Biometrics II

1
Biometrics II

• CUBS, University at Buffalo
• http//www.cubs.buffalo.edu
• http//www.cedar.buffalo.edu/govind/CSE717
• govind_at_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
• FARN NFAR
• FRRN FRR
• Number of False accepts scales as O(N2)
• 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

• Combination of the matching results of different
  biometric features provides higher accuracy.

7
Sequential combination of matchers
Fingerprint matching
Combination algorithm 1
No
Desired confidence achieved?
Yes
Signature matching
Combination algorithm 2
Yes
No
Desired confidence achieved?
Hand geometry matching
Combination algorithm 3
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 biometricsFusion 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
Biometric Sensor
Feature Extraction
Database
Enrollment
Feature Extraction
Biometric Sensor
Matching
Authentication
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_at_cedar.buffalo.edu