Human iris recognition under uncontrolled environment using convolutional neural network and compare accuracy with hidden Markov model classifier

The primary goal of this research is to evaluate the accuracy of human iris recognition using a Convolutional Neural Network in an uncontrolled environment image to that of a Hidden Mark ov Model. Convolutional neural network (CNN) model and Hidden Markov model (HMM) were used to segment and recogni...

Full description

Saved in:
Bibliographic Details
Main Authors: Karthik, B., Ramkumar, G.
Format: Conference Proceeding
Language:English
Subjects:
Accuracy
Artificial neural networks
Classifiers
Markov chains
Statistical analysis
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The primary goal of this research is to evaluate the accuracy of human iris recognition using a Convolutional Neural Network in an uncontrolled environment image to that of a Hidden Mark ov Model. Convolutional neural network (CNN) model and Hidden Markov model (HMM) were used to segment and recognise iris in uncontrolled environment photos with 50 samples per group using images from the MMU iris dataset. Group 1 had 25 samples taken using CNN, while group 2 had 25 samples taken using HMM and analysed using gpower 80 percent. In a MATLAB simulation, CNN segmented iris with a precision of 95%, while AAM segmented iris with a precision of 78%. In SPSS statistical analysis, significant accuracy of P 0.05 was achieved. In iris recognition tasks, the suggested CNN performs much better than the HMM classifier for the supplied images.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0158597