An Enhanced Feature Extraction using Chebyshev Wavelet Filter for Iris Recognition System

Abstract

This research explores an enhanced feature extraction method for iris recognition, addressing the limitations of existing techniques like Gabor filters and wavelet transforms, which struggle with noise and image variations in non-ideal conditions. The objective of this research is to design and evaluate a feature extraction method using Chebyshev wavelet filter, testing its performance in terms of False Acceptance Rate and False Rejection Rate, and assessing its effectiveness in handling variations. The study utilizes three publicly available datasets (CASIA, MMU, and UBIRIS), the methodology involves preprocessing iris images from the datasets and applying the Chebyshev wavelet filter for feature extraction and employing support vector machine for classification. This proposed system aims to improve recognition accuracy and robustness in non-ideal conditions. The expected outcome is to achieve higher accuracy rates and lower FAR and FRR compared to traditional methods. This research contributes to the development of more efficient iris recognition systems.