Iris Recognition using Color Models with Artificial Neural Network
S. B. Kulkarni1, U. P. Kulkarni2, Siddu Tushara M. S.3
1Dr. S B Kulkarni, Asst. Prof, Department of Computer Science and Engineering, S.D.M College of Engineering and Technology, Dharwad, Karnataka, India.
2Dr.U P. Kulkarni, Prof, Department of Computer Science and Engineering, S.D.M College of Engineering and Technology, Dharwad, Karnataka, India.
3Siddu Tushara M S, M. Tech, Department of Computer Science and Engineering, SDM College of Engineering and Technology, Dharwad, Karnataka, India.
Manuscript received on May 22, 2014. | Revised Manuscript received on June 05, 2014. | Manuscript published on June 30, 2014. | PP: 339-341 | Volume-3, Issue-5, June 2014. | Retrieval Number: E3254063514/2013©BEIESP
Open Access | Ethics and Policies | Cite
© 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: Biometrics plays a vital role for an extensive array of highly secure identification and personal verification systems. Iris Recognition is the recognition of an individual based on iris features. It is regarded as the most promising biometric identification system available. In this paper, the iris recognition is applied on UBIRIS database. Image is segmented using circular Hough transform, then converted into a fixed sized rectangular block using Daugman’s Rubber sheet model. Iris features are extracted using CMYK color model and a feature vector is created using 2D Walsh Hadamard transform, finally these are classified based on Artificial Neural Network(ANN) using MLP. Based on the database size ROC(Receiver Operating Characteristic) curve is plotted using true positive rate and false positive rate in order to analyze for what size efficiency may be good.
Keywords: Artificial neural network, Biometrics, Receiver operating characteristic curve.