BIOMETRIC IDENTIFICATION METHOD AND DEVICE USING ONE

• US 20170177961A1
• Filed: 12/22/2016
• Published: 06/22/2017
• Est. Priority Date: 12/22/2015
• Status: Active Grant

• Current Assignee: IDEMIA Identity and Security France SAS (Advent International Corporation)
• Original Assignee: Safran Identity & Security
• Inventors: RAGOT, Marcelin, BERTHIER, Mael, BOCKTAELS, Yves, BRINGER, Julien
• Granted Patent: US 10,489,667 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06V 40/1365 Matching; Classification
• CPC Class Codes: G06V 40/40 Spoof detection, e.g. liveness detection
• CPC Class Codes: G06V 40/70 Multimodal biometrics, e.g. combining information from different biometric modalities

Abstract

A biometric identification method includes the steps of comparing a candidate print with a reference print and of validating the identification according to a certain number of characteristics common to both prints and a predetermined validation threshold. The method includes the steps of creating at least one deceiving print and comparing same with a candidate print. An identification device designed for implementing this method is also provided.

Claims

1. A method for biometric recognition by comparing a candidate print with at least one reference print for determining a score depending on the similarity of the two prints and a predetermined validation threshold which the score will be compared with, the method comprising the following steps:

capturing a first candidate print;

comparing the first candidate print with the reference print and computing a score of the first candidate print for the reference print;

at least when the score is under the validation threshold, comparing the first candidate print with at least one first deceiving print, with said first deceiving print having been created for having characteristics enabling the first candidate print to obtain a score at least equal to a first predetermined threshold;

computing a score of the first candidate print for the first deceiving print.

2. The method according to claim 1, which comprises the following steps:

capturing a second candidate print;

comparing the second candidate print with the reference print and computing a score of the second candidate print for the reference print;

comparing the second candidate print with the first deceiving print and computing a score of the second candidate print for the first deceiving print;

if the score is under the validation threshold for the reference print and under the first predetermined threshold for the first deceiving print, generating a second deceiving print to have characteristics enabling the second candidate print to obtain a score at least equal to the first predetermined threshold.

3. The method according to claim 2, comprising the step of computing a score of the second candidate print for the second deceiving print.

4. The method according to claim 1, wherein, when the candidate print obtains a score above the first threshold with the deceiving print, the method comprises the step of triggering a securing action when the score is also above a second predetermined threshold above the first predetermined threshold.

5. The method according to claim 4, wherein the second predetermined threshold is between the first predetermined threshold and the validation threshold.

6. The method according to claim 4, wherein the securing action comprises the step of modifying the deceiving print for which the candidate print has a score above the second predetermined threshold.

7. The method according to claim 4, wherein the securing action comprises the step of aborting the identification.

8. The method according to claim 4, comprising a step of configuring, during which a securing action is selected from a list of predetermined securing actions.

9. The method according to claim 2, wherein the generated deceiving print has been generated and has been saved in place of the existing deceiving print after the candidate print has been compared with the reference print and with the existing deceiving print, with the score obtained with the reference print being under the validation threshold and the score obtained with the existing deceiving print being under the first predetermined threshold.

10. The method according to claim 9, comprising a step of configuring during which at least one deceiving print is saved beforehand, with the deceiving print being a random print.

11. The method according to claim 1, wherein the first candidate print is compared with the first deceiving print and a score of the first candidate print is computed for the first deceiving print, even though the score of the first candidate print is at least equal to the validation threshold for the reference print.

12. The method according to claim 1, wherein the reference print and the deceiving prints are saved in the same memory space, with the deceiving prints comprising biometric characteristics so predetermined as to identify the deceiving prints and distinguish same from the reference print.

13. The method according to claim 1, comprising the steps of:

creating a dummy deceiving print when the candidate print obtains a score above the validation threshold for the reference print or above the first predetermined threshold for one of the deceiving prints;

comparing the candidate print with the dummy deceiving print;

computing a score for this dummy deceiving print.

14. The method according to claim 13, comprising the step of saving the dummy deceiving print into the memory.

15. The method according to claim 13, comprising the step of simulating the writing of the dummy deceiving print into the memory.

16. The method according to claim 1, wherein the deceiving print has been generated to have characteristics enabling the candidate print to obtain a score between the first predetermined threshold and the validation threshold.

17. A device for biometric identification comprising at least one print capture device and a computer unit connected to the capture device, with the computer unit being so designed as to implement the method according to claim 1.