Selected Topics CSE 891
Popular in Course
Popular in Computer Science and Engineering
This 86 page Class Notes was uploaded by Donnell Kertzmann on Saturday September 19, 2015. The Class Notes belongs to CSE 891 at Michigan State University taught by Anil Jain in Fall. Since its upload, it has received 67 views. For similar materials see /class/207402/cse-891-michigan-state-university in Computer Science and Engineering at Michigan State University.
Reviews for Selected Topics
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/19/15
Biometric Recognition joimcsoumsomodo http biometricscsemsu edu vl w a identification 0 Identifying fellow human beings has been crucial to the fabric of human society 0 In the early days of civilization people lived in small communities and everyone knew each other 0 With the population growth and increase in mobility we started relying on documents and secrets to establish identity 0 Person identification is now an integral part of the infrastructure needed for diverse business sectors such as banking border control law enforcement Jain 2004 Person Identification Associating an identity with a person is called personal identification Jain 2004 Questions related to the identity of an individual Is this the person he or she claims to be Has this applicant been here before Should this individual be given access to our system Are the rendered services being accessed only by a legitimate user Jain 2004 clerltlflcaltlon Problem Security Threats We now live in a global society of increasingly desperate and dangerous people whom we can no longer trust based on identification documents which may have been compromised Senator Terrorist A Watch List Stops Kennedy at Airport Senator Edward M Kennedy Democrat of Mass discussed the problems faced by ordinary citizens mistakenly placed on terrorist watch lists Between March 1 and April 6 airline agents tried to block Mr Kennedy from boarding airplanes on five occasions because his name resembled an alias used by a suspected terrorist who had been barred from flying on airlines in the United States RACHEL L SWARNSNY TimesAug 20 2004 Jain 2004 ltiderl tii39iication Problems Identity Theft Identity thieves steal PIN eg date of birth to open credit card accounts withdraw money from accounts and take out loans 33 million identity thefts in US in 2002 67 million victims of credit card fraud Surrogate representations of identity such as passwords and ID cards no longer suffice Jain 2004 Too Many Passwo ds to Remember Entry about the odor I have all my passwnrds tattooed between my toes 0 Heavy web users have an average of 21 passwords 81 of users select a common password and 30 write their passwords down or store them in a file 2002 NTA Monitor Passwordsurvey 1m2004 The Secret PIN 391 WV wmw M39M on 09M am am 7 an aaomc mm NAH39W usaer THE B LIISFB 139 by An amp Chip Sunsum v39 39ma u wNmrALoroFrmubHY L quotx HAD TO sugar vx acme w CODL rOK THE NEW MM ND ku SEWLED on ma MN ANV me wow mu coma Y 5 wmua L rm N 7 1 N ALL mm Maemw AMPJ 5 W7 T Camer 9 2m uma Fezmre swam m 39A 1 What Is Biometrics Jain 2004 Blorrletrics Automatic recognition of people based on their distinctive anatomical eg face fingerprint iris retina hand geometry and behavioral eg signature gait characteristics Recognition of a person by his body then linking that body to an externally established identity forms a very powerful tool Iris based ATM access Jain 2004 Security Level a Why Biometrica Something you know I Something you have Method Jain 2004 BiOm etric F u nationalities Positive Identification Is this person truly known 1 y to the system 37 Provide login access to a valid user Query image Templateimage Vincent Vincent 0 Large Scale Identification Is this person in the database Prevent issuing multiple driver licenses to the same person 0 Screening Is this a wanted person Query image Airport watchlist Only biometrics can provide negative identification ie I am not he capability Jain 2004 0 Advantages Discourages fraud Enhances security Cannot be easily transferred forgotten lost or copied Eliminates repudiation claims Imparts convenience o Disadvantages Output is a matching score instead of a YesNo decision Cannot be reset if compromised unlike passwords