Jenna Lutton February 26th, 2007 Biometrics Jenna Lutton February 26th, 2007
Roadmap Introduction to Biometrics Facial Recognition Iris Scanning Fingerprint Scanning Palm-Vein Biometrics Behaviometrics Summary Questions
Introduction Biometrics – Use of physical characteristics to identify a person Two goals: Verification – Am I who I claim I am? Identification – Who am I? Future of security Border patrol ID Cards ATM access
Facial Recognition Compare a person’s face from a still image (or video capture) against a face stored in a database Two approaches: Geometric matching Distance between facial features Template matching Facial image – 2D array of pixel intensity values Compare against multiple templates of whole face
Issues Illumination Time Delay Pose Variation Facial Expression Multiple faces in image Wearing glasses, etc.
Fingerprint Scanning Very popular method of authentication Every person has unique fingerprint Scanner gets an image of your fingerprint Compares pattern of specific fingerprint features (minutiae) against another pattern in a database
Only finds adequate number of common minutiae patterns
Problems Some can be fooled by images of fingerprints Some can be fooled by molds Mythbusters: http://www.youtube.com/watch?v=ZncdgwjQxm0 Developing ways to bypass these: Scanner checks if cells are alive Scanner checks if blood flowing through finger
Iris Scanning Use infrared to capture an image of the iris (visible ring that surrounds the pupil)
Iris Scanning (cont.) Each eye has a unique iris, has measurable features (striations, pits, furrows) Controls the amount of light that enters the eye After an image of the iris is taken, a pattern or code is generated based on its features
Advantages Have to be present at scanner Non-invasive Not affected by contacts or glasses More points of comparison
Palm-Vein Biometrics Veins are also unique among people Uses near-infrared to identify all of the veins in your hand and captures that image
Advantages Change little with age Hard to spoof – not all veins visible Sanitary – don’t have to physically touch the scanner Used today as extra security layer at ATM machines
Behaviometrics Using behavioral characteristics of a person to verify or identify a person Example: Identify a person by the way they type Analyze keystroke timing Determined by finger length, motor skills, etc. Example: Handwriting analysis Speed/rhythm of writing Sequence in which you form letters
Summary Biometrics revolutionizing security Not all methods are reliable or secure Trying to find more ways to uniquely identify a person
Questions?
Sources http://www.iridiantech.com/index2.php http://www.howstuffworks.com/ http://www.face-rec.org/general-info/ http://www.face-rec.org/interesting-papers/General/gross_ralph_2001_4.pdf http://www.face-rec.org/interesting-papers/General/cr1182.pdf http://www.biometric-control.co.uk http://www.discover.com/web-exclusives-archive/computer-recognition1124/ Early, Mark. “ARE Biometrics the Key to Data Security?” EContent; Sep2006; Vol. 29 Issue 7; Pg 38-42. Hall, Kenji. “Biometrics: Vein Scanners Show Progress.” Business Week Online; Pg 22