Presentation is loading. Please wait.

Presentation is loading. Please wait.

Biometrics II CUBS, University at Buffalo

Similar presentations


Presentation on theme: "Biometrics II CUBS, University at Buffalo"— Presentation transcript:

1 Biometrics II CUBS, University at Buffalo http://www.cubs.buffalo.edu http://www.cedar.buffalo.edu/~govind/CSE717 govind@buffalo.edu

2 Maintaining Biometric Databases  Enrollment  Base truth established through seed documents  Positive Enrollment  Negative Enrollment  Problems in enrollment  Biometric System is only as secure as its enrollment  Fake Identities  Duplicate Identities  Quality  The biometric zoo  Sheeps,Goats,Wolves,Lambs

3 Large Scale Databases  How large is ‘large scale’?  Application  US-VISIT  National ID – Thailand, Hong Kong  Voter Registration – Peru  Accuracy issue  FAR N = N*FAR  FRR N = FRR  Number of False accepts scales as O(N 2 )  Scalability and retrieval  Indexing  Filtering  Binning

4 Indexing Problem  Biometric template have no natural order  Natural groups cannot be inferred from statistical descriptors

5 Indexing Problem: Solutions Binning and pruning Natural feature representation

6 Combination of biometric matchers Fingerprint matching Hand geometry matching Signature matching Alice Bob : 26 12 : Alice Bob : 0.31 0.45 : Alice Bob : 5.54 7.81 : Alice Bob : 0.95 0.11 : Combination algorithm Combination of the matching results of different biometric features provides higher accuracy.

7 Sequential combination of matchers Fingerprint matching Hand geometry matching Signature matching Alice Bob : 0.95 0.11 : Combination algorithm 1 Desired confidence achieved? Combination algorithm 2 Desired confidence achieved? Combination algorithm 3 No Yes

8 Information Fusion  Multiple sensors  Compensates environmental variation  Multiple samples  Allows interaction induced variation  Multiple instances of the same biometric  Multiple matchers  May use multiple representations  Serial and hierarchical combination  Multi-modal biometrics  Feature level Fusion  Score level Fusion  Decision level Fusion  Loosly coupled integration  Combination of biometrics and tokens

9 Multimodal biometrics:Fusion approaches

10 Security of biometric data  Issues in biometrics  Biometrics is secure but not secret  Permanently associated with user  Used across multiple applications  Can be covertly captured  Types of circumvention  Denial of service attacks  Fake biometrics attack  Replay and Spoof attacks  Trojan horse attacks  Back end attacks  Collusion  Coercion

11 Attacks on a Biometric System Database Biometric Sensor Feature Extraction Biometric Sensor Feature Extraction Matching ID : 8809 Authentication Enrollment Result

12 Gummy finger

13 Fake biometrics

14 Synthetic fingerprint

15 Securing biometric templates  Liveness testing  Cancelable biometrics  Signal domain distortion  Feature domain distortion  Hashing  Uses non invertible transform  Watermarking and steganography  Ensures that sensor input can be trusted  Encryption and digital signature  Combines security of encryption with non-repudiation of bometrics  Challenge response systems  Conversational biometrics  CAPTCHAs

16 Securing password information

17 Hashing

18 Cancelable biometrics

19 Privacy, social and ethical concerns  Aspects of privacy  Secrecy  Solitude  Anonymity  Unintended functional scope  Retinal pattern is capable of revealing diabetes  Hand irregularities can be correlated with genetic defect  This information may be used for discrimination  Unintended application scope  Can compromise anonymity of an individual  Biometric can be used for covert surveillance  Can be used to track purchases leading to spam  Social acceptance  Stigma associated with biometrics  Reduces level of expected privacy

20 Avenues for research  Face recognition  Palm verification  Key stroke dynamics  Speaker recognition  Soft biometrics  Multimodal biometrics  Classifier combination  Binning and Indexing  Security of biometrics  Interchange standards and API

21 Summary  Enrollment into biometric databases  Large scale databases  Filtering, binning and indexing biometric templates  Combination of classifiers  Multimodal biometrics  Attacks on Biometric System  Securing Biometric Templates  Privacy Social and Ethical Concerns

22 Thank You ssc5@cedar.buffalo.edu


Download ppt "Biometrics II CUBS, University at Buffalo"

Similar presentations


Ads by Google