Iris Recognition Based on DWT and PCA

Author

Kumar, D. R. Shashi ; Raja, K.B. ; Nuthan, N. ; Sindhuja, B. ; Supriya, P. ; Chhotaray, R.K. ; Pattnaik, Sabyasachi

Author_Institution

Dept. of CSE, Cambridge Inst. of Technol., Bangalore, India

fYear

2011

fDate

7-9 Oct. 2011

Firstpage

489

Lastpage

493

Abstract

An Iris is unique compared to all other biometric and is used to authenticate a person. In this paper, we propose iris recognition based on DWT and PCA (IRDP). The iris region is localized using morphological process and a template is formed by using the left and right part of the iris to the pupil by discarding the upper and lower parts of the iris as they are usually occluded by eyelids and eyelashes. The DWT is applied on the histogram of iris template. The features are generated from the approximation band of the DWT using PCA. The classifiers viz., KNN, RF and SVM are used for matching. It is observed that the proposed algorithm has better performance parameters compared to existing algorithm.

Keywords

discrete wavelet transforms; image classification; image matching; iris recognition; principal component analysis; support vector machines; DWT; KNN; PCA; RF; SVM; classifiers; iris recognition; iris template histogram; morphological process; Databases; Discrete wavelet transforms; Feature extraction; Iris recognition; Principal component analysis; Vectors; Biometrics; DWT; Iris; KNN; Morphology; PCA; RF; SVM;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence and Communication Networks (CICN), 2011 International Conference on

Conference_Location

Gwalior

Print_ISBN

978-1-4577-2033-8

Type

conf

DOI

10.1109/CICN.2011.102

Filename

6112915