Share Email Print
cover

Proceedings Paper

Iris identification system based on Fourier coefficients and singular value decomposition
Author(s): Sawet Somnugpong; Suphakant Phimoltares; Saranya Maneeroj
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Nowadays, both personal identification and classification are very important. In order to identify the person for some security applications, physical or behavior-based characteristics of individuals with high uniqueness might be analyzed. Biometric becomes the mostly used in personal identification purpose. There are many types of biometric information currently used. In this work, iris, one kind of personal characteristics is considered because of its uniqueness and collectable. Recently, the problem of various iris recognition systems is the limitation of space to store the data in a variety of environments. This work proposes the iris recognition system with small-size of feature vector causing a reduction in space complexity term. For this experiment, each iris is presented in terms of frequency domain, and based on neural network classification model. First, Fast Fourier Transform (FFT) is used to compute the Discrete Fourier Coefficients of iris data in frequency domain. Once the iris data was transformed into frequency-domain matrix, Singular Value Decomposition (SVD) is used to reduce a size of the complex matrix to single vector. All of these vectors would be input for neural networks for the classification step. With this approach, the merit of our technique is that size of feature vector is smaller than that of other techniques with the acceptable level of accuracy when compared with other existing techniques.

Paper Details

Date Published: 13 January 2012
PDF: 8 pages
Proc. SPIE 8350, Fourth International Conference on Machine Vision (ICMV 2011): Computer Vision and Image Analysis; Pattern Recognition and Basic Technologies, 83501W (13 January 2012); doi: 10.1117/12.920979
Show Author Affiliations
Sawet Somnugpong, Chulalongkorn Univ. (Thailand)
Suphakant Phimoltares, Chulalongkorn Univ. (Thailand)
Saranya Maneeroj, Chulalongkorn Univ. (Thailand)


Published in SPIE Proceedings Vol. 8350:
Fourth International Conference on Machine Vision (ICMV 2011): Computer Vision and Image Analysis; Pattern Recognition and Basic Technologies
Safaa S. Mahmoud; Zhu Zeng; Yuting Li, Editor(s)

© SPIE. Terms of Use
Back to Top