Extracting and Combining Multimodal Directional Iris Features

  • Chul-Hyun Park
  • Joon-Jae Lee
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3832)


In this paper, we deal with extracting and combining multimodal iris features for person verification. In multibiometric approaches, finding reasonably disjoint features and effective combining methods are crucial. The proposed method considers the directional characteristics of iris patterns as critical features, and first decomposes an iris image into several directional subbands using a directional filter bank (DFB), then generates two kinds of feature vectors from the directional subbands. One is the binarized output features of the directional subbands on multiple scales and the other is the blockwise directional energy features. The former is relatively robust to changes in illumination or image contrast because it uses the directional zero crossing information of the directional subbands, whereas the latter provides another form of rich directional information though it is a bit sensitive to contrast change. Matching is performed separately between the same kind of feature vectors and the final decision is made by combining the matching scores based on the accuracy of each method. Experimental results show that the two kinds of feature vectors used in this paper are reasonably complementary and the combining method is effective.


Feature Vector Iris Image Equal Error Rate Iris Feature False Acceptance Rate 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Daugman, J.G.: High Confidence Visual Recognition of Persons by a Test of Statistical Independence. IEEE Trans. Pattern Anal. Machine Intell. 15(11), 1148–1161 (1993)CrossRefGoogle Scholar
  2. 2.
    Wildes, R.P.: Iris Recognition: An Emerging Biometric Technology. Proc. IEEE 85(9), 1348–1363 (1997)CrossRefGoogle Scholar
  3. 3.
    Boles, W.W., Boashash, B.: A Human Identification Technique Using Images of the Iris and Wavelet Transform. IEEE Trans. Signal Processing 46(4), 1185-1988 (1998)CrossRefGoogle Scholar
  4. 4.
    Lim, S., Lee, K., Byeon, O., Kim, T.: Efficient Iris Recognition through Improvement of Feature Vector and Classifier. ETRI Journal 23(2), 61–70 (2001)CrossRefGoogle Scholar
  5. 5.
    Park, S., Smith, M.J.T., Mersereau, R.M.: Improved Structures of Maximally Decimated Directional Filter Banks for Spatial Image Analysis. IEEE Trans. Image Processing 13(11), 1424–1431 (2004)CrossRefGoogle Scholar
  6. 6.
    Park, C.-H., Lee, J.-J., Oh, S.-K., Song, Y.-C., Choi, D.-H., Park, K.-H.: Iris Feature Extraction and Matching Based on Multiscale and Directional Image Representation. In: Griffin, L.D., Lillholm, M. (eds.) Scale-Space 2003. LNCS, vol. 2695, pp. 576–583. Springer, Heidelberg (2003)CrossRefGoogle Scholar
  7. 7.
    Park, C.-H., Lee, J.-J., Smith, M.J.T., Park, K.-H.: Iris-Based Personal Authentication Using a Normalized Directional Energy Feature. In: Kittler, J., Nixon, M.S. (eds.) AVBPA 2003. LNCS, vol. 2688, pp. 224–232. Springer, Heidelberg (2003)CrossRefGoogle Scholar
  8. 8.
    Rosiles, J.G., Smith, M.J.T.: Texture Classification with a Biorthogonal Directional FilteBank. In: Proc. IEEE Intl. Conf. on Acoustics, Speech, and Signal Processing, vol. 3, pp. 1549–1552 (2001)Google Scholar
  9. 9.
    Ma, L., Tan, T., Wang, Y., Zhang, D.: Personal Identification Based on Iris Texture Analysis. IEEE Trans. Pattern Anal. Machine Intell. 25(12), 1519–1533 (2003)CrossRefGoogle Scholar
  10. 10.
    Jain, A.K., Ross, A.: Multibiometric Systems. Communications of the ACM, Special Issue on Multimodal Interfaces 47(1), 34–40 (2004)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Chul-Hyun Park
    • 1
  • Joon-Jae Lee
    • 2
  1. 1.School of Electrical and Computer EngineeringPurdue UniversityWest LafayetteUSA
  2. 2.Dept. of Computer and Information EngineeringDongseo UniversityBusanKorea

Personalised recommendations