Presentation is loading. Please wait.

Presentation is loading. Please wait.

An Introduction to Biometric Recognition Anil K

Similar presentations


Presentation on theme: "An Introduction to Biometric Recognition Anil K"— Presentation transcript:

1 An Introduction to Biometric Recognition Anil K
An Introduction to Biometric Recognition Anil K. Jain, Fellow, IEEE, Arun Ross, Member, IEEE, and Salil Prabhakar, Member, IEEE, IEEE Transactions on Circuits and Systems for Video Technologies, vol. 14, no. 1, Jan. 2004 Multimedia Security

2 Outline (1/2) Part Ⅰ. Introduction Part Ⅱ. Biometric System
Part Ⅲ. Biometric System Errors Part Ⅳ. Comparison of Various Biometrics Part Ⅴ. Application of Biometric Systems

3 Outline (2/2) Part Ⅵ. Advantage and Disadvantage of Biometrics
Part Ⅶ. Limitation of (Unimodal) Biometric Systems Part Ⅷ. Multimodal Biometric Systems Part Ⅸ. Social Acceptance and Privacy Issues

4 Ⅰ. Introduction (1/5) The term biometric comes from the Greek words bios (life) and metrikos (measure). Biometrics – individuals’ physiological and/or behavioral characteristics.

5 Ⅰ. Introduction (2/5) Biometric Recognition
“who she is” vs. “what she possesses”

6 Ⅰ. Introduction (3/5) What biological measurements qualify to be a biometric? Universality Distinctiveness Permanence Collectability

7 Ⅰ. Introduction (4/5) In a practical biometric system, there are a number of other issues that should be considered… Performance Acceptability Circumvention

8 Ⅰ. Introduction (5/5) In conclusion, the system should meet… Accuracy
Speed Resource requirement Be harmless to the users Be accepted by the intended population Be sufficient robust to various attack

9 Ⅱ. Biometric System (1/10) A biometric system is essentially a pattern recognition system.

10 Ⅱ. Biometric System (2/10) A biometric system is designed using the following four main modules. Sensor module (encapsulating a quality checking module) Feature module Matcher module (encapsulating a decision making module) System database module

11 The templates in the system database may be updated over time.
Ⅱ. Biometric System (3/10) A sample flow chart: Feature Extractor Sensor Qualify checker System Database True / False Matcher Decision Maker template The templates in the system database may be updated over time.

12 Ⅱ. Biometric System (4/10) A biometric system may operate either in verification mode or identification mode. Verification mode: “Does this biometric data belong to Bob? ” Identification mode: “Whose biometric data is this? ”

13 Ⅱ. Biometric System (5/10) Enrollment System Database Login Interface
Get Name & Snapshot Quality Checker Check Quality Feature Extractor Enrollment Template

14 Ⅱ. Biometric System (6/10) Verification System Database True / False
Login Interface Get Name & Snapshot One template Feature Extractor Extract Features Matcher One match Verification Claimed identity

15 Ⅱ. Biometric System (7/10) Identification System Database
User’s identity or “user unidentified” Login Interface Get Name & Snapshot N templates Feature Extractor Extract Features Matcher N match Identification

16 Ⅱ. Biometric System (8/10) “Recognition” is the generic term of verification and identification. We do not make a distinction between verification and identification.

17 Ⅱ. Biometric System (9/10) Describing the verification problem:
An input feature vector: XQ A claimed identity: I The biometric template corresponding to I : XI The similarity between XQ and XI: S(XQ, XI) The predefined threshold of similarity: t True (a genuine user): ω1 ; False (an imposter): ω2

18 Ⅱ. Biometric System (10/10) The identification problem…
The identity enrolled in the system: Ik, k=1, 2,…, N The reject case: IN+1 The biometric template corresponding to Ik : XIk

19 Ⅲ. Biometric System Errors (1/9)
A biometric verification system makes two types of errors: mistaking biometric measurements from two different persons to be from the same person (called false match) mistaking two biometric measurements from the same person to be from two different persons (called false non-match)

20 Ⅲ. Biometric System Errors (2/9)
Hypothesis testing: H0: input XQ does not come from the same person as the template XI H1: input XQ comes from the same person as the template XI

21 Ⅲ. Biometric System Errors (3/9)
Decision: D0: person is not who she claims to be D1: person is who she claims to be. If S (XQ , XI) ≧ t , then decide D1 , else decide D0 .

