Article Advances in Feature Extraction and Selection for Iris Recognition Systems: A Review

Abstract

The growing demand for security and reliable authentication methods has positioned biometric recognition systems at the forefront of modern security infrastructure. Among various biometric modalities, including facial and fingerprint, iris recognition has emerged as the most accurate and reliable, due to its rich texture, stability over time, and resistance to ageing and surgical alteration. However, existing iris recognition systems are primarily optimized for frontal, high-quality images and often fail to accurately process angular or partially captured iris images. To overcome these issues, techniques such as feature extraction, segmentation, and normalization need to be improved. Feature selection is essential in dimensionality reduction, and when combined with feature extraction techniques, the process will improve both time complexity and accuracy. This review explores the advancements in iris recognition, focusing on feature extraction and selection techniques that are crucial for improving system accuracy and efficiency. This study offers a comprehensive overview of various feature extraction and selection techniques, serving as a valuable reference for researchers.