Show simple item record

dc.contributor.advisorBusch, Christoph
dc.contributor.advisorFranke, Katrin
dc.contributor.authorSchuch, Patrick
dc.date.accessioned2019-10-29T13:30:45Z
dc.date.available2019-10-29T13:30:45Z
dc.date.issued2019
dc.identifier.isbn978-82-326-4081-2
dc.identifier.issn1503-8181
dc.identifier.urihttp://hdl.handle.net/11250/2625184
dc.description.abstractSummary Biometric recognition is a typical means to identify individuals or to verify claimed identities. Use cases are manifold. For example, users can unlock their smartphones for convenience by presenting their faces or fingerprints. Or one’s identity is verified when crossing borders. Today, biometric recognition already has many points of contact with our daily life and there are more to come. Besides iris and face, fingerprint is the most wide spread biometric trait used for recognition. Fingerprints are assumed to be unique for each and every finger. This makes it an ideal trait for recognition. In addition, fingerprint recognition has more than a century of tradition in the field of biometric recognition. A great amount of expertise and engineering skill made it a quite mature technology over time. Only few false positive and false negative errors are made in recognition in today’s deployed systems. However, fingerprint recognition is still far from being perfect. In contrast to popular opinion, fingerprint recognition is not a solved problem. Actually, there is still a lot of work to do. As biometric systems become larger and become more inclusive, even new challenges arise. Systems need to deal with large amounts of data while keeping performance with respect to recognition performance as well as transaction times in a reasonable order. Recognition shall work for everyone and shall not exclude a certain ethnic group or subset of the population. It will work in unconstrained conditions. However, it shall still make no erroneous decisions. Engineering may have come to its limits at this stage. In contrast to classical engineering, machine learning based on artificial neural networks may be a reasonable alternative. The emerging technologies of Deep Learning achieve tremendous successes in many domains of image processing and pattern recognition. This work assesses the application of such innovative machine learning concepts to fingerprint recognition. Three central aspects and challenges in fingerprint recognition are inspected in detail: fingerprint sample enhancement, orientation field estimation, and efficient processing structures.nb_NO
dc.description.abstractSammendrag Biometrisk gjenkjenning er en vanlig måte brukt for å identifisere enkeltpersoner eller for å verifisere en hevdet identitet. Bruksområdene er mangfoldige. For eksempel kan brukere låse opp smarttelefoner på en enkel må te ved å presentere deres ansikt eller fingeravtrykk. Et annet eksempel er når ens identitet blir verifisert når man krysser en landegrense. I dag kommer vi i kontakt med biometrisk gjenkjenning flere steder i det daglige liv, og det vil komme flere fremover. Utenom iris og ansikt, er fingeravtrykk det biometriske kjennetegnet som er mest vanlig når det kommer til gjenkjenning. Fingeravtrykk er antatt å være unik for hver enkelt finger. Dette gjør det til et ideelt kjennetegn for gjenkjenning. I tillegg har fingeravtrykk mer enn hundre års tradisjon innenfor biometrisk gjenkjenning. En stor andel ekspertise og teknisk utvikling har gjort det til en ganske moden teknologi over tid. Kun noen få falske positive og falske negative feil blir gjort i systemer som er tatt i bruk i dag. Derimot er gjenkjenning av fingeravtrykk fremdeles langt fra å være perfekt. I motsetning til hva folk flest tror, er ikke gjenkjenning basert på fingeravtrykk et løst problem. Det er fremdeles mye arbeid som gjenstår. Etter hvert som biometriske systemer blir større og mer inklusive oppstår det nye utfordringer. Systemer må håndtere store mengder data samtidig som ytelse når det gjelder gjenkjenning og transaksjonstider opprettholdes på et rimelig nivå. Gjenkjenning skal virke for alle og skal ikke ekskludere basert på etnisitet eller undergrupper