Biometric identification systems and methods
An exemplary embodiment of the present invention provides a method of verifying an identity of a person-to-be-identified using biometric signature data. The method includes creating a face sample database based on biometric signature data from a plurality of individuals, calculating a feature database by extracting selected features of entries in the sample database, calculating positive samples by calculating a feature absolute value distance for a same position of any two different images from one person, calculating negative samples by calculating a feature absolute value distance for a same position of different people, calculating a key bin feature using a learning algorithm, calculating a classifier from the key bin feature for use in identifying and authenticating an acquired face image of a person-to-be-identified and identifying and authenticating the person-to-be-identified using the classifier and the acquired face image of the person-to-be-identified.
1. A method of identity verification using biometric signature data, comprising:
creating a face sample database based on a plurality of acquired face samples;
calculating a feature database by extracting selected features of entries in the face sample database;
calculating positive samples by calculating a feature absolute value distance for a same position of any two different images from one person;
calculating negative samples by calculating a feature absolute value distance for a same position of different people;
calculating a key bin feature using a learning algorithm, the key bin feature distinguishing each of the positive samples and negative samples;
calculating a classifier from the key bin feature for use in identifying and authenticating an acquired face image of a person-to-be-identified; and
identifying and authenticating the person-to-be-identified using the classifier and the acquired face image of the person-to-be-identified.
2. The method of claim 1 , wherein calculating a feature database comprises calculating at least one of local binary pattern features and local ternary pattern features from entries in the face sample database.
3. The method of claim 1 , wherein the classifier is one of a left eye classifier, right eye classifier, left eye coarse detection classifier, right eye coarse detection classifier or a left eye and right eye classifier.
4. The method of claim 1 , wherein the learning algorithm is an adaptive boosting learning algorithm.
5. The method of claim 1 , wherein the identifying and authenticating further comprises:
receiving a face image of the person;
extracting at least one feature from the face image; and
using the classifier and the extracted at least one feature to determine the identity of the person-to-be-identified.
6. The method of claim 1 , wherein the plurality of acquired face samples are acquired using an analog-to-digital infrared sensor.
7. A system for identity verification using biometric signature data, the system comprising:
a processor; and
a memory storing logical instructions that, when executed by the processor, are configured to:
create a face sample database based on a plurality of acquired face samples;
calculate a feature database by extracting selected features of entries in the face sample database;
calculate positive samples by calculating a feature absolute value distance for a same position of any two different images from one person;
calculate negative samples by calculating a feature absolute value distance for a same position of different people;
calculate a key bin feature using a learning algorithm, the key bin feature distinguishing each of the positive samples and negative samples;
calculate a classifier from the key bin feature for use in identifying and authenticating an acquired face image of a person-to-be-identified; and
identify and authenticate the person-to-be-identified using the classifier and the acquired face image of the person-to-be-identified.
8. The system of claim 7 , wherein the feature database is calculated by calculating at least one of local binary pattern features and local ternary pattern features from entries in the face sample database.
9. The system of claim 7 , wherein the classifier is one of a left eye classifier, right eye classifier, left eye coarse detection classifier, right eye coarse detection classifier or a left eye and right eye classifier.
10. The system of claim 7 , wherein the learning algorithm is an adaptive boosting learning algorithm.
11. The system of claim 7 , wherein the identify and authenticate a person further comprises:
receive a face image of the person;
extract least one feature from the face image; and
use the classifier and the extracted at least one feature to determine the identity of the person-to-be-identified.
12. The system of claim 7 , wherein the plurality of acquired face samples are acquired using an analog-to-digital infrared sensor.
13. A non-transitory computer-readable medium having instructions stored therein which, when executed by a processor, are configured to:
create a face sample database based on a plurality of acquired face samples;
calculate a feature database by extracting selected features of entries in the face sample database;
calculate positive samples by calculating a feature absolute value distance for a same position of any two different images from one person;
calculate negative samples by calculating a feature absolute value distance for a same position of different people;
calculate a key bin feature using a learning algorithm, the key bin feature distinguishing each of the positive samples and negative samples;
calculate a classifier from the key bin feature for use in identifying and authenticating an acquired face image of a person-to-be-identified; and
identify and authenticate the person-to-be-identified using the classifier and the acquired face image of the person-to-be-identified.
14. The non-transitory computer-readable medium of claim 13 , wherein the feature database is calculated by calculating at least one of local binary pattern features and local ternary pattern features from entries in the face sample database.
15. The non-transitory computer-readable medium of claim 13 , wherein the classifier is one of a left eye classifier, right eye classifier, left eye coarse detection classifier, right eye coarse detection classifier or a left eye and right eye classifier.
16. The non-transitory computer-readable medium of claim 13 , wherein the learning algorithm is an adaptive boosting learning algorithm.
17. The non-transitory computer-readable medium of claim 13 , wherein the identify and authenticate a person, further comprises:
receive a face image of the person;
extract least one feature from the face image; and
use the classifier and the extracted at least one feature to determine the identity of the person-to-be-identified.
18. The non-transitory computer-readable medium of claim 13 , wherein the plurality of acquired face samples are acquired using an analog-to-digital infrared sensor.