Introduction to Biometrics: in a general format

This is an outline of starting studying biometrics.

1. Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi Introduction to Biometrics Course CodeCSEM-1141

2. Reference Books: Introduction to Biometrics By Anil K. Jain, Arun A. Ross & Karthik Naddakumar Handbook of Multibiometrics By Professor David D. Zhang & Professor Anil K. Jain Biometrics By John D. Woodward, Jr. Nicholas M. Orlans & Peter T. Higgins @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi 2

3. What is Biometric System ? A Biometric system is a person identification system that recognizes a person uniquely using his/her physiological and behavioral characteristics. 3 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

4. Introduction The term biometric comes from the Greek words bios (life) and metrikos (measure) . Biometrics – individuals’ physiological and/or behavioral characteristics. 4 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

5. Application of Biometrics: 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, etc. Forensic Terrorist identification, Missing children, Dead body identification etc . 5 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

6. Applications: Traditional methods involving passwords and PIN numbers Biometrics provide highest level of security Problems: Misplaced, Get lost, Fake @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi 6

7. Applications: DNA means deoxyribonucleic acid @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi 7

8. Application of Biometric Systems 8 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

9. @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi 9

10. Biometric System A biometric system is designed using the following four main modules. Sensor module (encapsulating a quality checking module) Feature module Matching module (encapsulating a decision making module) System database module 10 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

11. Biometric System A sample flow chart: 11 Feature Extractor Sensor Qualify checker System Database True / False Matcher Decision Maker template The templates in the system database may be updated over time. @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

12. Biometric System A biometric system may operate either in verification mode or identification mode . Verification mode: “ Does this biometric data belong to Mr. Babor ? ” Identification mode: “ Whose biometric data is this? ” 12 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

13. Biometric System 13 System Database Login Interface Get Name & Snapshot Quality Checker Check Quality Feature Extractor Enrollment Template @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

14. Biometric System 14 System Database True / False Login Interface Get Name & Snapshot One template Feature Extractor Extract Features Matcher One match Verification Claimed identity @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

15. Biometric System 15 System Database User’s identity or “ user unidentified ” Login Interface Get Name & Snapshot N templates Feature Extractor Extract Features Matcher N match Identification @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

16. Biometric System “Recognition” is the generic term of verification and identification. We do not make a distinction between verification and identification. 16 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

17. Biometric System Describing the verification problem: An input feature vector: X Q A claimed identity: I The biometric template corresponding to I : X I The similarity between X Q and X I : S(X Q , X I ) The predefined threshold of similarity: t True (a genuine user): ω 1 ; False (an imposter): ω 2 17 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

18. Biometric System The identification problem… The identity enrolled in the system: I k , k=1, 2,…, N The reject case: I N+1 The biometric template corresponding to I k : X I k 18 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

19. Biometric System Errors 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 ) 19 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

20. Biometric System Errors Hypothesis testing: H : input X Q does not come from the same person as the template X I H 1 : input X Q comes from the same person as the template X I 20 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

21. Biometric System Errors Decision: D : person is not who he/she claims to be D 1 : person is who he/she claims to be. 21 If S ( X Q , X I ) ≧ t , then decide D 1 , else decide D . @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

22. Biometric System Errors Such a hypothesis testing formulation contains two type of error: Type Ⅰ(α): false match ( D 1 , when H ) Type Ⅱ (β): false non-match ( D , when H 1 ) 22 FMR is the probability of Type I error FNMR is the probability of Type II error @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

23. Biometric System Errors 23 Decision Threshold ( t ) Matching Score ( s ) Probability ( p ) ∞ -∞ Imposter Distribution p ( s | H ) Genuine Distribution p ( s | H 1 ) FNMR = P ( D | H 1 ) FMR = P ( D 1 | H ) @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

24. Biometric System Errors The errors in identification mode: FMR N : the identification false match rate FNMR N : the identification false non-match rate FMR N ＝ 1 – (1 – FMR) N ～ N × FMR FNMR N ～ FNMR 24 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

25. Biometric System Errors Some situation may lead to following formulation of FMR N and FNMR N . FNMR N ＝ RER + (1 - RER) × FNMR RER: retrieval error rate FMR N ＝ 1 – (1 – FMR) N×P P: the average percentage of database searched during the identification 25 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

26. Biometric System Errors 26 False Non-match Rate (FNMR) False Match Rate (FMR) Forensic Applications High-security Applications Civilian Applications @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

27. Biometric System Errors Important specifications in a biometric system: FMR : false match rate FNMR: false non-match rate FTC: failure to capture (e.g., a faint (poorly, weak) fingerprint) FTE: failure to enroll 27 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

