An approach to human iris recognition using quantitative analysis of image features and machine learning

TL;DR

This paper presents a comprehensive iris recognition framework combining segmentation, feature extraction, reduction, and neural network classification, achieving high accuracy on a large dataset for reliable human identification.

Contribution

It introduces a novel multi-step iris recognition approach integrating advanced image analysis and machine learning techniques, demonstrating superior accuracy.

Findings

Achieved 99.64% recognition accuracy

Effective feature extraction with multiple methods

Reliable identification on CASIA-Iris-Interval dataset

Abstract

The Iris pattern is a unique biological feature for each individual, making it a valuable and powerful tool for human identification. In this paper, an efficient framework for iris recognition is proposed in four steps. (1) Iris segmentation (using a relative total variation combined with Coarse Iris Localization), (2) feature extraction (using Shape&density, FFT, GLCM, GLDM, and Wavelet), (3) feature reduction (employing Kernel-PCA) and (4) classification (applying multi-layer neural network) to classify 2000 iris images of CASIA-Iris-Interval dataset obtained from 200 volunteers. The results confirm that the proposed scheme can provide a reliable prediction with an accuracy of up to 99.64%.