EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Design of Yolo-Based Electrical Panel Box for Solenoid Drives and Telegram Notification

Akbar Maulana, Ayusta Lukita Wardani, Fithrotul Irda Amaliah

Journal of Intelligent Computing and Engineering · 2025

DOI: 10.67561/jice.v1i2.1 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Intelligent Computing and Engineering, 2025. Verfasst von Akbar Maulana und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Face recognition detection technology on electrical panel boxes for automatic dual security needs that can be easily monitored. This article presents the design and performance results of an electrical panel box based on the YOLO version 5 method for performing face recognition detection, which then controls the solenoid as a door lock and uses Telegram to send messages in the form of text and images of the device's operating conditions. The YOLO version 5 method was chosen for its efficiency, detection speed, and advanced dataset configuration. The dataset was created using Roboflow for collecting face data, labeling or classifying face data, and configuring face data variations. Node-red and Telegram were chosen as the monitoring system because they are user-friendly and open source. The device program system is run by Raspberry pi 4B, which ensures that the YOLO version 5 method is successful, then receives data in the form of detection results which are then used to control the solenoid and send data to both monitoring platforms, namely Node-red and Telegram, so that users know the working condition of the device. An experimental approach was used to evaluate system performance in terms of successful face recognition detection using the YOLO version 5 method with several conditions and device response performance. The results show that the average detection success rate is 91.67% and the average confidence value is 0.84, with an optimal detection distance of 1 to 2 meters. The device performance system runs properly and the device response performance shows a result of 5.17 seconds. This study shows that the YOLO version 5 method used was successful for face recognition detection with the help of a dataset processed by Roboflow. The performance of the built tool was successful and had a good response. With these results, the built tool can be used as a dual security system option on electrical panel doors and become a development in face recognition detection.

Zitieren

APA

Maulana, A., Wardani, A. L., & Amaliah, F. I. (2025). Design of Yolo-Based Electrical Panel Box for Solenoid Drives and Telegram Notification. Journal of Intelligent Computing and Engineering. https://doi.org/10.67561/jice.v1i2.1

Vancouver

Maulana A, Wardani AL, Amaliah FI. Design of Yolo-Based Electrical Panel Box for Solenoid Drives and Telegram Notification. Journal of Intelligent Computing and Engineering. 2025. doi:10.67561/jice.v1i2.1

BibTeX
@article{Maulana2025,
  author = {Maulana, Akbar and Wardani, Ayusta Lukita and Amaliah, Fithrotul Irda},
  title = {Design of Yolo-Based Electrical Panel Box for Solenoid Drives and Telegram Notification},
  journal = {Journal of Intelligent Computing and Engineering},
  year = {2025},
  doi = {10.67561/jice.v1i2.1}
}
Kostenlos starten →