Sistem Deteksi Alat Pelindung Diri Berbasis YOLOv8n pada Jetson Nano untuk Industri Migas
DOI:
https://doi.org/10.31328/jasee.v7i1.04Keywords:
PPE detection, NVIDIA Jetson Nano, embedded system, computer vision, oil and gas safetyAbstract
Occupational safety in the oil and gas industry requires strict compliance with Personal Protective Equipment (PPE); however, manual inspection is prone to human error and lacks effectiveness. This research designs and implements a PPE Smart Station based on the YOLOv8n algorithm on an NVIDIA Jetson Nano A02 embedded system to automatically and in real-time detect the completeness of three main PPE types—helmet, coverall, and gloves—within the industrial environment of PT Husky-CNOOC Madura Limited. The system employs a state machine with PASS/DENIED output, a real-time 2D visualization of missing PPE on the worker’s anatomical regions, and Indonesian-language text alerts with automatic audio warnings. Results are reported at two levels: (i) model evaluation on a limited test batch yielded 99.0% precision and 100% recall under near-ideal capture conditions; and (ii) field system evaluation on 30 samples at a single installation point, assessed with a confusion matrix at the PASS/DENIED decision level, yielded 93.3% accuracy, 96.0% precision, 96.0% recall, 96.0% F1-measure, and a 6.7% error rate. The results indicate the system’s initial feasibility as an AI-based occupational-safety monitoring solution under the evaluated conditions, while larger-scale, multi-site validation remains necessary.
References
[1] K. S. Alfidyani, D. Lestantyo, and I. Wahyuni, “Hubungan pelatihan K3, penggunaan APD, pemasangan safety sign, dan penerapan SOP dengan terjadinya risiko kecelakaan kerja,” Jurnal Kesehatan Masyarakat, vol. 8, no. 4, pp. 478–483, 2020.
[2] Republik Indonesia, Undang-Undang Nomor 1 Tahun 1970 tentang Keselamatan Kerja. Jakarta: Sekretariat Negara, 1970.
[3] Republik Indonesia, Peraturan Pemerintah Nomor 11 Tahun 1979 tentang Keselamatan Kerja pada Pengolahan Minyak dan Gas Bumi. Jakarta: Sekretariat Negara, 1979.
[4] G. Gallo, F. Di Rienzo, F. Garzelli, P. Ducange, and C. Vallati, “A smart system for personal protective equipment detection in industrial environments based on deep learning at the edge,” IEEE Access, vol. 10, pp. 110862–110878, 2022, doi: 10.1109/ACCESS.2022.3215148.
[5] A. Barlybayev, A. Sharipbay, G. Ulyukova, T. Sabyrov, and B. Kuzenbayev, “Personal protective equipment detection using YOLOv8 architecture on object detection benchmark datasets: a comparative study,” Cogent Engineering, vol. 11, no. 1, art. 2333209, 2024, doi: 10.1080/23311916.2024.2333209.
[6] A. Husain, M. Tamboli, S. Surwade, and M. H. Qureshi, “Automated PPE detection using YOLOv8 for real-time workplace safety monitoring,” International Journal of Innovative Science and Research Technology, vol. 9, no. 4, pp. 1–7, 2024.
[7] N. D. Nath, A. H. Behzadan, and S. G. Paal, “Deep learning for site safety: real-time detection of personal protective equipment,” Automation in Construction, vol. 112, art. 103085, 2020, doi: 10.1016/j.autcon.2020.103085.
[8] Z. Wang, Y. Wu, L. Yang, A. Thirunavukarasu, C. Evison, and Y. Zhao, “Fast personal protective equipment detection for real construction sites using deep learning approaches,” Sensors, vol. 21, no. 10, art. 3478, 2021, doi: 10.3390/s21103478.
[9] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You only look once: unified, real-time object detection,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2016, pp. 779–788, doi: 10.1109/CVPR.2016.91.
[10] A. Hayat and F. Morgado-Dias, “Deep learning-based automatic safety helmet detection system for construction safety,” Applied Sciences, vol. 12, no. 16, art. 8268, 2022, doi: 10.3390/app12168268.
[11] M. Ferdous and S. M. M. Ahsan, “PPE detector: a YOLO-based architecture to detect personal protective equipment (PPE) for construction sites,” PeerJ Computer Science, vol. 8, art. e999, 2022, doi: 10.7717/peerj-cs.999.
[12] NVIDIA Corporation, NVIDIA Jetson Nano Developer Kit Technical Reference Manual. Santa Clara, CA: NVIDIA Corporation, 2021. [Online]. Available: https://developer.nvidia.com/jetson-nano
[13] G. Jocher, A. Chaurasia, and J. Qiu, “Ultralytics YOLOv8,” 2023. [Online]. Available: https://github.com/ultralytics/ultralytics
[14] J. Terven, D.-M. Córdova-Esparza, and J.-A. Romero-González, “A comprehensive review of YOLO architectures in computer vision: from YOLOv1 to YOLOv8 and YOLO-NAS,” Machine Learning and Knowledge Extraction, vol. 5, no. 4, pp. 1680–1716, 2023, doi: 10.3390/make5040083.
[15] Republik Indonesia, Peraturan Pemerintah Nomor 50 Tahun 2012 tentang Penerapan Sistem Manajemen Keselamatan dan Kesehatan Kerja (SMK3). Jakarta: Sekretariat Negara, 2012.
[16] Kementerian Tenaga Kerja dan Transmigrasi Republik Indonesia, Peraturan Menteri Tenaga Kerja dan Transmigrasi Nomor PER.08/MEN/VII/2010 tentang Alat Pelindung Diri. Jakarta, 2010.
[17] Raspberry Pi Ltd., High Quality Camera (IMX477) Datasheet. Cambridge, UK: Raspberry Pi Ltd., 2020. [Online]. Available: https://www.raspberrypi.com
[18] V. V. Kukartsev, R. A. Ageev, and A. S. Borodulin, “Deep learning for object detection in images: development and evaluation of the YOLOv8 model using Ultralytics and Roboflow libraries,” in Lecture Notes in Networks and Systems, 2024, pp. 476–484.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 JASEE Journal of Application and Science on Electrical Engineering
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