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… 28 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

29. Ⅳ. Comparison of Various Biometrics (2/10) DNA fragments show unique patterns from one person to the next. Used in paternity disputes and as forensic evidence. DNA 1-D ultimate unique code identical twins have identical DNA patterns contamination and sensitivity automatic real-time recognition issues privacy issues @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi 29

30. Ⅳ. Comparison of Various Biometrics (2/10) 30 Ears The shape of the ear the structure of the cartilaginous tissue of the pinna . @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

31. 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

32. 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 32 手掌靜脈辨識系統 資料來源： FUJITSU, Taiwan @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

33. Ⅳ. Comparison of Various Biometrics (5/10) Fingerprint: Minutiae, core, delta, crossover, bifurcation, dot features of a fingerprints are considered to identify a person. 33 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

34. Ⅳ. Comparison of Various Biometrics (6/10) Gait Hand and finger Geometry 34 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

35. 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

36. Ⅳ. Comparison of Various Biometrics (8/10) Keystroke Odor Palmprint 36 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

37. 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

38. Ⅳ. Comparison of Various Biometrics (10/10) 38 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

39. 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

40. Ⅴ. Application of Biometric Systems (2/3) 40 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

41. Ⅴ. Application of Biometric Systems (3/3) 41 SOURCE: The `123' of Biometric Technology REVENUE (US$m) @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

42. Ⅵ. Advantage and Disadvantage of Biometrics (1/2) Advantage All the users of the system have equal security level. Between 20% and 50% of all help desk calls are for password resets. 42 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

43. Ⅵ. Advantage and Disadvantage of Biometrics (2/2) Disadvantage Speed is perceived as the biggest problem. FMR will increase when scaling up an identification application. 43 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

44. Ⅶ. Limitation of (Unimodal) Biometric Systems (1/2) Noise in sensed data Intra-class variations 44 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

45. Ⅶ. Limitation of (Unimodal) Biometric Systems (2/2) Distinctiveness e.g. Hand geometry, face, etc. Non-universality Spoof attacks 45 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

46. 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

47. Ⅷ. Multimodal Biometric Systems (2/19) 47 decision Feature Extraction Biometric snapshot Matching Decision Making Feature Extraction Biometric snapshot Fusion System Database features features @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

48. 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

49. 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

50. 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

51. Ⅷ. Multimodal Biometric Systems (6/19) 51 decision Feature Extraction Biometric snapshot Matching Decision Making Feature Extraction Biometric snapshot Fusion System Database Matching rank values rank values @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

52. 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. 0-100] 52 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

53. 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

54. Ⅷ. Multimodal Biometric Systems (9/19) The most popular schemes are: Weight sum Weight product Decision trees ( base on if-then-else ) 54 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

55. Ⅷ. Multimodal Biometric Systems (10/19) 55 Classifier 1 Classifier 2 Classifier 3 Score 1 > t 1 Score 2 > t 2 Score 3 > t 3 False True Yes Yes Yes No No No No Yes Score 2 > t 2 False False True @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

56. Ⅷ. Multimodal Biometric Systems (11/19) 56 decision Feature Extraction Biometric snapshot Matching Fusion System Database Matching Decision Making Decision Making Feature Extraction Biometric snapshot decision decision @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

57. 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

58. Ⅷ. Multimodal Biometric Systems (13/19) 58 Classifier 1 Classifier 2 Classifier 3 class 2 class 1 class 3 class 1 class 2 class 3 class 2 class 3 class 1 class 2=2 class 1=1 class 3=0 class 1=2 class 2=1 class 3=0 class 2=2 class 3=1 class 1=0 class 2=5 class 1=3 class 3=1 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

59. 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

60. 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. 60 System 1 System 2 System 1 System 2 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

61. Ⅷ. Multimodal Biometric Systems (16/19) The AND combination improves the False Acceptance Ratio. The OR combination improves the False Rejection Ratio. 61 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

62. Ⅷ. Multimodal Biometric Systems (17/19) 62 Multimodal Biometrics Multiple matchers Multiple snapshots Multiple units Multiple biometrics Multiple sensors right index & middle fingers optical & capacitance sensors minutiae & non-minutiae based matchers face & fingerprint two attempts of right index finger @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

63. 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

64. 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

65. Ⅸ. Social Acceptance and Privacy Issues (1/3) Social Acceptance 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

66. Ⅸ. Social Acceptance and Privacy Issues (2/3) Privacy Issues 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 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi

67. 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. 67 @ Prof. Dr. A K M Akhtar Hossain, Dept. of CSE, University of Rajshahi