Biometric image protection based on discrete cosine transform watermarking technique

Rizal Isa, Salem Aljareh

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    This paper presents combined face recognition and watermarking technique to secure a biometric image and maintain the recognition rate. These days, with advanced technology, the biometric data can be stolen and faked which may be used in other applications that utilize the same biometric feature. The discrete cosine transform (DCT) watermarking technique is used to certify the biometric image belong to the legitimate user. We have tested the proposed scheme under several attacks. The results of the experimentations show that the face recognition rate performance almost does not degrade due to watermark embedding and the proposed scheme is robust against several attacks.
    Original languageEnglish
    Title of host publicationProceedings of the 2012 International Conference on Engineering and Technology (ICET)
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    Pages1-5
    Number of pages5
    ISBN (Electronic)978-1467348102
    ISBN (Print)978-1467348089
    DOIs
    Publication statusPublished - 31 Dec 2012
    Event2012 International Conference on Engineering and Technology (ICET) - German University in Cairo (GUC), Cairo, Egypt
    Duration: 10 Oct 201211 Oct 2012

    Conference

    Conference2012 International Conference on Engineering and Technology (ICET)
    Country/TerritoryEgypt
    CityCairo
    Period10/10/1211/10/12

    Keywords

    • watermarking
    • face recognition
    • face
    • discrete cosine transforms
    • principal component analysis
    • robustness
    • biometrics
    • face recognition rate performance degradation
    • biometric image protection,
    • DCT watermarking technique
    • PCA

    Fingerprint

    Dive into the research topics of 'Biometric image protection based on discrete cosine transform watermarking technique'. Together they form a unique fingerprint.

    Cite this