Sistem Keamanan Loker Berbasis Verifikasi Wajah dan Notifikasi Telegram di Ponpes Baitussyakirin
TL;DR
The experimental results indicate that the face verification system achieves a 100% accuracy rate under bright lighting conditions and successfully rejects all unregistered faces, and the integration with the Telegram Bot is highly responsive, yielding an average notification latency of under 2 seconds for locker activities.
Abstract
The security of personal storage in Islamic boarding schools (pesantren) often faces challenges due to the vulnerabilities of conventional locker locking systems, which are prone to lost or duplicated manual keys. This study aims to design and implement a smart locker security system (Smart Lock) utilizing biometric-based face verification technology and automated notifications. The research method applied is Research and Development (R&D) with an iterative Prototyping model, conducted at Pondok Pesantren Ta'limil Qur'an Baitussyakirin, Kendal. The system is built by integrating the ESP32-CAM microcontroller, a servo motor as the locking actuator, and the Telegram API platform for two-way wireless communication. The experimental results indicate that the face verification system achieves a 100% accuracy rate under bright lighting conditions and successfully rejects all unregistered faces. The integration with the Telegram Bot is highly responsive, yielding an average notification latency of under 2 seconds for locker activities. Furthermore, the remote control feature via the /buka command proved to be an effective emergency mitigation function when the camera sensor experienced reading failures under total darkness. Overall, this prototype system is viable and reliable for enhancing real-time locker monitoring and security.