IRIS Recognition based on PCA based Dimensionality Reduction and SVM

There are several Iris recognition0 techniques. But method proposed by Daugman is considered to be most efficient technique for IRIS segmentation and feature extraction. Recent studies have shown that there is better classifier which when properly trained with sufficient numbers of features are bett...

Full description

Saved in:
Bibliographic Details
Published in:International journal of computer applications 2012-01, Vol.49 (3), p.28-31
Main Authors: Tiwari, Upasana, Kelkar, Deepali, Tiwari, Abhishek
Format: Article
Language:English
Subjects:
Classifiers
Feature extraction
Optimization
Recognition
Reduction
Segments
Support vector machines
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:There are several Iris recognition0 techniques. But method proposed by Daugman is considered to be most efficient technique for IRIS segmentation and feature extraction. Recent studies have shown that there is better classifier which when properly trained with sufficient numbers of features are better than the hamming distance based classifier. But more number of features increases the computational complexity due to the need for feature optimization by kernel based classifiers. Hence in this work we propose a unique technique of first extracting huge numbers of features from the IRIS images and then reducing the features by using PCA based linear dimensionality reduction technique. We first segment the IRIS images with a technique proposed by Daugman, further Gabor features are extracted from the segmented IRIS image. These features are reduced using feature reduction technique. The features are classified using Multiclass support vector machine. We show that the accuracy of the IRIS recognition technique is very high using this method.
ISSN:0975-8887
0975-8887
DOI:10.5120/7609-0652