bio metrics overview

8/3/2019 Bio Metrics Overview

1/10

Biometrics Overview

This Document Last Updated: 27 March 2006 Page 1 of 10

Introduction

Biometrics is a general term used alternatively to describe acharacteristic or a process.

As a characteristic:

1. A measurable biological (anatomical and physiological)and behavioral characteristic that can be used forautomated recognition.

As a process:

2. Automated methods of recognizing an individual basedon measurable biological (anatomical and physiological)and behavioral characteristics.

Biometric systems have been researched and tested for a fewdecades, but have only recently entered into the publicconsciousness because of high profile applications, usage inentertainment media (though often not realistically) andincreased usage by the public in day-to-day activities. Exampledeployments within the United States Government include theFBIs Integrated Automated Fingerprint Identification System(IAFIS), the US-VISIT program, the Transportation WorkersIdentification Credentials (TWIC) program, and the RegisteredTraveler (RT) program. Many companies are also implementing

biometric technologies to secure areas, maintain time records,and enhance user convenience. For example, for many yearsDisney World has employed biometric devices for season ticketholders to expedite and simplify the process of entering its parks,while ensuring that the ticket is used only by the individual towhom it was issued.

A typical biometric system is comprised of five integratedcomponents: A sensor is used to collect the data and convert theinformation to a digital format. Signal processing algorithmsperform quality control activities and develop the biometrictemplate. A data storage component keeps information that new

biometric templates will be compared to. A matching algorithmcompares the new biometric template to one or more templateskept in data storage. Finally, a decision process (eitherautomated or human-assisted) uses the results from the matchingcomponent to make a system-level decision.
http://www.fbi.gov/hq/cjisd/iafis.htmhttp://www.fbi.gov/hq/cjisd/iafis.htmhttp://www.dhs.gov/dhspublic/display?theme=91&content=3768http://www.tsa.gov/public/interapp/editorial/editorial_multi_image_with_table_0218.xmlhttp://www.tsa.gov/public/interapp/editorial/editorial_multi_image_with_table_0218.xmlhttp://www.tsa.gov/public/display?theme=8&content=09000519800b4dddhttp://www.tsa.gov/public/display?theme=8&content=09000519800b4dddhttp://www.tsa.gov/public/display?theme=8&content=09000519800b4dddhttp://www.tsa.gov/public/display?theme=8&content=09000519800b4dddhttp://www.tsa.gov/public/interapp/editorial/editorial_multi_image_with_table_0218.xmlhttp://www.tsa.gov/public/interapp/editorial/editorial_multi_image_with_table_0218.xmlhttp://www.dhs.gov/dhspublic/display?theme=91&content=3768http://www.fbi.gov/hq/cjisd/iafis.htmhttp://www.fbi.gov/hq/cjisd/iafis.htm


2/10

Biometrics Overview

Biometric Modalities

Commonly implemented or studied biometric modalities includefingerprint, face, iris, voice, signature and hand geometry. Many

other modalities are in various stages of development andassessment. There is not one biometric modality that is best forall implementations. Many factors must be taken into accountwhen implementing a biometric device including location,security risks, task (identification or verification), expectednumber of users, user circumstances, existing data, etc. It is alsoimportant to note that biometric modalities are in varying stagesof maturity.

Fingerprint Recognition

Manual comparison of fingerprints for recognition has been in use

for many years, and has become an automated biometricidentification technique over the past two decades. Fingerprintshave an uneven surface of ridges and valleys that form a uniquepattern for each individual. For most applications, the primaryinterest is in the ridge patterns on the top joint of the finger.

Figure 1: Fingerprint Recognition1

An important distinction to make is the difference between theFBIs IAFIS2 system and the commercial fingerprint systems usedfor verification purposes. The FBI IAFIS system was developed to

compare submitted fingerprint information against a database ofseveral million fingerprints to determine if the individual haspreviously submitted fingerprints, and thus has a potentialcriminal history. IAFIS systems require information from all tenfingers, either ink-based or electronic, and preferably rolledimpressions. Submitted fingerprints are compared against thefingerprints on file and are verified by 0, 1, or 2 fingerprintexaminers. The process usually takes about two hours.



3/10

Biometrics Overview

Commercial fingerprint systems that are used for verificationpurposes usually require only one finger to compare thefingerprint to the one on file to confirm the individuals claimedidentity. This process is completely automated and usually takes

less than a second. The two types of systems are not connectedat all.

Face Recognition

Humans recognize familiar faces with considerable ease, but theyare not good at recognizing unfamiliar individuals. Since the1960s, machine vision researchers have been developingautomated methods for recognizing individuals via their facialcharacteristics. Despite the volumes of research, there are noagreed-upon methods for automated face recognition as there arefor fingerprints. Multiple approaches have existed for several

years using low resolution 2D images. Recent work in highresolution 2D and 3D shows the potential to greatly improve facerecognition accuracy.

