Smart Door Lock Menggunakan Identifikasi Wajah dan Bot Telegram Sebagai Kendali Jarak Jauh Berbasis IoT

Abstract

Technological advances are expected to be able to meet growing human needs. One of the things that needs to be considered is door security, and one method to achieve this level of security is by utilizing technology. This research will explore the working principles of door locks that adopt facial identification as an alternative to conventional keys. In this research, conventional keys will be replaced by a facial identification system that operates based on a specified command sequence. The data resulting from the facial identification process will be collected by the ESP32-CAM device, then transferred to the computer to be compared with the available data. The computer uses the Arduino IDE programming language. This research will adopt quantitative methods, which include literature study and direct observation as data collection techniques. The collected data will be analyzed to reach conclusions. In this research, Telegram is also used as a means of remote control which will connect automatically when the system is activated and will be processed by the NodeMCU-ESP8266. The results of this research show that the ESP32-CAM can identify faces with a response time of between 2.69 s – 3.24 s and the distance for face reading is up to a maximum distance of 80 cm. Test results on Telegram show that the response time obtained by the Telegram bot to control the key on the NodeMCU-ESP8266 is between 4.19 s – 4.48 s for each different location.