Biometric system itself could be attacked Privacy concerns Jain 2004 F ml m trl ill at e w Bertillon system 1882 took a subject s photograph and recorded height the length of one foot an arm and index finger GaltonHenry system of fingerprint classification adopted by Scotland Yard in 1900 0 FBI set up a fingerprint identification division in 1924 AFIS installed in 1965 with a database of 810000 fingerprints First face recognition paper published in 1971 Goldstein et al 0 FBI installed IAFIS in 2000 with a database of 47 million 10 prints average of 50000 searches per day 15 of searches are in lights out mode 2 hour response time for criminal search Emphasis now is to automatically perform reliable person identification in unattended mode often remotely or at a distance Jain 2004 F oertlllon item The Bertillon system 1882 entailed photographing the subject looking directly at the camera then in profile with the camera centred upon the right ear Besides the two photographs the subject39s height was recorded together with the length of one foot an arm and index finger httpwwwtdjcueduauhiststatsbert Jain 2004 Biometric Applications Jain 2004 3 Bioirrleitrl c AppJ oat on Corpse National ID Card ATM Identification Criminal Driver s License Access Control Investigation Voter Registration Computer Login Parenthood Welfare Cellular Phone Determination Disbursement Missing Border Crossing E commerce Children US VISIT program Internet Banking Smart Card There are 500 million border crossingsyear in the US Jain 2004 Homeland Secur1ty IlSVISI39I39quot 39 Ev wwwdhsgovusvisit an information is then used to assist the a a border inspector in determining whether or V not to admit the traveler These enhanced intsmnner edlt procedures will add only seconds to the VISItor 5 overall processmg time The swarm Mam allows inspectors to identities of visitors against those on terrorist Watdi lists By Stephen J Buitanu AP From the DHS USVISI T 39t Iain 2004 4 Km 6 Al l l liiprlal Biometric ID icaglli cls UK to consider national biometric ID cards database By Laura Rohde COMPUTERWORLD Nov 29 2003 The UK government is set to consider legislation next year for the establishment of compulsory biometric identity cards and a central database of all UK subjects it was announced by the government this week The information that the government is considering for inclusion on the card includes personal details such as a person39s home address and telephone number his National Insurance number the equivalent of the US Social Security number medical information and criminal convictions as well as the biometric information most likely in the form of an iris fingerprint or palm print scan Jain 2004 L 3b ii iii Applicationa r itl EEz39i i Fingerprint at check out counter Cell Phone with Fingerprint Sensor earr r 39t ems Ben Gurion Airpor Smart gun Jain 2004 ACCESS CQFJ EFQJ mm WWWhEgnp cum Iain 2004 PAPERBALLUTERA denWWuvw m 2 2mm Enquwev Jam 2004 xgpliee itb rng Video Surveillance Online or offline Face Scan at Airports 1 52 1 9 hA 739 wt a V 39 ji Wivi ri E3 7 I I39f 1 if I 39 5 fr if 1 1 iquot 5 1 394 i l7 r I39 quoti i vb3 m L H a r j quot39 39 1 39 V 39w a a 7 a 39 a f 39 a 39 u 391 f quot 3 I M quotZ39Iv i r u g f e I I r 1 7 I b a 39 i Iv a a r a v a a r V 1 k I 4 The 51 P h 39 39 39 39 39 39 I gtqu nan meters The passengers vases instantly are compared to a database of images 39 39 5 first people to pass through the new security system f d Jain 2004 Galp Energia SGPS SA of Lisbon won the technology innovation award for developing a payment system in which gasolinestation customers can settle their bills simply by pressing a thumb against a glass pad Scanning technology identifies the thumbprint and sends the customer39s identification information into Galp39s backoffice system for payment authorizationquot THE WALL STREEi JOURNAL November 15 2004 Imn 2004 noun muE The Nine Zero hotel in Boston just installed a new system which uses digital photos of the irises of employees vendors and VIP guests to admit them to certain areas the same system used in high security areas at airports such as New York39s JFK USATODAV 7222004 Jam 2004 Foreigners entering the United State in three cities including Port Huron were ngerprinted