i befolkningen. Det skal virke under alle forhold. Likevel, skal det ikke gjøres noen feilaktige beslutninger. Den tekniske utviklingen kan ha nådd sine begrensninger på dette stadiet. I motsetning til klassisk ingeniørvitenskap kan maskinlæring basert på kunstige nevrale nettverk være et rimelig alternativ. Den fremvoksende teknologien innenfor dyp læring oppnår fremragende suksess innenfor flere områder av bildeprosessering og mønster gjenkjenning. Dette arbeidet undersøker anvendelsen av slike innovative maskinlæringskonsepter på gjenkjenning av fingeravtrykk. Tre sent rale aspekter og utfordringer innenfor fingeravtrykk gjenkjenning blir vurdert i detalj: Forbedring av fingeravtrykkprøver, estimering av orienteringsfelt, og effektive prosesseringsstrukturernb_NO
dc.language.isoengnb_NO
dc.publisherNTNUnb_NO
dc.relation.ispartofseriesDoctoral theses at NTNU;2019:242
dc.relation.haspartPaper 1: Schuch, Patrick; Schulz, Simon; Busch, Christoph. Survey on the impact of fingerprint image enhancement. IET Biometrics 2018),7(2):102 - (c) Institution of Engineering and Technology The copy of record is available at the IET Digital Library http://dx.doi.org/10.1049/iet-bmt.2016.0088
dc.relation.haspartPaper 2: Schuch, Patrick; Schulz, Simon; Busch, Christoph. De-Convolutional Auto-Encoder for Enhancement of Fingerprint Samples. I: 2016 Sixth International Conference on Image Processing Theory, Tools and Applications (IPTA). IEEE 2017 http://doi.org/10.1109/IPTA.2016.7821036 - © 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
dc.relation.haspartPaper 3: Schuch, Patrick; Schulz, Simon; Busch, Christoph. Minutia-based Enhancement of Fingerprint Samples. I: 2017 International Carnahan Conference on Security Technology (ICCST). IEEE conference proceedings 2017 https://doi,org/10.1109/CCST.2017.8167824 - © 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
dc.relation.haspartPaper 4: Schuch, Patrick; Schulz, Simon-Daniel; Busch, Christoph. ConvNet Regression for Fingerprint Orientations. I: Image Analysis 20th Scandinavian Conference, SCIA 2017 2017 Proceedings, Part I. Springer ISBN 978-3-319-59125-4. s. 325-336 - The final published version available at https://doi.org/10.1007/978-3-319-59126-1_27
dc.relation.haspartPaper 5: Schuch, Patrick; Schulz, Simon; Busch, Christoph. Deep Expectation for Estimation of Fingerprint Orientation Fields. IJCB 2017 https://doi.org/10.1109/BTAS.2017.8272697 - © 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
dc.relation.haspartPaper 6: Schuch, Patrick; Schulz, Simon-Daniel; Busch, Christoph. Intrinsic Limitations of Fingerprint Orientation Estimation. BIOSIG 2017 https://doi.org/10.23919/BIOSIG.2017.8053513 - © 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
dc.relation.haspartPaper 7: Schuch, Patrick. Survey on features for fingerprint indexing. IET Biometrics 2018 ;Volum 8.(1) - (c) Institution of Engineering and Technology The copy of record is available at the IET Digital Library http://dx.doi.org/10.1049/iet-bmt.2017.0279
dc.relation.haspartPaper 8: Schuch, Patrick; May, Jan Marek; Busch, Christoph. Unsupervised Learning of Fingerprint Rotations. 2018 International Conference of the Biometrics Special Interest Group (BIOSIG) https://doi.org/10.23919/BIOSIG.2018.8553096 - © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
dc.relation.haspartPaper 9: Schuch, Patrick; May, Jan Marek; Busch, Christoph. Learning Neighbourhoods for Fingerprint Indexing. SITIS 2018 14th International Conference on Signal-Image Technology & Internet-Based Systems https://doi.org/10.1109/SITIS.2018.00018 - © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
dc.relation.haspartPaper 10: Schuch, Patrick; May, Jan Marek; Busch, Christoph. Estimating the Data Origin of Fingerprint Samples. BIOSIG 2018 http://dx.doi/10.23919/BIOSIG.2018.8553096 - © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
dc.titleDeep Learning for Fingerprint Recognition Systemsnb_NO
dc.typeDoctoral thesisnb_NO
dc.subject.nsiVDP::Technology: 500::Information and communication technology: 550::Computer technology: 551nb_NO


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record