modified census transform (MCT) type, or a geometrically warped type.

2. The method of claim 1 , wherein the modifying includes modifying the fingerprint image in at least one of a spatial domain and a frequency domain.

3. The method of claim 1 , wherein the extracting includes extracting enrollment phase information from the fingerprint image based on a frequency-domain transform scheme.

4. The method of claim 1 ,

wherein the matching the input modified image and the enrollment modified image comprises:

receiving an input fingerprint image;

extracting first property information from a query fingerprint included in the input fingerprint image;

generating the input modified image by modifying the input fingerprint image; and

matching the input modified image and the enrollment modified image based on the extracted first property information and the enrollment property information; and

wherein the determining comprises:

recognizing the query fingerprint based on a result of the matching.

5. The method of claim 4 , wherein the determining includes determining the score indicating a degree to which the input modified image matches the enrollment modified image in the overlapping region of the input modified image and the enrollment modified image.

6. The method of claim 5 , wherein the determining includes determining the score further based on at least one of

a similarity between the input modified image and the enrollment modified image, and

information associated with the input fingerprint image.

7. The method of claim 4 , wherein the matching includes determining at least one of translation information, rotation information, and scale information of the input modified image and the enrollment modified image, based on the first property information and the enrollment property information.

8. The method of claim 4 ,

wherein the extracting includes extracting phase information from the query fingerprint based on a frequency-domain transform scheme, and

wherein the matching includes matching the input modified image and the enrollment modified image based on a phase of correlation (POC) between the phase information and enrollment phase information associated with the enrollment fingerprint.

9. A non-transitory computer-readable medium comprising program code that, when executed by a processor, performs functions according to the method of claim 1 .

10. A fingerprint recognition apparatus comprising:

a sensor configured to receive a fingerprint image corresponding to a fingerprint to be enrolled;

a processor configured to generate an enrollment modified image by modifying the fingerprint image and extract enrollment property information from the fingerprint image; and

a memory configured to store the enrollment modified image and the enrollment property information such that the enrollment modified image and the enrollment property information are mapped,

wherein the processor is further configured such that the modifying includes converting a type of the fingerprint image to a modified type, the modified type being a local binary pattern (LBP) type, a histogram of oriented gradients (HOG) type, a Haar feature type,

wherein the processor is further configured to

match an input modified image and the enrollment modified image,

divide an overlapping region of the input modified image and the enrollment modified image into sub-regions,

calculate, for each of the sub-regions, histogram information of the input modified image relative to the enrollment modified image, and

determine a score corresponding to the matching based on the histogram information.

11. The apparatus of claim 10 , wherein the processor is configured to generate the enrollment modified image by performing modification in at least one of a spatial domain and a frequency domain.

12. The apparatus of claim 10 , wherein the processor is configured to extract enrollment phase information from the fingerprint image based on a frequency-domain transform scheme.

13. The apparatus of claim 10 ,

wherein the sensor is configured to receive an input fingerprint image, and

wherein the processor is configured to

extract first property information from a query fingerprint appearing in the input fingerprint image,

generate the input modified image by modifying the input fingerprint image,

match the input modified image and the enrollment modified image based on the extracted first property information and the enrollment property information, and

recognize the query fingerprint based on a result of the matching.

14. The apparatus of claim 13 , wherein the processor is configured to determine the score indicating a degree to which the input modified image matches the enrollment modified image in the overlapping region of the input modified image and the enrollment modified image.

15. The apparatus of claim 14 , wherein the processor is configured to determine a similarity in the overlapping region between the input modified image and the enrollment modified image.

16. The apparatus of claim 14 , wherein the processor is configured to determine the score further based on at least one of

a similarity between the input modified image and the enrollment modified image, and

information associated with the input fingerprint image.

17. The apparatus of claim 13 , wherein the processor is configured to determine at least one of translation information, rotation information, and scale information of the input modified image and the enrollment modified image, based on the first property information and the enrollment property information.

18. The apparatus of claim 13 , wherein the processor is configured to extract phase information from the query fingerprint based on a frequency-domain transform scheme, and match the input modified image and the enrollment modified image based on a phase of correlation (POC) between the phase information and enrollment phase information associated with the enrollment fingerprint.