photographed and subjected to background checks on Monday in a test of a program that will eventually be extended to every land border crossing nationwidequot Lansing State Journal Nov 15 2004 lam 2004 New Passports lCAO TAGMRTDNTWG RESOLUTION N001 Berlin 28 June 2002 ICAO TAGMRTDINTWG endorses the use ofas the globally interoperable biometric for machine assisted len y con lrmation with machine readable travel documents ICAO TAGMRTDINTWG further recognizes that Member States may elect to use fingerprint andor iris recognition as additional biometric technologies in support of machine assisted identity confirmation Endorsement Unanimous httpwwwicaoint PASSPORT The new passports have an embedded contactless ISO 14443 smartcard chip that stores personal information and a biometric quotmm template Two problems reliability and privacy of Amexlca Jain 2004 Beepcard a company in California has designed a credit card that works only when it recognizes the voice of its rightful owner Enclosed in the card is a tiny microphone a loudspeaker and a speech recognition chip that compares the spoken password with a recorded sample If the voices match the card emits a set of beeps that authorize a transaction over the telephone or the Internet If the voices do not match the card will not beep The system tolerates some variations in voice to accommodate cold or background noise But it might not work if there is a blaring music in the background Imn2004 N secure JVlultirnedia 0 Pirates can easily distribute multimedia data without being tracked Copyright owners are reluctant to release their content without appropriate protection mechanisms 0 Watermarking Not very robust to attacks Encryption Illegal key exchange 0 Use biometric characteristics as the key Alice now needs Bob s biometric data in order to decrypt the file This requires an invariant representation of the biometric Jain 2004 u Biometrics for Personalization Automatic personalization of vehicle settings Seat position Steering wheel position Mirror positions Lighting Radio station preferences Climate control settings httpwww visteoncom 0 URLs at your fingertips Jain 2004 Arrlsterdam Schip39nol Airport Frequent flyers can sign up for the so called Privium club which not only allows members to bypass the lengthy line at passport control but also lets them park closer to the departure hall and use fast check in counters rm 2004 Application Clolalr acterlatlc Uli o Overt vs Covert 0 Attended vs Unattended o Cooperative vs Non cooperative o Scalable vs Non scalable o Habituated vs Non habituated 0 Private vs Public 0 Closed vs Open 0 Lowlevel vs Highlevel security Jain 2004 b g lhg 1 9 h f arm 4 him f EPDJJCCJCIOFJ Crlcll cJCer gfle o Overt vs Covert 0 Attended vs Unattended o Cooperative vs Non cooperative o Scalable vs Non scalable o Habituated vs Non habituated 0 Private vs Public 0 Closed vs Open 0 Reduce vs Eliminate fraud Jain 2004 9 Domams of Appljcation DDMAIN S OF APPLICATIDN 2003 Physical Ancess CDII39TI EI39DI 29 Computer 41 45 Financial 19 All other 11 Jain 2004 Biorrletric Traits Jain 2004 F Behavioral vs P a Physical Characteristics Iris ReUna Vein Pattern Hand Geometry Face Fingerprint Behavioral Characteristics Keystroke dynamics Signature dynamics 0 Voice Jain 2004 Jain 2004 o One dimensional unique code for one s individuality but identical twins have identical DNA patterns 0 Three issues limit the utility of DNA 0 Contamination 0 Automatic real time recognition issues 0 Privacy issues information about susceptibilities of a person to certain diseases could be gained from the DNA pattern Jain 2004 Cost f eat accurajcy g I s Hand ReUna Signature Hn er Face 9 Vote Accuracy IEEE Computer Jain 2004 M3151 Ln 1 39x1 mt indudln AFIS mmui 1mm F593 In 5 m r 39 in I vx iq madmanIf 1 quotam th p i 13 513 If FIE 111 I uw g hl ch 1M IMOm l nll Biamu Bump Jain 2004 31mm 3 tr C H ark at Growth Prepare to be scalare10 quotmmquot mm gggngET cs m moan mman alouE rn nEvEuuss NissanN Us sV MILLIONS Z Z S g g 1353 mu 4 a mum a rAc was 350 a mu anus sang MILLmN Hm Enmmv Zoo 3 mum 2008 354 MILLION Gaq nls on 33 mum zoos suns WLLmN mlnnLEWAn a zuu 46 um u muchworn 20a 5am on 200D 22 mum 75 vnlc zooa gzsyuLLoN SIGNATURE zoo sewn on was syn L Inn FINGERPRI