INTRODUCTION TO BIOMETRICS Comp. Lab. C Dulce-Garcia Bldg. University of San Agustin February 28, Ronald C. Tantiado Computer Engineer
OUTLINE 1. The Basics 2. Biometric Technologies 3. Multi-model Biometrics 4. Biometric Applications 5. Constraints 6. Research
SECTION I: THE BASICS Why Biometric Authentication? Frauds in industry Identification vs. Authentication
WHAT IS BIOMETRICS? The automated use behavioral and physiological characteristics to determine or veiry an identity. Know HaveBe Rapid!
FRAUDS IN INDUSTRY HAPPENS IN THE FOLLOWING SITUATIONS: Safety deposit boxes and vaults Bank transaction like ATM withdrawals Access to computers and s Credit Card purchase Purchase of house, car, clothes or jewellery Getting official documents like birth certificates or passports Obtaining court papers Drivers licence Getting into confidential workplace writing Checks
WHY BIOMETRIC APPLICATION? To prevent stealing of possessions that mark the authorised person's identity e.g. security badges, licenses, or properties To prevent fraudulent acts like faking ID badges or licenses. To ensure safety and security, thus decrease crime rates
IDENTIFICATION VS. AUTHENTICATION IdentificationAuthentication It determines the identity of the person. It determines whether the person is indeed who he claims to be. No identity claim Many-to-one mapping. Cost of computation number of record of users. Identity claim from the user One-to-one mapping. The cost of computation is independent of the number of records of users. Captured biometric signatures come from a set of known biometric feature stored in the system. Captured biometric signatures may be unknown to the system.
SECTION II: BIOMETRIC TECHNOLOGIES Several Biometric Technologies Desired Properties of Biometrics Comparisons
TYPES OF BIOMETRICS Fingerprint Face Recognition Session III Hand Geometry Iris Scan Voice Scan Session II Signature Retina Scan Infrared Face and Body Parts Keystroke Dynamics Gait Odour Ear DNA
FINGER PRINT RECOGNITION Minutiae Pattern Matching Problems: sometimes unusable
VASCULAR PATTERN MATCHING LED infrared light Fingers and back of hand Not completely viable
IRIS RECOGNITION Uses infrared light Converts Images to vectors Needs further development
FACIAL RECOGNITION Location and position of facial features Dependent on background and lighting conditions
VOICE VERIFICATION Factors: pitch, intensity, quality and duration Text dependent Text independent Problems: include background noise
HAND GEOMETRY Scan both sides of hand Primarily used for verification Not as accurate as other methods
DYNAMIC SIGNATURE Factors: velocity, acceleration and speed Mainly used for verification Problems: forgers could reproduce
RETINA RECOGNITION One of the most secure means of biometrics Unique to each person Unique to each eye Problems: require effort on the part of subjects
OTHER TYPES Keystroke Gait DNA Odor
DESIRED PROPERTIES Universality Uniqueness Permanence Collectability Performance User’s Accpetability Robustness against Circumvention
COMPARISON Biometric TypeAccuracyEase of UseUser Acceptance FingerprintHighMediumLow Hand GeometryMediumHighMedium VoiceMediumHigh RetinaHighLow IrisMedium SignatureMedium High FaceLowHigh
SECTION III: A MULTI-MODEL BIOMETRICS Multi-modal Biometrics Pattern Recognition Concept A Prototype
MULTIMODAL BIOMETRICS
PATTERN RECOGNITION CONCEPT SensorsExtractors Image- and signal- pro. algo. Classifiers Biometrics Voice, signature acoustics, face, fingerprint, iris, hand geometry, etc Data Rep. 1D (wav), 2D (bmp, tiff, png) Feature Vectors Negotiator Scores Decision: Match, Non-match, Inconclusive EnrolmentTraining Submission Threshold
AN EXAMPLE: A MULTI-MODEL SYSTEM SensorsExtractorsClassifiersNegotiator Accept/ Reject 1D (wav) 2D (bmp) ID Face Extractor Voice Extractor Face Feature Voice Feature Face MLP Voice MLP AND Objective: to build a hybrid and expandable biometric app. prototype Potential: be a middleware and a research tool
Basic Operators 3D2D1D Data Representation Ex-qVoice ExFace Ex Extractors Cl-qVoice MLPFace MLP Learning-based Classifiers … … Signal Processing, Image Procesing Different Kernels (static or dynamic) NN, SVM, Negotiation Logical AND Diff. Combination Strategies. e.g. Boosting, Bayesian {LPC, FFT, Wavelets, data processing} {Fitlers, Histogram Equalisation, Clustering, Convolution, Moments} Biometrics Voice, signature acoustics Face, Fingerprint, Iris, Hand Geometry, etc. Face ABSTRACTION
cWaveProcessing fWaveProcessing cWaveOperator cWaveStackcFFTcFFiltercWaveletcLPCcDataProcessing cWaveObject Output data Input data Operators Operants * cPeripherique Audio 1 AN EXTRACTOR EXAMPLE: WAVE PROCESSING CLASS
LSIIT, CNRS-ULP, Groupe de Recherche en Intelligence Artificielle Pour plus de renseignements : Pr J. Korczak, Mr N. Identité Accepter, Rejeter w1 w2 Effacer les silences Transformation de l’ondelette C 0 C 1 C 2 C 3 C 4 C 5 C 6 C 7 C 9 C 10 C 11 C 12 C 13 C 14 C 15 Fréquence Temps Normalisation + Codage Réseau des neurones Apprentissage et Reconnaissance Détection des yeux Trouver X Trouver Y Filtre de base Inondation + Convolution Extraction Normalisation + Codage Moment Vert Bleu Hue Saturation Intensité Réseau des neurones Apprentissage et Reconnaissance Visage Voix Base des données Décision SYSTEM ARCHITECTURE IN DETAILS
SECTION V: APPLICATIONS Authentication Applications Identification Applications Application by Technologies Commercial Products
COMMERCIAL APPLICATIONS Computer login Electronic Payment ATMs Record Protection
GOVERNMENT APPLICATIONS Passport control Border control Access Control
FORENSIC APPLICATIONS Missing Persons Corpse identification Criminal investigations
APPLICATION BY TECHNOLOGIES Biometrics Vendors Market Share Applications Fingerprint9034%Law enforcement; civil government; enterprise security; medical and financial transactions Hand Geometry-26%Time and attendance systems, physical access Face Recognition1215%Transaction authentication; picture ID duplication prevention; surveillance Voice Authentication 3211%Security, V-commerce Iris Recognition19%Banking, access control
TYPE OF AUTHENTICATION Authentication 1:1 Verification 1:N
CONSTRAINTS ON BIOMETRICS Typical “Constrained” Image Constraints: Lighting Distance Pose Expression Time Lapse Occlusion
CONSTRAINTS ON BIOMETRICS “Unconstrained” Image
BIOMETRICS RESEARCH Aging Research
BIOMETRICS RESEARCH Demographics Older vs. Younger Males vs. Females Geographic origin of algorithms
BIOMETRICS RESEARCH Periocular Region Recognition Texture, color, eye shape Overcome facial occlusion
BIOMETRICS RESEARCH Ear Recognition Not affected by aging or expression Covert collection of images Little research performed
CONCLUSION Questions?
Thank You…