<![CDATA[Building access systems have enhanced security but often rely on costly foreign products. This research aims to develop an employee attendance application that integrates facial recognition using the Siamese network method with RFID for dual authentification. Designed to improve security and efficiency, the application allows for real-time monitoring and data collection of employee access without requiring additional hardware when expanding the system. The study designs an access system based on RFID and facial recognition with the Siamese network method. The system includes RFID nodes, a base station, a WiFi access point, a web dashboard, and a database. The RFID nodes read UID data, which is sent to the base station for verification. The system uses the ESP32 microcontroller and is integrated with a Next.js web server to monitor and manage employee attendance data. The test results indicate that the RFID can read RFID cards within a reading range of approximately 0 to 10 cm. The ESP-NOW communication can transfer data within a time range of 0 to 341 ms and a distance range of 50 to 400 cm between microcontrollers without using an external antenna. The HTTP communication protocol has a data transfer speed of approximately 700 ms with an internet download/upload speed of 10.5/5.41 Mbps. Facial recognition using the Siamese network model achieved an average threshold of 0.736 out of 0.5 under various lighting conditions and distances.]]>
Copyrights © 2025
