Latentfingerprintsaretheimpressionsofpartialridgesleftonthesurfaceofobjects touched unintentionally at crime scenes and constitute a valuable source of evidence in law enforcement agencies to helpsolve crimes. However,majority of the processing (marking regionof interest(ROI),singularpoints(SP),orientationfieldandminutiaepoints)for latentprintsidentificationisdonemanuallybyforensicexperts.Theexistingmethods involve forensicexpertstomanually markthe featuresin latentandtheninput ittothe systemforautomaticmatchingwithreferenceprints.Thematcherreturnsalistof candidates thatare manuallycheckedbyexperts totake final decision.Thepracticeof manuallymarkingfeaturesinlatentsislaborious,timeconsumingandhumandependent whichmayresultsinwrongidentification.Thereforethereisaneedtoautomatethis process to avoid aforesaidconstraints. This thesis describes an automatedapproach of segmentation andenhancement for latent fingerprints identification.Currently,afew attemptshavebeenmade inthisrespectand still remain a challenging problem due to: (i) poor quality, (ii) small friction ridge area,(iii) presence of non-linear distortion,(iv) blurringor smudging,and(v) complexbackground noise. In this research, an algorithm for automated segmentation of latent fingerprints is proposed. The latent image datais classifiedintoclusters using K-means clustering techniquewhich results in pixels having similar characteristicsto fit in one cluster (foreground) while pixels having opposite characteristicsto other cluster (background).Tophat filtering is appliedto enhance the clustered data and mask is generated on the basis of this enhanced information. Segmentation is achievedby applying thegenerated maskon latent image.The proposed algorithm forsegmentationof latent fingerprintsisautomatedwithout anysortof human involvement.Performanceofproposedalgorithmisevaluatedbycomputingthemissed detection rate (MDR) and false detection rate (FDR) and comparison of proposed method withotherexistingalgorithmsisdone.SimulationresultsonNISTSD-27(databaseof latentfingerprintimagescontaining258latentfingerprintsalongwiththeirmatedrolled prints)showsignificantperformanceenhancementofproposedmethodhavingaverage MDR and FDR of 4.77% and 26.06% respectively. Furthermore, subjective comparison is made usingvisual segmentation reliability(VSR)which istheratioof intersectional area ofautomated and ground truthlatent to manually marked segmented latent. VSR approaches to 90% for good quality images, 70-80% for bad quality images and 50-60% for ugly qualityimages.Matchingperformance isimprovedwhen thesegmentedinputis applied to commercial-off-the-shelf(COTS tenprint)matcherascompared with un- segmentedinput. Another contribution of proposed research is towards the enhancement of latent fingerprints. Enhancement of segmented latent isperformed using Gabor filter bank. It has five image-dependent-parameterslikeorientation ∅ ,standarddeviations ?�and ?�ofthe Gaussian function,timeperiodTandthe convolutionmasksize.Theselectionof these parametersplaysacrucial roleinfingerprintenhancementspecificallytheorientation∅ andstandarddeviation ?� and ?� .Thelatentimageis dividedintoblocks of WxW centered at pixel (i,j)and gradients ��� and ��� along x-axis andy-axis are computedby applying Sobel operator at everypixel.Orientation ∅ is computedon the basis ofthese computed gradients. Ridge frequency F(i, j) is estimated by calculating the grey level value of each pixel, housed in the block, and is projected in a direction perpendicular to the local ridgeorientationandridgespacingS(i,j).Animprovementinfrequencyestimationis achievedbyintroducingGaussianlowpassfilterthatminimizesthenoiselevels.Ridge orientation ∅ and frequency F(i, j) is used to design an even-symmetric Gabor filter. Spatial convolution of thelatentfingerprint withGaborfilter isperformedtogenerateenhanced latentimage.SimulationresultsonNISTSD-27showthatimprovementinmatchingis increased 11% in comparison toautomatedlatent fingerprint segmentation and enhancementalgorithmby Zhang et al in 2013.
Chapters
Title |
Author |
Supervisor |
Degree |
Institute |
Title |
Author |
Supervisor |
Degree |
Institute |
Title |
Author |
Supervisor |
Degree |
Institute |
Title |
Author |
Supervisor |
Degree |
Institute |
Book |
Author(s) |
Year |
Publisher |
Book |
Author(s) |
Year |
Publisher |
Chapter |
Author(s) |
Book |
Book Authors |
Year |
Publisher |
Chapter |
Author(s) |
Book |
Book Authors |
Year |
Publisher |
Similar News
Headline |
Date |
News Paper |
Country |
Headline |
Date |
News Paper |
Country |
Similar Articles
Article Title |
Authors |
Journal |
Vol Info |
Language |
Article Title |
Authors |
Journal |
Vol Info |
Language |
Similar Article Headings
Heading |
Article Title |
Authors |
Journal |
Vol Info |
Heading |
Article Title |
Authors |
Journal |
Vol Info |