The rise in missing person cases across many communities has created an urgent need for efficient, scalable, and intelligent identification systems. Conventional approaches such as manual search operations, posters, and media announcements are often slow, fragmented, and incapable of real-time response. This study presents the design and implementation of a real-time community-based missing person identification system using FaceNet and OpenCV. The system integrates deep learning-based facial recognition with real-time image processing to enable rapid detection of missing individuals from community cameras and user-submitted images. The architecture consists of face detection, feature embedding extraction, database matching, and automated alert notification modules. OpenCV employs face detection and preprocessing, while FaceNet generates 128-dimensional embeddings used for identity comparison within a structured database. The system supports community participation by allowing local users to submit sightings and monitor alerts, thereby decentralizing detection efforts. Experimental evaluation shows an average recognition accuracy of 94.2%, low false-positive rates, and near real-time processing performance. The findings demonstrate that embedding-based facial recognition significantly improves robustness against environmental variations such as lighting and pose changes. This research contributes to humanitarian technology by proposing a low-cost, scalable framework that empowers communities to participate in missing person identification while maintaining ethical safeguards related to privacy and controlled access.
Akintunde et al. (Wed,) studied this question.