Irr rnn cm mumL nternau om Biomemc Group Iain 2004 IiK FLPLVT 135 D H v uli iich Ejiorheir if 32h 0 Universality all users possess this biometric o Uniqueness varies across users 0 Permanence does not change over time o Collectability can be measured quantitatively 0 Performance low error rates and processing time o Acceptability is it acceptable to the users 0 Circumvention can it be easily spoofed No biometric is optimal Need to find the best match between a specific biometric and the application Jain 2004 1 CdiTipiirihd Eiidme im Fechrididgieg Biemetrics Un iVersality Uniqueness Permanence Cnilectabi lity Performance Acceptabi lit y Circumven ticn Face High 1 Low Medium High Low High Low Fingerprint Medium High High Medium High Medium Low Hand Medium Medium Medium High Medium Medium Medium Geometry Iris High High High Medium High Low High Retinal Scan High High Medium Low High Low High Signature Low Low Low High Low High Low Voice Medium Low Low Medium Low High Low Facial High High Low High Medium High High Thermogram There is no optimal biometric Jain 2004 Design and Performance Evaluation Jain 2004 Biometrics A Pattern Recognition System YesNo 4 Matcher Threshold t I Feature Extractor Template Database Authencation Enrollment False accept rate FAR Proportion of imposters accepted False reject rate FRR Proportion of genuine users rejected Faiure to enroll rate FI39E portion of population that cannot be enrolled Faiure to acquire rate FI39A portion of population that cannot be verified Jain2004 o Sensor 0 Feature extraction 0 Matcher 0 System Data base Jain 2004 Biometrics A Grand Challenge A fundamental problem in science and engineering with broad economic and scientific Impact Scale 101 Unusable Hard to Use Easy to Use 9 Transparent to User 9 99 99999 90 9999 Accuracy UsabHHy The grand challenge is to design a biometric system that would operate on the extremes of all these three axes simultaneously Jain 2004 Should meet specified recognition accuracy speed and resource requirements Be harmless to the users Be accepted by the intended population Must be sufficiently robust to various fraudulent attacks on the system Jain 2004 v lf rlat a Terri pl a A set of features extracted from the raw biometric data of an individual Represents typical value of a biometric Multiple templates per individual are often stored to account for intra class variations Template aging requires that templates should be updated over time Where should the template be stored central database or a smart cardmagnetic card How to protect the template Encryption Jain 2004 EXEJJ HDJES Oi Terripiates Fixedlength vector 65 53 59 52 62 4747 45 255 333 253 287 243 149 Hand feature set Lengths and widths of fingers width of palm Fixedlength vector 31591 44110 21235 9078 84012 43474 Face feature set Eigencoefficients 35 150 10 401 703 45 14234 50145 6 5116618 Fingerprint feature set Minutiae coordinates and local ridge orientation Jain 2004 Per rfo f T fl a n E v l u a ti o n o The overall performance of a biometric system is assessed in terms of its universality accuracy speed and storage 0 Factors like cost and ease of use also affect performance 0 Biometric systems are not perfect and can mistakenly accept an impostor as a valid user a false match or conversely reject a valid individual a false non match Best Practices wwwcesqqovuktechnoloqvbiometrics FRVT2002wwwfrvtorg FVC 2004 biascsruniboitfvc2004 NIST SV wwwnistqovsoeechtestssok Jain 2004 P l Perior39rrlarlce shal acterlzaClon o Impostor Distribution 0 Genuine Distribution 0 Threshold False Accept Rate FAR or False Match Rate False Reject Rate FRR or False Non match Rate Receiver Operating Characteristic ROC curve Equal Error Rate or Crossover Rate Failure to Enroll FI39E Failure to Acquire FTA or Failure to Capture Jain 2004 Error Rates Genuine Dismbminn ImpaslnrDislribminn FalseRejed False AmePI False Match False Accept Mistaking biometric measurements from two different persons to be from the same person False Nonmatch False reject Mistaking two biometric measurements from the same person to be from two different persons lmn 2004 Error vs Threshold Equal Error Rate Degree mum Human Jam 2004 False Accept Rate FAR ROC Curve Forensic Applications quot5 Equal