Biometrics with Topics in Face Recognition Dr Karl

Biometrics with Topics in Face Recognition Dr. Karl Ricanek, Jr. Assistant Professor Computer Science Dept University of North Carolina, Wilmington

Discussion Overview l Biometrics ¡ Definition/History ¡ Technologies l Face Recognition ¡ History/Issues ¡ Research l Questions Focus and Answers

Biometrics Definition (Merriam-Webster online): the statistical analysis of biological observations and phenomena. l Biometrics are automated methods of recognizing a person based on a physiological or behavioral characteristic. (http: //www. biometrics. org) l ¡ ¡ Phenotypic biometric – based upon features or behaviors that are acquired through experience and development. Genotypic biometric – based upon genetic characteristics or traits.

Biometrics History l First documented example: Egypt several thousand years ago. (Biometrics: Advanced Identity Verification the complete guide, Julian Ashbourn) ¡ Khasekem, assistant to chief administrator, used phenotypic biometrics for identification of food provisions. l Notes were kept on every worker (100, 000 or more) detailing physical characteristics (eg. age, height, weight, deformities) and behavioral characteristics (eg. General disposition, lisp/slurs in speech, etc. )

Biometrics History l Biblical Reference ¡ Judges 12: 5 -6: “Then said the men of Gilead unto him, Say now Shibboleth: and he said Sibboleth: for he could not frame to pronounce it right. Then they took him, and slew him at the passages of the Jordan: and there fell at that time of the Ephraimites forty and two thousand. ” ¡ Phenotypic biometric, in particular, voice, was used to identify Ephraimites, the enemy of the Gileadites. l Ephraimites pronounced “Sh” as “S”

Biometrics History l Modern ¡ ¡ ¡ Belgian mathematician and astronomer Adolphe Quetelet ushered in the modern use of biometrics with his treatise of 1871, “L’anthropometrie ou mesuare des diffenretes facultes de l’homme” Frenchman Alphonse Bertillon, applied Quetelet work to develop a system to identify criminals based on anatomical measures. Argentinean police officer Juan Vucetich was the first to use dactyloscopy in 1888. Dactyloscopy is the taking of fingerprints using ink.

Biometric Technologies: Selected l Fingerprint l Voice l Iris/retina l Gait l Face Recognition

Biometric Technologies l Fingerprint ¡ Pros: l l l ¡ Years of research and understanding Security community comfortable with technology Innately distinctive feature Cons: l l l Can be altered/worn over time Some ethnic groups exhibit poor discrimination of finger prints Automatic techniques not trusted

Biometric Technologies l Voice ¡ Pros l l ¡ Non-invasive Distinctive w. r. t. vocal chords, vocal tract, patalte, sinuses, and tissue w/in mouth Cons l l Easily corrupted with noise High false rates (positive and negative) w. r. t. physical ailments (colds, sinus drains, etc. )

Biometric Technologies l Iris/Retina ¡ Pros l l ¡ Innately unique No change over time (static) Left and right within themselves Genetic inheritance (Genotypic) Cons l Acquiring image • Alignment/position • Pupil size change

Biometric Technologies l Gait ¡ Pros l l l ¡ Non-invasive Discriminate under various conditions (eg, walking, jogging, running) Promising research Cons l l Can be altered Too early in research

Biometric Technologies: Face Recognition l History Kanade 1977, Kaya 1972, Bledsoe 1964 Feature Metric 1888 Galton Profile Id Akamtsu 1991 Brunelli 1992 Neural Network Ricanek 1999 Variable Lateral Pose Recognition Turk 1991 Hong 1991 Shirovich 1987 Statistical Ricanek, Patterson & Albert 200 X Craniofacial Morphology: Models for Face Aging (Research in progress) Psychophysic neuroscience approaches

Face Recognition Techniques l Image Based ¡ Statistical based on O(2 nd) l l l ¡ Template matching l l PCA/Eigenfaces (dominant) Fisherfaces (LDA) Etc. Spectral analysis Gabor filtering Etc. Feature Based ¡ ¡ ¡ Geometric Feature metrics (spatial relationships) Morphable models (shape/texture)

FRT Diagram Preprocessing Probe Preprocessing Face Recognition System Gallery (DB) Rank ordered lists from gallery set with confidence factor

Face Recognition Technologies: Field Reports ACLU Press Release: Data on Face-Recognition Test at Palm Beach Airport Further Demonstrates Systems' Fatal Flaws. May 14, 2002. l ACLU press release: Drawing a blank: Tampa police records reveal poor performance of face-recognition technology: Tampa officials have suspended use of the system. Jan. 3, 2002. l Etc. l l Reports that system in real world app was effective 53% of the time l “System logs obtained by the ACLU through Florida's openrecords law show that the system never identified even a single individual contained in the department’s database of photographs. ”

Face Recognition Technologies: Problems l l l Resolution/Quality Orientation Scale Disguise Lighting Image Currency ¡ ¡ Physiologic changes due to growth Physiologic changes due to aging

My Research Niche: Age Progression l Age Progression ¡ ¡ Growth – from infancy to full maturation (~18) Maturation – from full maturation to senescence (elderly years)

My Research Niche: Age Progression l Maturation Age Progression ¡ Face undergoes significant changes during the adult age progression which dramatically impacts face recognition technologies. Loss of epidermis elasticity causes the formation of rhytides and ptosis. l Elasticity loss is caused primarily by photoaging but contributory factors include smoking, alcohol consumption, drug use, and some prescribed medications. l Skin texture changes occur also, rougher skin, blotchiness/discoloration, hanging skin, etc. l

My Research Niche: Age Progression

Face Recognition Rates (offline) l Probe-Gallery (temporally current) ¡ Image based: mid 90% ¡ Feature based: mid 90% l Probe-Gallery ¡ Image (temporally displaced) based: 80% (1 yr) – 50% (5 yr) ¡ Feature based: unknown

Face Recognition Rank Curve: Normal

Face Recognition Rank Curve: Age Progression

Team’s Research Constructing the first craniofacial database where each subject contains multiple images that span from late adolescences through senescence. Formulate understanding of the mechanisms of morphological changes in the human face as it ages from late adolescence (i. e. , ages 18 -21 years) to senescence (i. e. , ages 60+ years). l l ¡ ¡ Develop models based on analysis of features for consistent patterns versus idiosyncratic variations of craniofacial change due to aging. Develop soft tissue texture map models that simulate aging of skin. Detailed evaluation of FRT against the database. l l ¡ l l Which features fundamentally change with age? Which features DO NOT change with age? How and why does the FRT algorithm fail? Develop FRT algorithm that is robust against aging. Develop face detection and tracking techniques.

Questions and Answers