Implementasi Sistem Keamanan Pintu Otomatis Berbasis Face Recognition di Proactive Robotic: Integrasi ESP32-Cam dan Telegram

Abstract

The research focuses on implementing an automatic door lock system based on face recognition using the ESP32-Cam microcontroller and integrating it with the Telegram platform. The system is designed to enhance security at Proactive Robotic by leveraging IoT technology for office door access control. System design began with needs analysis and identification of existing security issues which do not integrate with other smart home systems at the institution. Development methods included hardware and software design, circuit schematic creation, PCB production, and physical assembly. System testing ensured facial detection, recognition, and essential solenoid operation in varying lighting conditions. Test results showed the system accurately recognizes registered faces under standard lighting, automatically unlocks doors, and sends real-time notifications via Telegram. However, system accuracy under low-light conditions needs improvement, as well as enhanced user data security to protect stored information privacy. This research contributes to developing IoT and face recognition-based security applications. The system successfully implements cutting-edge technology to improve physical security and access management in office environments, laying the foundation for future advancements in security technology.The research focuses on implementing an automatic door lock system based on face recognition using the ESP32-Cam microcontroller and integrating it with the Telegram platform. The system is designed to enhance security at Proactive Robotic by leveraging IoT technology for office door access control. System design began with needs analysis and identification of existing security issues which do not integrate with other smart home systems at the institution. Development methods included hardware and software design, circuit schematic creation, PCB production, and physical assembly. System testing ensured facial detection, recognition, and essential solenoid operation in varying lighting conditions. Test results showed the system accurately recognizes registered faces under standard lighting, automatically unlocks doors, and sends real-time notifications via Telegram. However, system accuracy under low-light conditions needs improvement, as well as enhanced user data security to protect stored information privacy. This research contributes to developing IoT and face recognition-based security applications. The system successfully implements cutting-edge technology to improve physical security and access management in office environments, laying the foundation for future advancements in security technology.