Development of a Facial Recognition System for Suspects for Integration into a Video Surveillance Security System with Direct Alerts Using a Mavic 2 Pro Drone: Case of Burkina Faso

This research presents a facial recognition system based on cameras installed in public spaces in Burkina Faso, which sends real-time alerts with the location and time of detection of the suspect. The objective of this research is to develop a security system solution capable of helping Burkina Faso combat terrorism. The system is developed and tested using a drone operating at medium altitude to simulate its operation before integration with the cameras. The system uses ArcFace, an open-source model compatible with facial differentiation constraints. The results achieved a detection rate of 92% and an accuracy of 90% under ideal conditions, while maintaining good performance in a dynamic environment. This research is important because Burkina Faso is experiencing periods of insecurity linked to terrorism. To help Burkina Faso, we are implementing this scalable solution to strengthen security and enable the monitoring of suspects, which will contribute to the strategy for combating terrorism in Burkina Faso. In this research, we design the system architecture, build a Python script and function, and test the system’s effectiveness. The result shows a perfect detection of the suspect.