22 Ⅲ. Biometric System Errors (4/9)
Such a hypothesis testing formulation contains two type of error: Type Ⅰ(α): false match (D1, when H0) Type Ⅱ(β): false non-match (D0, when H1) FMR is the probability of Type I error FNMR is the probability of Type II error

23 Ⅲ. Biometric System Errors (5/9)
Decision Threshold (t ) Matching Score (s ) Probability (p ) -∞ Imposter Distribution p (s|H0) Genuine p (s|H1) FNMR = P (D0|H1) FMR = P (D1|H0)

24 Ⅲ. Biometric System Errors (6/9)
The errors in identification mode: FMRN: the identification false match rate FNMRN: the identification false non-match rate FMRN = 1 – (1 – FMR)N ~ N × FMR FNMRN ~ FNMR

25 Ⅲ. Biometric System Errors (7/9)
Some situation may lead to following formulation of FMRN and FNMRN. FNMRN = RER + (1 - RER) × FNMR RER: retrieval error rate FMRN = 1 – (1 – FMR)N×P P: the average percentage of database searched during the identification

26 Ⅲ. Biometric System Errors (8/9)
False Non-match Rate (FNMR) False Match Rate (FMR) Forensic Applications High-security Civilian

27 Ⅲ. Biometric System Errors (9/9)
Important specifications in a biometric system: FMR: false match rate FNMR: false non-match rate FTC: failure to capture (e.g., a faint fingerprint) FTE: failure to enroll

28 Ⅳ. Comparison of Various Biometrics (1/10)
Each biometric has its strengths and weaknesses. No biometric is “optimal”. A brief introduction of the commonly used biometrics is given below…

29 Ⅳ. Comparison of Various Biometrics (2/10)
DNA 1-D ultimate unique code identical twins have identical DNA patterns contamination and sensitivity automatic real-time recognition issues privacy issues Ears The shape of the ear the structure of the cartilaginous tissue of the pinna

30 Ⅳ. Comparison of Various Biometrics (3/10)
Face - Also used by humans the location and shape of facial attributes the overall analysis of the face image Requiring a simple background and illumination In practice, … Detect the face Locate the face Recognize the face

31 Ⅳ. Comparison of Various Biometrics (4/10)
Facial, hand, and hand vein infrared thermogram A thermogram-based system does not require contact and is non-invasive Infrared sensors are prohibitively expensive 手掌靜脈辨識系統 資料來源:FUJITSU, Taiwan

32 Ⅳ. Comparison of Various Biometrics (5/10)
Fingerprint A fingerprint scanner costs about US $20 Single vs. Multiple

33 Ⅳ. Comparison of Various Biometrics (6/10)
Gait Hand and finger Geometry

34 Ⅳ. Comparison of Various Biometrics (7/10)
Iris stabilize during the first two years of life the irises of identical twins are different extremely difficult to surgically tamper the texture of the iris

35 Ⅳ. Comparison of Various Biometrics (8/10)
Keystroke Odor Palmprint

36 Ⅳ. Comparison of Various Biometrics (9/10)
Retinal scan the most secure biometric reveal some medical conditions Signature professional forgers may be able to reproduce signatures that fool the system Voice a combination of physiological and behavioral biometrics

37 Ⅳ. Comparison of Various Biometrics (10/10)

38 Ⅴ. Application of Biometric Systems (1/3)
The application of biometric can be divided into three main groups: Commercial ATM, credit card, cellular phone, distance learning, etc. Government ID card, driver’s license, social security, passport control, etc. Forensic terrorist identification, missing children, etc.

39 Ⅴ. Application of Biometric Systems (2/3)

40 Ⅴ. Application of Biometric Systems (3/3)
REVENUE (US$m) SOURCE: The `123' of Biometric Technology

41 Ⅵ. Advantage and Disadvantage of Biometrics (1/2)
All the users of the system have equal security level. Between 20% and 50% of all help desk calls are for password resets.

42 Ⅵ. Advantage and Disadvantage of Biometrics (2/2)
Speed is perceived as the biggest problem. FMR will increase when scaling up an identification application.

43 Ⅶ. Limitation of (Unimodal) Biometric Systems (1/2)
Noise in sensed data Intra-class variations

44 Ⅶ. Limitation of (Unimodal) Biometric Systems (2/2)
Distinctiveness e.g. Hand geometry, face, etc. Non-universality Spoof attacks

45 Ⅷ. Multimodal Biometric Systems (1/19)
Data Fusion Level of Fusion Fusion at Sensor level Fusion at Feature level Fusion at Opinion level Fusion at Decision level

46 Ⅷ. Multimodal Biometric Systems (2/19)
decision Feature Extraction Biometric snapshot Matching Decision Making Fusion System Database features

47 Ⅷ. Multimodal Biometric Systems (3/19)
This combination strategy is usually done by a concatenation of the feature vectors extracted by each feature extractors. This yields an extended size vector set.

