Facial Biometrics for Access Control in a Company: A Systematic Review

Escriba Huaman J.P.; Monzón Porras M.R.; Sanchez Atuncar G. · Lecture Notes in Networks and Systems (2026) · DOI: 10.1007/978-3-032-22236-7_18

Resumen

This systematic review analyzes 60 studies (2020–2024) from Scopus and Web of Science on facial biometric systems for corporate access control, comparing their efficiency and security against traditional authentication methods. Facial recognition emerges as the predominant approach (40.3% of cases), with optimized models such as ECN + GWO + SAE reaching 97.29% accuracy in real-world settings. Multimodal frameworks demonstrate even stronger performance (up to 99.47% accuracy and 68% fraud reduction), particularly when integrated with deep neural networks and FMCW sensors. Compared to passwords and ID cards, biometric systems show clear advantages: faster authentication (40% improvement), reduced direct attacks (13.9%), and lower vulnerability to social engineering (12.1%). However, limitations persist in scenarios with poor lighting or low-resolution inputs (74.7% accuracy), and few studies critically assess statistical significance or methodological bias. Overall, the evidence suggests that hybrid systems—combining facial recognition with multifactor authentication and liveness detection—offer the most effective enterprise solutions, while future research should address technical gaps and standardization to ensure scalability and reliability. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

Instituciones: Universidad César Vallejo