• DocumentCode
    2529414
  • Title

    Iris Verification Algorithm Based on Texture Analysis and its Implementation on DSP

  • Author

    Fatt, R. ; Haur, T.Y. ; Mok Kai Ming

  • Author_Institution
    Comput. Vision & Intell. Syst. (CVIS) group, Univ. Tunku Abdul Rahman (UTAR), Petaling Jaya, Malaysia
  • fYear
    2009
  • fDate
    3-5 April 2009
  • Firstpage
    198
  • Lastpage
    202
  • Abstract
    This paper presents a Digital Signal Processor (DSP)implementation of the iris verification algorithm. Aninnovative method is presented to extract the iris features based on texture analysis. The algorithm was evaluated using CASIA iris image database version 1.0 [1]. Experimental results show that the approach has achieved high accuracy of 98.62%. The iris verification algorithm is designed to suit for DSP implementation. Templates are encoded in binary format because they are compact in size and can avoid the visibility of the individual iris images. The binary templates can be stored and processed effectively using DSP technology. The iris verification system is realized on state-of-the-art Analog Devices ADSP-BF561 to achieve fast verification performance. DSP technology provides a cost effective, low power and fast authentication solution within a compact hardware module.
  • Keywords
    biometrics (access control); digital signal processing chips; feature extraction; image recognition; image texture; Analog Devices ADSP-BF561; CASIA iris image database version 1.0; digital signal processor; iris verification algorithm; texture analysis; Algorithm design and analysis; Authentication; Costs; Digital signal processing; Digital signal processors; Feature extraction; Image databases; Image texture analysis; Iris; Signal processing algorithms; digital signal processor; iris verification; texture analysis;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Signal Acquisition and Processing, 2009. ICSAP 2009. International Conference on
  • Conference_Location
    Kuala Lumpur
  • Print_ISBN
    978-0-7695-3594-4
  • Type

    conf

  • DOI
    10.1109/ICSAP.2009.9
  • Filename
    5163854