BIOMETRIC AUTHENTICATION METHOD AND APPARATUS

• US 20160183812A1
• Filed: 10/15/2015
• Published: 06/30/2016
• Est. Priority Date: 12/24/2014
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: KIM, Sang-joon, ZHANG, Chao, FENG, Xuetao, LIU, Yang, BAE, Chisung
• Granted Patent: US 10,154,818 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: A61B 2560/0462 Apparatus with built-in sensors
• CPC Class Codes: A61B 2562/06 Arrangements of multiple sensors of different types
• CPC Class Codes: A61B 5/117 Identification of persons methods or arrangements for recognising patterns, e.g. fingerprints, G06F18/00, G06V40/00; identification of persons by analysing their voice or speech G10L17/00
• CPC Class Codes: A61B 5/7246 using correlation, e.g. template matching or determination of similarity
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: G07C 2209/14 With a sequence of inputs of different identification information
• CPC Class Codes: G07C 9/37 using biometric data, e.g. fingerprints, iris scans or voice recognition

Abstract

A biometric authentication apparatus and method are provided. The biometric authentication apparatus may obtain at least two types of biosignals, extract a same type of physiological features from each of the biosignals, and determine whether the biosignals are generated from a same living body based on the extracted same type of physiological features.

Claims

1. A biometric authentication method comprising:

extracting a same physiological feature from each of at least two different type biosignals; and

determining whether the biosignals are generated from a same living body based on a comparison of the extracted physiological features.

2. The biometric authentication method of claim 1, further comprising capturing each of the biosignals at a same time using one or more sensors configured to capture biometric features from a body.

3. The biometric authentication method of claim 2, wherein the capturing of the biosignals includes controlling the one or more sensors to measure the biosignals for a predetermined period of time.

4. The biometric authentication method of claim 1, wherein the biosignals are different types of biosignals from among a human facial image biosignal type, a fingerprint image biosignal type, a palm print image biosignal type, a blood vessel image biosignal type, an iris image biosignal type, a retinal image biosignal type, an ECG signal biosignal type, an EEG signal biosignal type, a photoplethysmogram (PPG) signal biosignal type, a blood pressure signal biosignal type, a heart sound signal biosignal type, an electromagnetic signal associated with a human body biosignal type, a signal associated with a movement of chest or abdominal muscles biosignal type, and a conductive signal of the human body biosignal type.

5. The biometric authentication method of claim 1, wherein the extracting of the physiological features includes extracting, for each of the biosignals, a first same physiological feature type from among a heartbeat physiological feature type, a respiration physiological feature type, a blood pressure physiological feature type, and a body temperature physiological feature type.

6. The biometric authentication method of claim 5, further comprising extracting an other same physiological feature from each of the biosignals, including extracting, from each of the biosignals, a second same physiological feature type from among the heartbeat physiological feature type, the respiration physiological feature type, the blood pressure physiological feature type, and the body temperature physiological feature type, different from the first same physiological feature type.

7. The biometric authentication method of claim 6, wherein the determining of whether the biosignals are generated from the same live body comprises:

calculating a first consistent level representing evaluated consistencies between the extracted physiological features;

calculating a second consistent level representing evaluated consistencies between the extracted other physiological features; and

determining whether the biosignals are generated from the same living body based on both the first and second consistent levels.

8. The biometric authentication method of claim 1, wherein the determining of whether the biosignals are generated from the same live body comprises:

calculating a consistent level representing evaluated consistencies between the extracted physiological features based on properties of the extracted physiological features; and

determining whether the biosignals are generated from the same living body based on the consistent level satisfying a preset condition.

9. The biometric authentication method of claim 8, wherein the determining of whether the biosignals are generated from the same living body based on the consistent level satisfying the preset condition comprises determining whether the consistent level satisfies the preset condition based on at least one of a consistent level between a same first physiological feature extracted from each of the biosignals, collected from a body at a same time, and a consistent level between a same second physiological feature extracted from at least two different types of other biosignals collected from one or more other bodies or collected at different times, where the first physiological feature is different from the second physiological feature.

10. The biometric authentication method of claim 8, wherein the calculating of the consistent level comprises calculating the consistent level based on an evaluated relationship between a waveform of each of the extracted physiological features and a collection time in which a predetermined physiological event is collected.

11. The biometric authentication method of claim 10, wherein the physiological event corresponds to a peak or a valley in at least one of the waveforms,

wherein the consistent level represents a difference or a similarity, and

wherein the difference is a variance of a deviation of the collection time corresponding to the peak or the valley, and the similarity is a reciprocal of the difference.

12. The biometric authentication method of claim 8, wherein the calculating of the consistent level comprises calculating the consistent level using a learner trained based on input physiological feature data and a preset consistent level.