Error Rate Civilian I A p plicati a n1 I High Emu city Access Applications I False Reject Rate FER Accuracy requirements of a biometric system are application dependent Iain 2004 rd A l 254 l 4 oval l u a ltli l Protocol 0 Define a protocol to test the system select the data and measure the performance 0 Evaluations should be conducted by an independent organization that is not involved in the design of the system 0 Test on biometric data previously unseen by the system 0 Size of the data set and representative examples of the data set should be provided for tuning algorithmic parameters 0 Face Fingerprint and Voice systems have undergone the most study and testing Jain 2004 Jain 2004 U illlby is biometric Recog ni tion Difficult Large number of classes eg millions of faces Intra class variability and inter class similarity Segmentation Noisy and distorted images Population coverage amp scalability System performance error rate speed cost Attacks on the biometric system Template ageing Non uniqueness of biometric characteristics Addressing privacy concerns Jain 2004 U lt Sl 2 m C Ity N39teu age Seven After an 1I1irlyLuven quotThanc an aicmxsmnms armquot 11 Age Mm or inwxiwaim um can MN 11 gnusun39s ranu ufkll quotmuer I mncnmi quot Jain 2004 Imtarclass Similarity Jam 2004 Jain 2004 Temporal Variatme Ti Time duration 2 yearS Jain 2004 Theo Pa vidis httphomeattnett pa vlidiscomphumanscomphuman htm Jain 2004 a Crowd a Games Magazine September 2001 Jain 2004 l Joi3y Images 0 3 of the population has poor quality fingerprint images Four impressions of a user s fingerprint Faded ngiawffifiiiis cost former welder a job wa39lmn Pam DECATUR The ears Chuck Suickler spent as a weld er provided 39 39 rience he needed as a welding inspector at power plants across c nation But the welding also has left Strickler 60 of Decatur lacking rims 1e has attesting to his identity no longer are 5 den I rst ran into a problem with it three or four years agoquot Strickler said quotThey said my ngerprints weren39t wlidr But at the time they accepted a picture B o 2 5ch psychological examina and criminal background check quotThe 39 3 a 39 Strickler needed It urns a e If A person has about 100 iden ti cationmarksonhis or er v gerprints and most adults have about 80 that can be use to 39 and em But because of his welding work Stliclder has only about u m 5 to am a inh inm a m up I I t I L I r r Bridgman where he had worked or the nuclear halides D harm Le r ln t bec he said Iain 2004 22522 avzega eea raisge 0 20o O 200 2F3OC4 chrElljrptfar gis 23907 207 ESE1 Xiihe jo39ri ggg 10 1 Egg 5 Indegee tdent 10392000 2500 At NY airports an average of 300000 passengers pass through daily If all of these used biometricauthenticated smart cards for identification there would be 600 falsely rejected and inconvenienced passengers per day for fingerprints 30000 for face and 45000 for voice Similar numbers can be computed for false accepts Jain 2004 Wj llillf lgj List Performance Metrics and Evaluation Fingerprint Verification Competition FVC 2000 2002 2004 httpbiascsruniboitfvc2004 Fingerprint Vendor Technology Evaluation FpVTE 2003 httpfpvtenistgov Face Recognition Vendor Tests FRVT 2000 2002 httpwwwfrvtorg Face Verification Contest on the BANCA dataset 2004 httpwwweesurreyacukbancaicpr2004 NIST Speaker Recognition Evaluations 1996 2004 httpwwwnistgovspeechtestsspk Signature Verification Competition SVC 2004 httpwwwcsusthksvc2004 Jain 2004 IVJLJJUDiQJ FJEtFiC39 Uplicnl rm Minuliac Filterbased mu waw an guvus wsl Iam 2004 i i v J Li iii o J om c3 Limited discrimination and nonuniversality of a biometric Hand geometry Fingerprint 7 7 Helps improve accuracy and popul y L V m ation coverage Jam 2004 SerialCascade Mode Parallel Mode Hierarchical Mode Jain 2004 0 Three commercial fingerprint matchers and one face matcher with EER values of 396 372 216 and 376 respectively were combined 0 972 individuals in the database oThe best EER values in individual columns rows are indicated with bold typeface star Normalization Fusion Technique TeChnique Sum Min Max MW uw MinMax 099 543 086 116 063 ZScore 171 528 179 172 186 Tanh 173 465 182 150 162 QLQ 094 543 063 116 063 MW Matcher Weighting UW User Specific Weights Jain 2004 ii f39ZchLirity of