Biometrics - A Tool in Fraud Prevention

Biometrics - A Tool in Fraud Prevention

Agenda Authentication Biometrics : Need, Available Technologies, Working, Comparison Fingerprint Technology About Enrollment, Matching and Verification Key Concepts - FAR/ FRR/ ERR Challenges Faced in implementation Need For Biometrics Standards Relevance in National ID Card

Authentication Authentication: To prove or establish as being genuine

Authentication Methods Something you know Passwords PINs Mother s maiden name Something you have ATM card Smart card Digital certificate Key/Token Something you are Biometrics : DNA IRIS FingerPrint etc.

What is Biometrics? The term biometric refers to a science Involving the standard analysis of biological characteristics. Biometric is a unique, measurable characteristic or trait of a human being for automatically recognising or verifying identity.

Why Biometrics Required? Uniqueness Non-repudiation No Proxy Strong identification and authentication

Need of Biometrics? As the level of security breaches and transaction fraud increases, the need for highly secure identification and personal verification technologies is becoming apparent. What you Know (Password) What you have (Token/Smart Card) Who you are (Biometrics) Lost Yes Yes No Stolen Yes Yes No Forgotten Yes Yes No Forged Yes Yes No Damaged No Yes Yes Loaned to others Yes Yes No

Available Biometrics Technology DNA Residue Dynamic Signature Voice recognition Iris recognition Hand morphology Face recognition Fingerprints

Technology Comparison Iris Face Finger Signature Voice Accuracy Very High Medium High High Medium Ease of use Medium Medium High High High Barrier to Very High Medium High Medium Medium Attack User Acceptability Medium Medium Very High High Medium Long term Stability High Medium High Medium Medium Interference Coloured Contact Light, Aging Glasses, Hair Dryness Dirt, Age And Race Changing Signature Atmospheric Noise

Why Fingerprint Technology? Technically most tested technology and most known to general public. Small size of the reader facilitating integration in majority of applications (laptops, computers, mobile phones). Low cost of the readers thanks to the new sensors of the type " silicon chip. Fast treatment. Good compromise between the false rejection rate and the false acceptance rate.

Available Technologies in Fingerprint Sensors Optical sensors. Capacitance based sensors Ultrasound sensors. Thermal sensors. Pressure Sensors. Optical: The technology is based on reflection changes at the spots where the finger papilar lines touch the readers surface. Capacitance: The silicon sensor acts as one plate of a capacitor, and the finger is the other. The capacitance between platen and the finger is converted into a 8-bit gray scale digital image.

About Optical Based Technology Optical Sensor Technology Optical System Design OPTICS Software Algorithms for Fingerprint Image Enhancement, Analysis & Matching LIGHT SOURCE CCD VIDEO SIGNAL FRAME GRABBER The scanning process starts when you place your finger on a glass plate, and a CCD camera takes a picture. The scanner has its own light source, typically an array of light emitting diodes, to illuminate the ridges of the finger. The CCD system actually generates an inverted image of the finger, with darker areas representing more reflected light (the ridges of the finger) and lighter areas representing less reflected light (the valleys between the ridges).

About Capacitance Based Technology Capacitance Sensor Technology The difference in the flow-out current generated by the difference between points of the contact and non contact on the fingerprint is measured. Fingerprint information of 76800 pixels is converted to 8- bit gray scale data.

How do Biometrics Work? Enrollment: Add a biometric identifier to a database Fingerprint, Voice, Facial or Iris Present biometric Capture Process Store Compare IDENTIFIED Match Verification: Match against an enrolled record No Match Present biometric Capture Process Denied

Process of Fingerprint Enrollment What is Enrollment? What is good enrollment? Errors/ tips in enrollment Position Pressure Rotation Clean finger

Methods of Matching There are Two methods employed to match the live template to the stored template Minutia Based Matching Pattern Based Matching

What is Minutia??

Minutia Extraction

Fingerprint Image Verification

Pattern Matching..

