Penerapan Smart Computer Vision untuk Deteksi Alat Pelindung Diri di Tempat Pembuangan Akhir Sampah
DOI:
https://doi.org/10.31004/r8jkzd21Keywords:
smart computer vision, alat pelindung diri, YOLOv8, Deep Learning, Keselamatan KerjaAbstract
Tempat Pembuangan Akhir (TPA) merupakan lingkungan kerja dengan risiko tinggi terhadap kecelakaan dan penyakit akibat kerja. Rendahnya kepatuhan pekerja dalam menggunakan Alat Pelindung Diri (APD) menjadi salah satu faktor utama yang meningkatkan risiko kecelakaan. Pengawasan manual yang dilakukan selama ini terbukti kurang efektif karena keterbatasan tenaga pengawas dan luasnya area kerja. Pengabdian ini mengusulkan penerapan sistem APDWatch berbasis Smart Computer Vision yang mengintegrasikan model deteksi objek YOLOv8m dan model klasifikasi multi-output VGG16 untuk mendeteksi penggunaan APD secara otomatis. Sistem ini mampu memberikan notifikasi real-time kepada pengawas ketika ditemukan pelanggaran penggunaan APD. Hasil evaluasi menunjukkan nilai mAP@50 sebesar 0,9514 untuk YOLOv8m dan rata-rata F1-score 0,8927 pada atribut helm, rompi, dan sepatu menggunakan VGG16. Implementasi sistem pada perangkat IoT berbasis Raspberry Pi dengan kamera menunjukkan kemampuan berjalan secara real-time dengan latensi kurang dari 2 detik. Program pengabdian ini dilaksanakan di CV Berkah Jaya Mandiri Semarang dengan luaran berupa aplikasi web monitoring, artikel ilmiah, dan HKI hak cipta aplikasi. Penerapan sistem ini diharapkan dapat meningkatkan kesadaran pekerja tentang pentingnya penggunaan APD serta mengurangi angka kecelakaan kerja di sektor pengelolaan sampah.
Downloads
References
Ahmad, A., Herdianzah, Y., Safutra, N. I., & Malik, R. (2024). Implementasi keselamatan dan kesehatan kerja pada proses pemilahan sampah bagi pemulung di tempat pembuangan akhir Tamangapa Kota Makassar. Idea Pengabdian Masyarakat, 4(1), 45-56. https://doi.org/10.53690/ipm.v4i01.194
Ajwad, A., Fauzi, A., & Wijaya, C. (2022). Evaluasi occupational safety and health management system policies di Dinas Lingkungan Hidup Kota Sukabumi. SIBATIK Jurnal, 4(2), 118-132. https://doi.org/10.54443/sibatik.v4i2.2598
Fang, W., Wang, L., & Ren, P. (2024). Automated detection of helmet wearing with YOLOv8 and real-time monitoring for factory safety. International Journal of Informatics and Knowledge Management. https://doi.org/10.52953/ijikm.0000.0000
Gao, R., Liu, Y., & Chen, Z. (2023). Real-time detection of PPE compliance using YOLO-based deep learning models. IEEE Transactions on Industrial Informatics. https://doi.org/10.1109/TII.2023.3287564
Hornsby, D. (2025). AI: Helping organizations with health and safety. Inside Supply Management Magazine. https://www.ismworld.org/supply-management-news-and-reports/news-publications/inside-supply-management-magazine/blog/2025/2025-09/ai-helping-organizations-with-health-and-safety/
Huang, T., & Sun, J. (2022). Applying AI-powered image processing for personal protective equipment monitoring in hazardous environments. IEEE International Conference on Image Processing. https://doi.org/10.1109/ICIP.2022.9801938
Kim, D., & Park, S. (2022). Computer vision-based personal protective equipment monitoring for smart workplaces. IEEE/CVF Conference on Computer Vision and Pattern Recognition. https://doi.org/10.1109/CVPR.2022.1359
Li, C., & Zhang, W. (2023). AI-powered safety compliance monitoring in construction and waste management sites. Automation in Construction. https://doi.org/10.1016/j.autcon.2023.104238
Martinez, R., & Gonzalez, H. (2021). Enhancing worker safety through AI-based PPE monitoring systems. IEEE Access. https://doi.org/10.1109/ACCESS.2021.3124782
Mendoza, J., & Castillo, P. (2021). AI-driven PPE recognition system for safety compliance in industrial and waste management sites. Journal of Hazardous Materials. https://doi.org/10.1016/j.jhazmat.2021.125491
Mohy, M. A., et al. (2025). Deep learning enabled computer vision model for construction safety. ITcon Journal. https://doi.org/10.36680/j.itcon.2025.057
Phasuk, K., et al. (2025). ESPCN-YOLO: A high-accuracy framework for personal protective equipment detection under low-light and small objects. Buildings, 15(10), 1609. https://doi.org/10.3390/buildings15101609
Protex AI. (2024). A complete guide to PPE and PPE detection. https://www.protex.ai/guides/a-complete-guide-to-ppe-and-ppe-detection
Sharma, N., & Patel, K. (2022). An intelligent deep learning approach for automated PPE detection in industrial and hazardous environments. Engineering Applications of Artificial Intelligence. https://doi.org/10.1016/j.engappai.2022.105837
Sui, H., et al. (2018). Occupational safety and health management system implementation in developing countries. Safety Science, 109, 224-233.
Wang, T., et al. (2025). An improved YOLOv8 safety helmet wearing detection network. PMC. https://pmc.ncbi.nlm.nih.gov/articles/PMC11289450/
Zhang, Y., Li, X., & Wang, L. (2023). Deep learning-based object detection for occupational safety monitoring in industrial workplaces. Expert Systems with Applications. https://doi.org/10.1016/j.eswa.2023.119488
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Prayitno, Wiktasari, Suko Tyas Pernanda, Aminuddin Rizal, Sunardi
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