Iris Recognition

The iris is the colored portion of an individuals eye. The conceptof using the iris for recognition purposes dates back to 1936.3 Thenext major advancement appeared in the late 1980s, with apatent being issued in 1994 for the algorithms that can performiris recognition automatically. To obtain a good image of the iris,identification systems typically illuminate the iris with near-infrared light, which can be observed by most cameras yet is notdetectable by, nor can itcause injury to, humans.A commonmisconception is thatiris recognition shines alaser on the eye toscan it. This isincorrect untrue. Irisrecognition simply takesan illuminated pictureof the iris withoutcausing any discomfortto the individual. Figure 2: Iris Recognition.4

Hand/Finger Geometry

One of the first successful commercial biometric products was ahand geometry system. Typically, a user enters a PIN code toclaim an identity, and then places his/her hand on the system,



4/10

Biometrics Overview

which takes a picture of the hand. Using mirrors, the pictureshows the view of the hand from the top and side. Measurementsare then taken on the digits of the hand and compared to thosecollected at enrollment.

Figure 3: Hand Geometry.5,6

Other Biometric Identification Systems

Many other identification methods are in various stages ofdevelopment and/or commercialization. Following are someexamples.


Speaker recognition uses an individuals speech, afeature influenced by both the physical structure of anindividuals vocal tract and the behavioralcharacteristics of the individual, for recognitionpurposes.

Dynamic Signature measures the speed and pressureone uses when signing his or her name (not what thesignature looks like).

Keystroke dynamics measures the typing patterns of anindividual.

Retina recognition takes an image of the back of theeye and compares blood vessels with existing data.

Gait/Body recognition measures how someone appearsas he or she walks. As in face recognition, this

technique is one that humans intuitively use torecognize someone.7

Facial Thermography measures how heat dissipates offthe face of an individual.


5/10

Biometrics Overview


Testing and Statistics

The accuracy of a biometric system is determined through a seriesof tests, beginning with an assessment of matching algorithm

accuracy (technology evaluation), then assessing performance in amock environment (scenario evaluation), followed by live testingon site (operational evaluation) before full operations begin.Each evaluation serves a different purpose and involves differenttypes of analyses.

Biometric terms, such as recognition, verification andidentification, are sometimes used randomly. This is not onlyconfusing, but incorrect as each term has a different meaning.

Recognition is a generic term and does not necessarilyimply either verification or identification. All biometric

systems perform recognition to again know aperson who has been previously enrolled.2

Verification is a task where the biometric systemattempts to confirm an individuals claimed identity bycomparing a submitted sample to one or morepreviously enrolled templates.

Identification is a task where the biometric systemattempts to determine the identity of an individual. Abiometric is collected and compared to all thetemplates in a database. Identification is closed-set

if the person is known to exist in the database. Inopen-set identification, sometimes referred to as awatchlist, the person is not guaranteed to exist in thedatabase. The system must determine if the person isin the database.

Because of these variances, different statistics must be used foreach task.

Verification

False Acceptance Rate (FAR)

The percentage of times a system produces a false accept,which occurs when an individual is incorrectly matched toanother individuals existing biometric. Example: Frankclaims to be John and the system verifies the claim.

Verification Rate

The rate at which legitimate end-users are correctlyverified.


6/10

Biometrics Overview


Open-Set Identification (Watchlist)

False Alarm Rate

The percentage of times an alarm is incorrectly sounded on

an individual who is not in the biometric systems database(the system alarms on Frank when Frank is not in thedatabase), or an alarm is sounded but the wrong person isidentified (the system alarms on John when John is in thedatabase, but the system thinks John is Steve).

Detection and Identification Rate

The rate at which individuals who are in a database cause asystem alarm and are properly identified in an open-setidentification (watchlist) application.

Closed-set Identification

Identification Rate

The rate at which an individual in a database is correctlyidentified.

Standards

Standards help users deploy and maintain their systems in aneasier manner, while also promoting longevity and enablinginteroperability. There are numerous national and international

efforts developing standards for:technical interfaces

data interchange formats

testing and reporting

societal issues

Conclusion

The NSTC Subcommittee on Biometrics developed this

introductory material in order to better communicate both withinthe government and with other interested parties. Stating factsand discussing related issues in a consistent, understandablemanner, will enable smoother integration of privacy-protectivebiometric solutions. Federal agencies are working to ensure thattheir outreach activities are consistent with, and occasionallyreference, this suite of documents so that the public, press andCongress are able to easily understand their plans and discuss


7/10

Biometrics Overview


them productively. The Subcommittee encourages other entitieto also use and reference this material.

This document serves as a general introd

s

uction to the field of.biometrics; other documents describe key items in more detail

These include:

Biometrics Frequently Asked Questions

Biometrics Glossary

Biometrics History

Biometrics Overview

Biometrics StandardsDynamic Signature

Face Recognition

itionFingerprint Recogn

Hand GeometryIris Recognition

nitionPalm Print Recog

Speaker Recognitiond StatisticsBiometrics Testing an

Vascular Pattern RecognitionThese chttp://www e

The Privacy of Biometricsdo uments are available at:

.biometricscatalog.org/NSTCSubcommitte .

ocument ReferencesD

1 International Biometric Group.

2

Orlans, and Peter T. Higgins,John D. Woodward, Jr., Nicholas M.Biometrics (New York: McGraw Hill Osborne, 2003).

