Fingerprint, seminar at IASRI, New Delhi
Upcoming SlideShare
Loading in...5
×
 

Fingerprint, seminar at IASRI, New Delhi

on

  • 2,874 views

 

Statistics

Views

Total Views
2,874
Views on SlideShare
2,869
Embed Views
5

Actions

Likes
2
Downloads
232
Comments
0

1 Embed 5

http://www.linkedin.com 5

Accessibility

Categories

Upload Details

Uploaded via as Microsoft PowerPoint

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

Fingerprint, seminar at IASRI, New Delhi Fingerprint, seminar at IASRI, New Delhi Presentation Transcript

  • Nishikant P. TaksandeIASRI, New Delhi, India
  • ContentsIntroductionBiometric RecognitionBiometric SystemsFingerprintsFingerprint Recognition SystemMinutiae ExtractionMinutiae MatchingApplicationsLimitationsConclusionsReferences
  • Introduction“Biometrics is the science of verifying and establishing identity of an individual through physiological features or behavioral traits.” Biometric is more about what you are than what you have or know Password and tokens are not reliable View slide
  • Introduction…Biometric has been used since 14thcentury in chinaEveryone have unique physical orbehavioral characteristics and so uniqueIdentityEnhanced convenience and augmentedsecurity measure View slide
  • Biometric RecognitionBiometric modalities Physical Biometrics: Face, Fingerprints, Iris-scans, Hand geometry Behavioral Biometrics: Speech, Signature, and Keystroke dynamics Chemical biometrics: Odor and the chemical composition of human perspiration
  • IrisIris is the area of the eyewhere the pigmented orcoloured circle, usuallybrown, blue, rings the darkpupil of the eyeImage is typically capturedusing a noncontact imaging Iris portionprocessIris should be at thepredetermined distancefrom the focal plane of thecamera
  • Face RecognitionMethod of acquiring faceimages is nonintrusiveFacial disguise is of concernin unattended recognitionapplicationsIt is challenging to developtechniques that can toleratethe effects of aging, facialexpression, variations in theimaging environment
  • Hand and Finger GeometryFeatures related to the humanhand are relatively invariant andpeculiar to an individualSystem requires cooperation ofthe subject to capture frontal andside view images of the palmflatly placed on a panel withoutstretched fingersOnly used for verification
  • Hand Vein RecognitionBack of a clenched fistused to determine handvein structureSystems for vein captureuse inexpensive infra-redlight emitting diodes
  • Voice RecognitionVoice capture is unobtrusive andonly feasible applicationsrequiring person recognitionover a telephoneVoice signal available forrecognition is typically degradedin quality by the microphone,communication channelVoice is affected by factors suchas a person‟s health, stress andemotional state
  • SignatureThe way a person signs hisname is known to be acharacteristic of that individualSignature is a behaviouralbiometric that changes overtimeProfessional forgers canreproduce signatures of othersto fool the unskilled eye
  • Biometric SystemsAn important issue in designing a practicalbiometric system is to determine how anindividual is going to be recognizedBiometric system Enrolment Verification system Identification system
  • Enrolment Subject Identifier Template Identifier Feature Template Extraction Data Creation Storage Sample Feature setCapture Enrolment Process
  • Verification Subject Identifier Claimed Identity Feature Extraction Matching Data Storage Sample Feature set One SubjectsCapture template Match or Non-Match Verification Process
  • Identification Subject Identifier Feature Pre-selection Extraction and Data Matching Storage Sample Feature set ‘N’ SubjectCapture Templates Subject Identified Or not Identified Identification Process
  • Architecture Biometric sensor Feature extraction Database Enrolment Biometric sensor Feature extraction Matching Authentication ResultGeneral architecture of a biometric system
  • Comparison of Biometric TraitsComparison of commonly used biometric traits
  • FingerprintSkin on human fingertips containsridges and valleys which togetherforms distinctive patternsThese patterns are fully developedunder pregnancy and are permanentthroughout whole lifetimeNo two persons have the samefingerprintsAutomatization of the fingerprintrecognition process turned out to besuccess in forensic applications
  • Fingerprint PatternsLeft Loop Right Loop
  • Fingerprint Patterns...Whorl Arch Twin Loop
  • Fingerprint Feature Minutiae Based ApproachMinutiae is the unique, measurable physicalpoint at which ridge bifurcate or ends Ridge Ending Ridge Bifurcation
  • Fingerprint Recognition System Minutiae Minutiae Extractor MatcherSensorArchitecture for Fingerprint Recognition System
  • Fingerprint Sensing and StorageOff- line scan or live scanA special kind of off-lineimages, extremelyimportant in forensicapplications, are the so-called latent fingerprintsfound at crime scenes
  • Live Scan DevicesFingerprint image acquisition isthe most critical step of anautomated fingerprintauthentication systemIdea behind each captureapproach is to measure in someway the physical differencebetween ridges and valleysPhysical principles likecapacitive, optical and thermalare used
  • Optical Sensors FTIR-based Fingerprint SensorFrustrated Total Internal Reflection (FTIR)
