Presentation is loading. Please wait.

Presentation is loading. Please wait.

Pattern Recognition 1/6/2009 Instructor: Wen-Hung Liao, Ph.D. Biometrics.

Similar presentations


Presentation on theme: "Pattern Recognition 1/6/2009 Instructor: Wen-Hung Liao, Ph.D. Biometrics."— Presentation transcript:

1 Pattern Recognition 1/6/2009 Instructor: Wen-Hung Liao, Ph.D. Biometrics

2 Outline n Basic Concepts n Fingerprint n Iris Scan n Hand Geometry n Face Recognition

3 Identification vs Verification n Identification: Who am I? One-to-many search n Verification: Am I who I claim I am? One-to- one search n Detection: Find out whether there is an instance of a given type of object in an environment. n Recognition: detection + identification

4 Terminology n False Acceptance Rate (FAR) : the probability that a biometric device will allow a ‘bad guy’ to pass. Related to security. n False Rejection Rate (FRR):the probability that a biometric device won't recognize a good guy. Related to convenience. n The point where false accept and false reject curves cross is called the "Equal Error Rate." The Equal Error Rate provides a good indicator of the unit's performance. The smaller the Equal Error Rate, the better.

5 Validity of Test Data n Testing biometrics is difficult, because of the extremely low error rates involved. n Some are based on theoretical models. n Some are obtained from actual field testing. n It's important to remember that error rates are statistical: they are derived from a series of transactions by a population of users.

6 What is a good biometric feature? n Uniqueness n Invariance n Non-intrusive n Easy (or not too difficult) to acquire n Low processing cost

7 Fingerprint n Finger-scan biometrics is based on the distinctive characteristics of the human fingerprint. n A fingerprint image is read from a capture device, features are extracted from the image, and a template is created. n If appropriate precautions are followed, what results is a very accurate means of authentication.

8 Fingerprints vs Finger-scans n Fingerprint images require 250kb per finger for a high-quality image. n Can be acquired using ink-and-roll procedure, optical or non-contact methods. n Finger-scan technology doesn't store the full fingerprint image. It stores particular data about the fingerprint in a much smaller template, requiring from 250- 1000 bytes.

9 AFIS n AFIS (Automated Fingerprint Identification Systems) - commonly referred to as "AFIS Systems" (a redundancy) - is a term applied to large-scale, one-to-many searches. n Although finger-scan technology can be used in AFIS on 100,000 person databases, it is much more frequently used for one-to- one verification within 1-3 seconds.

10 Fingerprint Characteristics n Can be classified according to the decades-old Henry system: u left loop u right loop u arch u whorl u tented arch

11 Feature Extraction Steps Minutiae, the discontinuities that interrupt the otherwise smooth flow of ridges, are the basis for most finger- scan authentication.

12 Accuracy n False Rejection Rates (FRR), or the likelihood that the system will not "recognize" an enrolled user's finger-scan, in the vicinity of 0.01%. n False Acceptance Rates (FAR), or the likelihood that the system will mistakenly "recognize" the finger-scan of a user who is not in the system, are frequently stated in the vicinity of 0.001%. n The point at which the FAR and FRR meet is the Equal Error Rate, frequently claimed to be 0.1%.

13 Iris Scan n Iris recognition is based on visible (via regular and/or infrared light) qualities of the iris. n A primary visible characteristic is the trabecular meshwork (permanently formed by the 8th month of gestation), a tissue which gives the appearance of dividing the iris in a radial fashion. n Other visible characteristics include rings, furrows, freckles, and the corona.

14 Iris Recognition Technology n Iris recognition technology converts the visible characteristics discussed before into a 512 byte IrisCode(tm), a template stored for future verification attempts.

15 Accuracy n The odds of two different irises returning a 75% match (i.e. having a Hamming Distance of 0.25): 1 in 10^16 n Equal Error Rate (the point at which the likelihood of a false accept and false reject are the same): 1 in 1.2 million n The odds of 2 different irises returning identical IrisCodes: 1 in 10^52

16 Benefits n Uniqueness n Established prior to birth and remains intact through out the life.

17 For more details n Check Dr. John Daugman’s web page: http://www.cl.cam.ac.uk/users/jgd1000

18 Hand Scan n Hand-scan reads the top and sides of the hands and fingers, using such metrics as the height of the fingers, distance between joints, and shape of the knuckles. n Although not the most accurate physiological biometric, hand scan has proven to be an ideal solution for low- to mid-security applications where deterrence and convenience are as much a consideration as security and accuracy.

19 Example n HandPunch 2000/3000 model developed by Recognition Systems

20 Pros and Cons n Advantages u Ease of use u Resistant to fraud u Template size u User perception n Disadvantages u Static design u Cost u Injury to hands u Accuracy

21 Face Recognition n Most natural because this is how we human recognize other people. n Remains a difficult subject.

22 Primary Facial Scan Technologies n Eigenfaces n feature analysis n neural network n automatic face processing

23 Typical Eigenfaces

24 Feature Analysis n The most widely utilized facial recognition technology n Local Feature Analysis (LFA) utilizes dozens of features from different regions of the face, and also incorporates the relative location of these features. n The extracted (very small) features are building blocks, and both the type of blocks and their arrangement are used to identify/verify.

25 ANN Approach n Features from both faces - the enrollment and verification face - vote on whether there is a match. n Neural networks employ an algorithm to determine the similarity of the unique global features of live versus enrolled or reference faces, using as much of the facial image as possible.

26 AFP n Automatic Face Processing (AFP) is a more rudimentary technology, using distances and distance ratios between easily acquired features such as eyes, end of nose, and corners of mouth. n Not as robust, but AFP may be more effective in dimly lit, frontal image capture situations.


Download ppt "Pattern Recognition 1/6/2009 Instructor: Wen-Hung Liao, Ph.D. Biometrics."

Similar presentations


Ads by Google