Implementasi Servo Lock System yang Terintegrasi Pengenalan Wajah untuk Keamanan Rak Senjata Otomatis

Abstract

Military armories require strict monitoring of their security systems to prevent unauthorized use. The purpose of this research is to implement and evaluate a prototype of an automated weapon rack locking system based on face recognition. A quantitative descriptive method with an experimental approach was used. The HIKVISION DS-K1T343MFX terminal serves as the authentication device, the ESP32 microcontroller as the main controller, and the MG996R servo motor as the actuator. Parameters including False Acceptance Rate (FAR), False Rejection Rate (FRR), response time, as well as distance and light intensity variations, were used to evaluate system performance. The test results show that the prototype has a high level of accuracy with a FAR of 0% and an FRR of 0%, and it is capable of rejecting unauthorized access attempts using facial photos (spoofing). The average system response time from the scanning process to unlocking is 2.20 seconds. The system operates optimally within a distance range of 30 cm to 150 cm, with the best point at a distance of 100 cm, and has proven to remain stable under various lighting conditions (morning, daylight, night). In conclusion, this face recognition-based servo lock prototype has been successfully implemented and has the potential to be an efficient alternative for supporting the management of defense equipment protection in military units.