3 Nalini Ratha and Ruud Bolle, Automatic Fingerprint RecognitionSystems (Springer: New York, 2004).

4 James Wayman et al, Biometric Systems Technology, Design andPerformance Evaluation (London: Springer, 2005).


8/10

Biometrics Overview


5 Maltoni, Davide, Maio, Jain, and Prabhakar, Handbook ofFingerprint Recognition (Springer: New York, 2005).

6 Secugen Biometrics Solutions.

7 Biometrics: Department of Def

.

8 Manfred Bromba, Bioidentification: Frequently AskedQuestions.

9 Anil K. Jain, Ruud Bolle, and Sharath Pankanti, PersonalIdentification in a Networked Society (Kluwer Academic

nd Technology Council

Publishing: Massachusetts, 1999).

About the National Science a

The National Science and Technology Council (NSTC) wasestablished by Executive Order on November 23, 1993. This

ecutivee

Cabinet

ars in a

the

der

ion is

Cabinet-level Council is the principal means within the exbranch to coordinate science and technology policy across thdiverse entities that make up the Federal research anddevelopment enterprise. Chaired by the President, the

membership of the NSTC is made up of the Vice President, theDirector of the Office of Science and Technology Policy,Secretaries and Agency Heads with significant science andtechnology responsibilities, and other White House officials.

A primary objective of the NSTC is the establishment of clenational goals for Federal science and technology investmentbroad array of areas spanning virtually all the mission areas ofexecutive branch. The Council prepares research anddevelopment strategies that are coordinated across Federalagencies to form investment packages aimed at accomplishingmultiple national goals. The work of the NSTC is organized un

four primary committees; Science, Technology, Environment andNatural Resources and Homeland and National Security. Each ofthese committees oversees a number of sub-committees andinteragency working groups focused on different aspects ofscience and technology and working to coordinate the variousagencies across the federal government. Additional informatavailable at http://ostp.gov/nstc.
http://www.secugen.com/images/faq02.gifhttp://www.bromba.com/faq/fpfaqe.htm#Fingerprint-Sensorenhttp://www.bromba.com/faq/fpfaqe.htm#Fingerprint-Sensorenhttp://ostp.gov/nstchttp://ostp.gov/nstchttp://www.bromba.com/faq/fpfaqe.htm#Fingerprint-Sensorenhttp://www.bromba.com/faq/fpfaqe.htm#Fingerprint-Sensorenhttp://www.secugen.com/images/faq02.gif


9/10

Biometrics Overview


About the Subcommittee on Biometrics

Biometrics is a technology that is rapidly becoming a usefule Federal

Government. Although the Federal Government is using theandocy

es

security, cost-savings and convenience tool for th

technology for many applications now, further developmentassessment is required to improve the technologys utility. Taddress these issues, the Office of Science & Technology Poli(OSTP) created the NSTC Subcommittee on Biometrics, reportingto the National Science & Technology Council (NSTC) Committeon Technology and Homeland & National Security. Additionalinformation is available athttp://www.biometricscatalog.org/NSTCSubcommittee.

Subcommittee on Biometrics

Co-chair: Duane Blackburn (OSTP)

epard (FBI Contractor)

Co-chair: Chris Miles (DOJ)

Co-chair: Brad Wing (DHS)

Executive Secretary: Kim Sh

Department Leads

Mr. Jon Atkins (DOS)F)

(EOP)

C)

Ms. Usha Karne (SSA)Dr. Michael King (IC)

SA)

Dr. Sankar Basu (NS

Mr. Duane Blackburn

Ms. Zaida Candelario(Treasury)

Dr. Joseph Guzman (DoD)

Dr. Martin Herman (DO

Mr. Chris Miles (DOJ)

Mr. David Temoshok (G

Mr. Brad Wing (DHS)

Mr. Jim Zok (DOT)
http://www.biometricscatalog.org/NSTCSubcommitteehttp://www.biometricscatalog.org/NSTCSubcommittee


10/10

Biometrics Overview


Communications ICP Team

Champion: Duane Blackburn (OSTP)

Members & Support Staff:

Mr. Richard Bailey (NSAContractor)

Mr. Jeffrey Dunn (NSA)

Ms. Valerie Lively (DHS S&T)

Mr. John Mayer-Splain (DHSUS-VISIT Contractor)

Ms. Susan Sexton (FAA)

Ms. Kim Shepard (FBIContractor)

Mr. Scott Swann (FBI)

Ms. Kimberly Weissman (DHSUS-VISIT)

Mr. Brad Wing (DHS US-VISIT)

Mr. David Young (FAA)

Mr. Jim Zok (DOT)

Document Source

This document, and others developed by the NSTC Subcommitteeon Biometrics, can be found athttp://www.biometricscatalog.org/NSTCSubcommittee.
http://www.biometricscatalog.org/NSTCSubcommitteehttp://www.biometricscatalog.org/NSTCSubcommittee

bio metrics overview

Documents