Binary spectral minutiae representation with multi-sample fusion for fingerprint recognition

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    5 Citations (Scopus)

    Abstract

    Biometric fusion is the approach to improve the biometric system performance by combining multiple sources of biometric information. The binary spectral minutiae representation is a method to represent a fingerprint minutiae set as a fixed-length binary string. This binary representation has the advantages of a fast operation and a small template storage. It also enables the combination of a biometric system with template protection schemes that require a fixed-length feature vector as input. In this paper, based on the spectral minutiae representation algorithm, we investigate the multi-sample fusion algorithms at the feature-, score-, and decision-level respectively. Furthermore, we propose different schemes to mask out unreliable bits. The algorithms are evaluated on the FVC2000-DB2 database and showed promising results.
    Original languageUndefined
    Title of host publicationProceedings of the 12th ACM Workshop on Multimedia and Security
    Place of PublicationNew York, NY, USA
    PublisherAssociation for Computing Machinery (ACM)
    Pages73-80
    Number of pages7
    ISBN (Print)978-1-4503-0286-9
    DOIs
    Publication statusPublished - Sep 2010
    Event12th ACM Workshop on Multimedia and Security, MM&Sec 2010 - Rome, Italy
    Duration: 9 Sep 201010 Sep 2010
    Conference number: 12

    Publication series

    Name
    PublisherACM

    Workshop

    Workshop12th ACM Workshop on Multimedia and Security, MM&Sec 2010
    Abbreviated titleMM&Sec
    CountryItaly
    CityRome
    Period9/09/1010/09/10

    Keywords

    • METIS-271066
    • IR-73657
    • Finger print
    • EWI-18582
    • Template Protection
    • SCS-Safety
    • Biometrics

    Cite this