Biometric System using Iris Pattern Recognition
Samarth S. Mabrukar1, Nitin S. Sonawane2, Jasmine A. Bagban3
1Samarth S. Mabrukar, Department of Electronics and Telecommunication, Sinhgad Academy of Engineering, Pune (Maharashtra), India.
2Nitin S. Sonawane, Department of Electronics and Telecommunication, Sinhgad Academy of Engineering, Pune (Maharashtra), India.
3Jasmine A. Bagban, Department of Electronics and Telecommunication, Sinhgad Academy of Engineering, Pune (Maharashtra), India.
Manuscript received on 15 April 2013 | Revised Manuscript received on 22 April 2013 | Manuscript Published on 30 April 2013 | PP: 54-57 | Volume-2 Issue-5, April 2013 | Retrieval Number: E0644032413/13©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Iris is unique body part which does not change with respect to time. Also every individual has unique and different pattern of the Iris for both the eyes. This helps in identifying a person, quite accurately. Initially, a filter must be employed to get rid of any kind of noises before pre-processing stage. Initially we detect the pupil-iris boundary. After that, we give it to Circular Hough transform to detect its center which will be used to extract iris from the image. Using Daugman’s Rubber sheet model, we normalize the iris pattern for making computations easy. Feature Extraction is done by using multi-scale Taylor series expansion of the iris texture. Feature vectors are extracted by binarizing the first and second order multi-scale Taylor coefficients. The proposed algorithm is tested against different images which gives better results in less computation time. The simulation is carried out using CASIA database on MATLAB.
Keywords: Hough Transform, Iris, Multi-Scale, Segmentation, Taylor Series Expansion.
Scope of the Article: Pattern Recognition