48 Ⅷ. Multimodal Biometric Systems (4/19)
Two drawbacks: There is little control over the contribution of each vector component on the result. Both feature extractors should provide identical vector rates.

49 Ⅷ. Multimodal Biometric Systems (5/19)
Although it is a common belief that the earlier the combination is done, the better result is achieved, state-of-the-art data fusion relies mainly on the opinion and decision level.

50 Ⅷ. Multimodal Biometric Systems (6/19)
decision Feature Extraction Biometric snapshot Matching Decision Making Fusion System Database rank values

51 Ⅷ. Multimodal Biometric Systems (7/19)
The score must be adjusted first: ( Normalization must be done. ) The similarity measures must be converted into distance measures. The score generated by each classifier must have same range. [ex ]

52 Ⅷ. Multimodal Biometric Systems (8/19)
The combination strategies can be classified into three main groups: Fixed rules / equal weight Trained rules / unequal weight Adaptive rules / adaptive weight

53 Ⅷ. Multimodal Biometric Systems (9/19)
The most popular schemes are: Weight sum Weight product Decision trees ( base on if-then-else )

54 Ⅷ. Multimodal Biometric Systems (10/19)
Classifier 1 Classifier 2 Classifier 3 Score1 > t1 Score2 > t2 Score3 > t3 False True Yes No

55 Ⅷ. Multimodal Biometric Systems (11/19)
decision Feature Extraction Biometric snapshot Matching Fusion System Database Decision Making

56 Ⅷ. Multimodal Biometric Systems (12/19)
In this last case, the Borda count method can be used for combining the classifiers’ outputs. This approach overcomes the scores normalization that was mandatory for the opinion fusion level.

57 Ⅷ. Multimodal Biometric Systems (13/19)
class 2 class 2=2 Classifier 1 class 1 class 1=1 class 3 class 3=0 class 1 class 1=2 class 2=5 Classifier 2 class 2 class 2=1 class 1=3 class 3 class 3=0 class 3=1 class 2 class 2=2 Classifier 3 class 3 class 3=1 class 1 class 1=0

58 Ⅷ. Multimodal Biometric Systems (14/19)
One problem that appears with decision level fusion is the possibility of ties. For verification applications, at least three classifiers are needed.

59 Ⅷ. Multimodal Biometric Systems (15/19)
An important combination scheme at the decision level is the serial and parallel combination, also known as “AND” and “OR” combination. System 1 System 2 System 1 System 2

60 Ⅷ. Multimodal Biometric Systems (16/19)
The AND combination improves the False Acceptance Ratio. The OR combination improves the False Rejection Ratio.

61 Ⅷ. Multimodal Biometric Systems (17/19)
Multiple matchers snapshots units biometrics sensors optical & capacitance sensors minutiae & non-minutiae based matchers face & fingerprint two attempts of right index finger right index & middle fingers

62 Ⅷ. Multimodal Biometric Systems (18/19)
Example of Multimodal Biometric Systems “Person Identification Using Multiple Cues” Face, Voice “Expert Conciliation for Multimodal Person Authentication Systems using Bayesian Statistics” Face, Speech “Integrating Faces and Fingerprints for Personal Identification” Face, fingerprint “Personal Verification using Palmprint and Hand Geometry Biometric” Palmprint and Hand Geometry “Bioid: A Multimodal Biometric Identification System” voice, lip motion, face

63 Ⅷ. Multimodal Biometric Systems (19/19)
A combination of uncorrelated modalities is expected to result in a better improvement in performance. A combination of uncorrelated modalities can significantly reduce the FTE. However, the cost of the system may increase and the system may cause inconvenience.

64 Ⅸ. Social Acceptance and Privacy Issues (1/3)
The ease and comfort in interaction with a biometric system contribute to its acceptance. Biometric characteristics captured without the knowledge of the user is perceived as a threat to privacy by many individuals.

65 Ⅸ. Social Acceptance and Privacy Issues (2/3)
Biometrics can be used as one of the most effective means for protecting individual privacy. Biometric characteristics may provide additional information about the background of an individual.

66 Ⅸ. Social Acceptance and Privacy Issues (3/3)
Legislation is necessary to ensure that such information remains private and that its misuse is appropriately punished. Most of the commercial biometric systems available today store a template in an encrypted format.


Download ppt "An Introduction to Biometric Recognition Anil K"

Similar presentations


Ads by Google