13. The biometric authentication method of claim 8, wherein the properties of the extracted physiological features include at least one of a time domain property, a frequency domain property, and a statistical property of the extracted physiological features.

14. The biometric authentication method of claim 13, wherein the time domain property comprises at least one of a moment at which a predetermined physiological event occurs in the extracted physiological features, a moment at which the physiological event changes, a duration of the physiological event, and a signal waveform of each of the extracted physiological features, and

wherein the frequency domain property comprises a signal frequency spectrum distribution or a signal frequency of each of the extracted physiological features.

15. The biometric authentication method of claim 1, wherein the determining of whether the biosignals are generated from the same living body comprises:

classifying the biosignals using a classifier based on one or more properties of the extracted physiological features, the classifier using a first type of samples acquired from first biosignals collected at a same time from a same body and a second type of samples acquired from second biosignals collected from one or more other bodies or collected at different times; and

determining whether the biosignals are generated from the same living body based on a result of the classifying.

16. The biometric authentication method of claim 15, wherein classifying of the biosignals is based on inputting to the classifier the one or more properties of the extracted physiological features, wherein the one or more properties are at least one of a time domain property, a frequency domain property, and a statistical property of the extracted physiological features.

17. The biometric authentication method of claim 16, wherein the time domain property comprises at least one of a moment at which a predetermined physiological event occurs in the extracted physiological features, a moment at which the physiological event changes, a duration of the physiological event, and a signal waveform of each of the extracted physiological features, and

wherein the frequency domain property comprises a signal frequency spectrum distribution or a signal frequency of each of the extracted physiological features.

18. The biometric authentication method of claim 1, wherein the extracted physiological features are physiological features that vary over time for a living body.

19. The biometric authentication method of claim 1, wherein the extracted physiological features comprise a select one of a heartbeat, a respiration, a blood pressure and a body temperature.

20. The biometric authentication method of claim 1, further comprising performing an authentication or identification of an identity of the living body based on the biosignals being determined to be from the same living body.

21. The biometric authentication method of claim 20, wherein the performing the authentication or identification comprises:

determining whether identity feature information extracted from at least one of the biosignals or an other biosignal from the living body matches registered identity feature information; and

performing the authentication or identification of the identity of the living body based on a result of the determining of whether the extracted identity feature information matches the registered identity feature information.

22. The biometric authentication method of claim 21, wherein the extracted identity feature information comprises at least one of a human facial image, a fingerprint image, a palm print image, a blood vessel image, an iris image, a retinal image, an audio signal, a gait feature, a feature of a signature or handwriting, an electrocardiogram (ECG) signal, and an electroencephalogram (EEG) signal.

23. A non-transitory recording medium comprising processor readable code to control at least one processing device to implement the method of claim 1.

24. A biometric authentication apparatus comprising:

a processor configured to extract a same physiological feature from each of at least two different type biosignals and to determine whether the biosignals are generated from a same living body based on a comparison of the extracted physiological features.

25. The biometric authentication apparatus of claim 24, further comprising one or more sensors configured to capture biometric features from a body, the processor being further configured to control a capturing of each of the biosignals at a same time using the one or more sensors.

26. The biometric authentication apparatus of claim 25, wherein the processor is configured to control the one or more sensors to measure the biosignals for a predetermined period of time.

27. The biometric authentication apparatus of claim 24, further comprising a communication device to communicate with one or more remote sensors configured to capture biometric features from a body, the processor being further configured to control or selectively receive a measuring of each of the biosignals, measured at a same time by the one or more sensors.

28. The biometric authentication apparatus of claim 24, wherein the biosignals are each analog or digital information respectively representing a different physical characteristic or behavioral characteristic of a body.

29. The biometric authentication apparatus of claim 28, wherein, to determine whether the biosignals are generated from the same live body, the processor is configured to evaluate consistencies between the extracted physiological features and determines whether the biosignals are generated from the same living body based on the evaluated consistencies.

30. The biometric authentication apparatus of claim 29, wherein the processor is configured to:

extract an other same physiological feature from the biosignals;

evaluate consistencies between the extracted other physiological features; and

determine whether the biosignals are generated from the same living body based on the evaluated consistencies between the extracted physiological features and the evaluated consistencies between the extracted other physiological features.

31. The biometric authentication apparatus of claim 29, wherein the processor is configured to:

determine whether identity feature information extracted from at least one of the biosignals or an other biosignal from the living body matches registered identity feature information; and

perform an authentication or identification of the identity of the living body based on the determination by the processor of whether the biosignals are generated from the same living body and the determination of whether the extracted identity feature information matches the registered identity feature information.