Diorr ietr ic i quot Ei7ii fiquoti i Like any security system biometric systems are not foolproof 3 Override Feature Extractor 6 Modify Template l lYesNo 2 ReplaYI 4 Synthesized 1 Fake Old Data l f ft rTf itoi Biometric 5 Override Matcher Jain 2004 U Attacks on Biometric Sygtem Spoofing a biometric trait Enlarge View Dummy finger created from a lifted impression Artificial skinfingers httpwwwivingskincom Jain 2004 5m ummmmgm quotmm Iain 2004 Digital Biometrics optical sensor 508x480 Veridicom capacitive sensor Fidelica pressure sensor 300x300 256x256 Sensors used during enrollment and verification may be different Jain 2004 V i39l 7 ensor I n terooerabiiity Sensors used during enrollment and verification may be different Jain 2004 i o Biometrics Big Brother 0 Will biometric data be used to track people secretly violating their right to privacy 0 Biometric data may provide medical history eg retinal data may divulge information about diabetes or hypertension 0 Will biometric data be used only for their intended purpose Will biometric systems be linked o Government s interest is motivated by desire to decrease fraud and waste increasing efficiency and public safety 0 Trade off between public safety and personal freedom 0 Government regulation may be needed to safeguard biometric data 0 By restricting access to personal information biometric actually provides effective privacy protection Jain 2004 xx g rm 4x y cw K0quot 1 4n 1 dewlcb ujwli F i i r T Biometric can help in protecting individual privacy because biometrics provides stronger identification than password it can be used to guard personal amp sensitive information Health Information Privacy Protection Act Will biometric data be used to track people secretly violating their right to privacy Functionality creep Will biometric data be used only for their intended purpose Will various biometric databases be linked MINORITY REPCDKF Fmquot 7quotquot Ti RUNS JUNE 21 ARE quotDU 7 THAT 5 NILMNG TD 1 DEPENM TRAPEQWL F ON THE 3 meshes exg ANEE FQR RATE 3 secumr r V I L e 739 ZPZCJ gH l4 r 7 7 r 4 V 39 V 39 M9 Jain 2004 National ID Card 1e credibcard companies or H16 Minela are keenly sensi ve lo a we just as on 39 card years is proposing a a uniform 200 million 5 liaenses iver s License has be personal ID in lhe coun we assume you re legit yearly anyone can get a 3 did Having those lie cover to mingle in ing detected rgues let s make it the chief require to drive This The Case for a National ID Card Big Brother already knows where you live Why not let him make you safer Time January 21 2002 Jain 2004 a iii on all 1D rd 0 ID card would contain basic information about the holder including their SSN and a photograph along with biometric data will be used to verify the identity of the holder and check her legal status 0 The question is not whether the government should issue ID cards and maintain databases they already do The question is whether the ones we have can be made more effective especially when it comes to finding criminalsquot Larry Ellison W5 0 A recent Harris Poll found that 68 of Americans favor some form of national identification Jain 2004 My Ynatamvmnmmxammm al Solutions new quotVerichipquot about the size of a gra of rlce seen at left In this handout image is the rstever computer ID chip that could be embedded beneath a persons skin YahunINews 27 Feb 3902 mn2004 ym1aao ARE quotDU 7 THAT 5 NILMNG TD 1 DEPENM TRAPEQWL F ON THE 3 meshes exg ANEE FQR RATE 3 secumr r V I L e 739 ZPZCJ gH l4 r 7 7 r 4 V 39 V 39 M9 Jain 2004 r 4 M mquot p rquot 39 m g HCJIC Card Jl id DID mama Biometric template resides in the personal smart card of a user Biometric verification takes place via a built in chip on the card Template does not leave the card no centralized biometric database is required Precise Biometrics 5m Sense from Veridicom Siemens Matcher on Card Version 11 lam 2004 I 4 ReligioLiaCulturai QDJections Mail vi en in ESOE Q39 13 OE A i MARK OF THE BEAST wig f d 1 39 39 39 or on 39 umber ofhis name Here is wisdom Let him who his war cm 2 M g z W E x Q mixng as e J r 5391 Li a a a w g g g A a 9 m Orlando Sentinel This is no different than acceptance of some other technologies Jain 2004
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'