Machine Learning/Deep Learning in Biometric Systems
Mahesh K. Singh, Ankur Mittal · 2025
Biometric recognition systems have become most popular and trustworthy systems for organizations of all sizes. This chapter provides a comprehensive overview of biometric systems and recent trends, including fundamental concepts, physiological and behavioral traits, and their applications in security, authentication, and identification. It highlights the role of machine learning (ML) algorithms and in enhancing the efficiency, accuracy, and adaptability of biometric systems. Various ML methods, including deep learning, supervised learning, unsupervised learning, and reinforcement learning, are integrated with biometric systems, fingerprint recognition, voice verification, and iris and retina recognition verification, are also explored using ML methods like SVMs, Hidden Markov Models (HMMs), and deep neural networks. The chapter concludes with a summary of key attributes of different biometric recognition systems and discusses future prospects, especially privacy and ethical issues.