  • Minutia ExtractionPre-ProcessingFingerprint Image Enhancement Fingerprint Image enhancement is to make the image clearer for easy further operations Increase the contrast between ridges and furrows and connect the broken points
  • Minutia Extraction...Histogram EqualizationHistogram equalization is to expand the pixel valuedistribution of an image so as to increase theperceptional information The Original Histogram of a Histogram after histogram fingerprint image equalization
  • Minutia Extraction...Histogram Equalization Original Image Enhanced Image after Histogram Equalization
  • Minutia Extraction...Fingerprint ImageBinarization 8-bit Gray fingerprint image transform to a 1-bit image with 0-value for ridges and 1- value for furrows Ridges in the fingerprint are highlighted with black colour while furrows are white
  • Minutia Extraction...Fingerprint Image Segmentation Region of Interest (ROI) is useful to be recognized for each fingerprint image Two step method is based on Morphological methodsROI extraction by Morphologicaloperations Two Morphological operations called „OPEN‟ and „CLOSE‟ are adopted
  • Minutia Extraction...ROI extraction by Morphological operations Original Image Area After CLOSE operationAfter OPEN operation ROI + Bound
  • Minutia Extraction...Minutia ExtractionFingerprint Ridge Thinning Ridge Thinning is to eliminate the redundant pixels of ridgesMinutia Marking If the central pixel is 1 and has exactly 3 one-value neighbours, then the central pixel is a ridge branch If the central pixel is 1 and has only 1 one-value neighbour, then the central pixel is a ridge ending
  • Minutia Extraction...Minutia MarkingBifurcation Termination A fingerprint after minutiae extraction
  • Minutia Extraction...Constellation Creation Technique
  • Minutia Extraction...Constellation Creation Technique
  • Minutia Extraction...Constellation Creation Technique
  • Minutia Extraction...Constellation Creation Technique
  • The Matching ModuleThe matching module use a pre-processed patterncomposed by the fingerprint, it‟s minutia-file andit‟s constellations,These patterns are extracted from enrolledtemplatesFeature extraction is done then its patterns arecompared to the reference onesComparison between the two fingerprints Constellation matching Minutiae matching
  • Constellation MatchingWhen a new fingerprint is scanned it is passedby the feature extraction module, a set ofconstellations and their respective parametersare createdComparison with the genuine constellations setextracted during system enrolment F1 = {C1.0, C1.1, C1.2, C1.3} F2 = {C2.0, C2.1, C2.3, C2.4, C2.5}Euclidean distance between the point patternsof constellations centres
  • Constellation MatchingRejects a fingerprint template No constellation is matched D = 0 A unique constellation that includes less than 15 minutiae is matched D = 0 Co1.0 Co2.0 Co2.1 Co1.1 Co2.3 Co2.2 Co1.3 Co1.2 Fingerprint constellation matching
  • Minutiae MatchingMinutiae matching are performed within aconstellationThe minutiae matching are used as a secondlevel verificationminutiae matching algorithm proceeds asfollow Associate minutiae points Compute distance between minutiae Decide
  • Minutiae Matching Flow Chart Start Decision=reject Fingerprint and constellation Constellation files MatchingNO D<CTH CTH Minutiae MatchingNO D<MTH MTH Decision=Accept END
  • Applications of Fingerprint Recognition SystemPhysical Access ControlLogical Access ControlTransaction AuthenticationDevice Access ControlTime and AttendanceCivil IdentificationForensic Identification
  • LimitationsDirt , grime and woundsPlacement of fingerToo big database to processLiveness importantIn case of permanent finger injury
  • Barriers to AdoptionBusiness value is difficult to quantify in terms ofreturn on investmentFingerprint recognition systems, being an emergingtechnology, is sometimes confronted with unrealisticperformanceThe quality varies quite dramatically from oneapplication to another and from one vendor toanotherSeveral fingerprint system vendors are notfinancially stable
  • ConclusionsMost highly used methods for human recognitionFingerprint system strongly relies on the precisionobtained in the minutia extraction processIndustrial commitment led to next generationfingerprint technologyIt has got broad acceptance from forensic tohandheld devicesUsed for securing international borders
  • References[1] FBI Fingerprint guide. Available online at http://www.fbi.gov/hq/cjisd/takingfps.html[2] Fingerprint. Available online at http://en.wikipedia.org/wiki/Fingerprint[3] Fingerprint recognition. Available online at http://en.wikipedia.org/wiki/Fingerprint recognition[4] Goyal, A., Study of Fingerprints Minutia Extraction and matching Technique. Available online at http://www.advancedcenterpunjabi.org/MTechP/Anjna%20Goyal_ M.Tech(CS).pdf
  • References...[5] Human Fingerprints. Available online at http://ww.fingerprinting.com/human-fingerprints.php[6] Maltoni, D., Maio, D., Jain, A., Prabhakar, S.,(2009), Handbook of Fingerprint Recognition, second edition, Springer Limited.[7] Murmu, N., Otti, A., Fingerprint recognition. Available online at http://ethesis.nitrkl.ac.in/955/[8] Rokbani, N., Alimi, A., Fingerprint identification using minutiae constellation matching. Avialable online at http://www.regim.org/publications/conferences/2005/2005_MCCSIS_ RA.pdf