• Mano Sri Ram K
• Abilash C R
• Sreehari K
• Mr. Jeya Ganesan J

The rapid advancement of Artificial Intelligence (AI) has significantly enhanced surveillance and security systems in both defense and civilian domains. This paper presents Sentinel-Face, an AI-based facial recognition and behavioral analysis framework designed to strengthen identity verification and anomaly detection in high-security environments. The proposed system integrates Haar Cascade-based face detection with deep learning-based feature extraction using Convolutional Neural Networks (CNNs) and triplet loss optimization for accurate identity verification. Beyond recognition, the system incorporates LSTM-based behavioral analysis and Vision-Language Models (SmolVLM) to monitor unusual activities and detect potential threats in real time. The framework supports criminal registration, continuous monitoring through CCTV integration, and automated alert generation for unauthorized access or suspicious behavior. Experimental results demonstrate a face detection rate of 98.2%, recognition accuracy of 96.5%, and real-time processing capability at 12 frames per second per camera. The modular and scalable architecture ensures adaptability across defense zones, public infrastructures, and smart surveillance systems while addressing concerns related to privacy and data protection. The proposed solution contributes toward intelligent, automated, and proactive security management systems.

Artificial Intelligence (AI), Facial Recognition, Behavioral Analysis, Convolutional Neural Network (CNN), Triplet Loss, Haar Cascade, LSTM, SmolVLM, Surveillance Systems, Anomaly Detection, Real-Time Monitoring.

"Sentinel-Face: AI-Based Facial Recognition and Behavior Analysis", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.13, Issue 3, page no.c316-c320, March-2026, Available :http://www.jetir.org/papers/JETIR2603240.pdf

"Sentinel-Face: AI-Based Facial Recognition and Behavior Analysis", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.13, Issue 3, page no. ppc316-c320, March-2026, Available at : http://www.jetir.org/papers/JETIR2603240.pdf