Key Concepts... False Acceptance Rate (FAR) The FAR is the frequency that a non authorized person is accepted as authorized. False Rejection Rate (FRR) The FRR is the frequency that an authorized person is rejected access. Equal Error Ratio (ERR) It is the point which shows optimal values where FAR and FRR are equal.

Application Overview Applicant Applicant Enrollment Enrollment Web Web Enrollment Enrollment Data Data Base Base collection collection and and storing storing in in ID ID Card Card PC Verification Verification One One to to one one match match

Selection of a Biometric Technology.. Some Tips Study and define the application Study and define the environment... Study and define the user profile.. Study the requirements for verification accuracy Find a balance between Overall System cost, capabilities and security level.

Possible Places of use Boot Time/Network Logon Application Authentication: When Modifying a data When Adding a data When Deleting data

What Biometrics can offer.. Strong Authentication Traceability Management of Authentication Process.

Need For Standards Lack of interoperable hardware and software Standard. Privacy concern and inconsistent performance claims, Develop an OS independent standard. Growth of Biometrics

Standard Biometrics API s Flow of Standard API

Thank you Mr. Y.D. Wadaskar (Managing Director) WYSE Biometrics Systems Pvt. Ltd. WYSE Biometrics Systems Pvt. Ltd. Survey No. 82/1, Plot 20, Sahakar Nagar-I, Parvati, Pune 411 009. INDIA. Ph. No.- +91-20-2422 4058, 2422 5268 Fax: +91-20-2422 7115. Email: ydw@wyse.co.in Website: www.wyse.co.in Copyright@1993-2004 WYSE Biometrics Systems Pvt. Ltd.

Features of Capacitance Fingerprint Sensor A silicon semiconductor device performs stable with no optical mechanisms Silicon semiconductor, can be very easily integrated into various control circuits, memories, interfaces, and so on. The devices combine a new technology with proven technology for low power consumption. Highly durable with a rigid protective membrane.

Advantages of Capacitance over Optical Sensors Capacitance Requires Live Finger Gives Better image in Less surface area Immune to ambient light. Fingerprint will be retrieved as 1:1 image size. Greasy, Muddy, Oily, soiled, and tainted fingers can be detected Optical Photocopy can also fool the system. Needs more Surface area. Ambient light can cause poor image quality. Image may get magnified or reduced Acceptance of greasy, Muddy, Oily fingerprint is low.

Disadvantages of capacitance over Optical Capacitance Capacitance sensors need to have a way for ESD to be discharged before touching the device. Optical Optical devices can not be damaged by ESD(Electro static discharge) events.

About WYSE Biometrics Systems Pvt Ltd. Incorporated in 1993 in the name of WYSE Systems. WYSE Systems is ISO 9001:2000, IT Security Provider having a large client base in India. First Indian Manufacturer of Capacitance based Fingerprint Device. Areas of operation Hardware based software Security Solutions - Sentry Network Security Solutions - SentryNet Internet/Extranet Security Solutions - esentry Data Protection - Sentry Secure Click Biometrics Security - BioSentry

About Capacitance based BioSentry Hardware Specifications: Sensor type : Capacitive solid state device Recognition rate : 0.4 seconds for ONE to ONE match Resolutions : 500 DPI Effective Sensing Area : 12.8mm X 15.0 mm. Sensor Array size : 256 X 300 Pixel Sensor array FRR (False Rejection Ratio) : 1/1000 FAR (False Acceptance Ratio) : 1/1,00,000 Operating Temperature : 0-60 Degree Centigrade Power Supply : DC 5V Maximum Power consumption: 80 ma Platform ( Drivers for OS ) : WIN 98,WIN 2000,ME,XP Interface : USB ( Cont. )

About Capacitance based BioSentry Hardware Specifications: Physical Dimensions : 63 MM X 90 MM X 28 MM Exceptionally Hard abrasion and chemical protective coating. Compressed Template size 400 to 1024 bytes (depending upon number of minutia detected) World-class detection and matching algorithms. Algorithms consumes less than 1 MB of CPU memory at any point of time. Complete SDK with documentation and sample code. Verification time is less that 2sec to match finger print within a database of 1000 Fingerprint templates.