Iris recognition Using Directional Filter Bank 


Vol. 27,  No. 12, pp. 1208-1216, Dec.  2002


PDF Full-Text
  Abstract

We propose a new iris recognition method using a directional filter bank (DFB). The proposed method obtains accurate directional information using a DFB, and extracts the features on multiple scales by adding a low pass filter before the DFB, and then generates an iris code used for subsequent matching. The resultant subband outputs through the DFB are thresholded using a zero value and one part of an iris code is then generated by sampling the thresholded subband outputs. As such, the low pass filtered image is fed into the DFB to construct an additional code at different scales. The overall iris code is completed by combining the codes obtained at three different scales. Recognition is performed by finding the class to which the iris image with the minimum distance in the database belongs. Experimental results showed that the proposed method could be used successfully for personal identification and verification

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

C. Park, S. Oh, D. Kwak, j. Oh, J. Lee, K. Park, "Iris recognition Using Directional Filter Bank," The Journal of Korean Institute of Communications and Information Sciences, vol. 27, no. 12, pp. 1208-1216, 2002. DOI: .

[ACM Style]

Chul-hyun Park, Sang-keun Oh, Dong-min Kwak, jong-hwan Oh, Joon-jae Lee, and Kil-houm Park. 2002. Iris recognition Using Directional Filter Bank. The Journal of Korean Institute of Communications and Information Sciences, 27, 12, (2002), 1208-1216. DOI: .

[KICS Style]

Chul-hyun Park, Sang-keun Oh, Dong-min Kwak, jong-hwan Oh, Joon-jae Lee, Kil-houm Park, "Iris recognition Using Directional Filter Bank," The Journal of Korean Institute of Communications and Information Sciences, vol. 27, no. 12, pp. 1208-1216, 12. 2002.

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880