US2016283603A1PendingUtilityA1

Methods and systems for testing performance of biometric authentication systems

48
Assignee: LANGLEY RICHARD JAYPriority: Dec 22, 2010Filed: Jun 1, 2016Published: Sep 29, 2016
Est. expiryDec 22, 2030(~4.4 yrs left)· nominal 20-yr term from priority
H04L 63/0861G06F 17/30876G06F 16/955H04L 9/3231
48
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

A method of testing the performance of a biometric authentication system includes conducting an initial biometric authentication transaction for an individual using data associated with the individual, and generating an initial result for the initial transaction with the biometric authentication system. Moreover, the method includes obtaining additional data associated with the individual when the additional data is required for conducting a subsequent biometric authentication transaction or after randomly determining that the subsequent transaction is to be conducted. Furthermore, the method includes conducting the subsequent transaction, generating a subsequent result for the subsequent transaction with the biometric authentication system, and updating cumulative performance records based on the initial and subsequent results. The method also includes generating a summary on a temporal or quantifiable basis that includes the cumulative performance records, and determining the performance of the biometric authentication system using the cumulative performance records or the summary.

Claims

exact text as granted — not AI-modified
What is claimed is: 
     
         1 . A method of testing the performance of a biometric authentication computer system, said method comprising:
 conducting, by a processor, using a first algorithm, an identification authentication transaction using an identification probe, the identification probe including user biometric template data;   generating an initial result, the initial result including at least one potentially matching data record;   replacing the identification probe user biometric template data with different user biometric template data;   conducting, using a second algorithm, a subsequent identification authentication transaction using the identification probe;   generating a subsequent result, the subsequent result including at least one potentially matching data record; and   comparing the initial and subsequent potentially matching data records against each other, potentially matching data records included in the initial and subsequent results being true matches.   
     
     
         2 . The method in accordance with  claim 1 , further comprising reviewing, by an operator, each potentially matching data record not included in both the initial and subsequent results. 
     
     
         3 . The method in accordance with  claim 1 , further comprising using the number of true matches included in the initial result only and the number of true matches included in the subsequent result only to calculate relative performance metrics between the first and second algorithms. 
     
     
         4 . A non-transitory computer-readable storage medium having instructions stored thereon for testing the performance of biometric authentication computer systems, the instructions when executed by a processor of the computer system cause the computer system to:
 conduct, using a first algorithm, an identification authentication transaction using an identification probe, the identification probe including user biometric template data;   generate an initial result, the initial result including at least one potentially matching data record;   replace the identification probe user biometric template data with different user biometric template data;   conduct, using a second algorithm, a subsequent identification authentication transaction using the identification probe;   generate a subsequent result, the subsequent result including at least one potentially matching data record; and   compare the initial and subsequent potentially matching data records against each other, potentially matching data records included in the initial and subsequent results being true matches.   
     
     
         5 . The non-transitory computer-readable storage medium of  claim 4 , the instructions when executed by a processor of the computing system further cause the computing system to use the number of true matches included in the initial result only and the number of true matches included in the subsequent result only to calculate relative performance metrics between the first and second algorithms. 
     
     
         6 . A computing system for testing the performance of a biometric authentication computer system comprising:
 a processor; and   a memory configured to store data, said computing system being associated with a network and said memory being in communication with said processor and having instructions stored thereon which, when executed by said processor cause said computing system to:   conduct, using a first algorithm, an identification authentication transaction using an identification probe, the identification probe including user biometric template data;   generate an initial result, the initial result including at least one potentially matching data record;   replace the identification probe user biometric template data with different user biometric template data;   conduct, using a second algorithm, a subsequent identification authentication transaction using the identification probe;   generate a subsequent result, the subsequent result including at least one potentially matching data record; and   compare the initial and subsequent potentially matching data records against each other, potentially matching data records included in the initial and subsequent results being true matches.   
     
     
         7 . The computing system of  claim 6 , the instructions when executed by said processor further cause said computing system to:
 compare the illumination characteristic differences against a threshold score; and   determine the action was executed, and that the biometric data was captured from a live person, when at least one of the differences is at least equal to the threshold score.

Cited by (0)

No later patents cite this yet.

References (0)

No backward citations on record.