SavedAdd your journal

Journals / Afyon Kocatepe Üniversitesi Fen ve Mühendislik Bilimleri Dergisi / 2026 / Cilt: 26 Sayı: 3

Performance Analysis of YOLO Algorithms for Safety Helmet Detection in Construction Sites

İnşaat Sahalarında Güvenlik Kaskı Algılama için YOLO Algoritmalarının Performans Analizi

Mahmut Sağlıyan(KAPADOKYA ÜNİVERSİTESİ)0009-0006-9115-3943
Ahmet Özcan(KAPADOKYA ÜNİVERSİTESİ)0000-0002-1098-7078
Pages
615–625
Publication Date
2026-06-03
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

These findings inform the selection and deployment of robust, AI-driven compliance monitoring solutions and contribute to the design of safer, more effective occupational safety systems.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:The construction industry consistently reports a high incidence of work-related injuries.
📊
Key Findings & Data:Although contemporary deep learning methods can automatically detect helmet compliance, their performance may deteriorate under adverse conditions (e.g., low illumination), noncompliant wear (e.g., loosely fastened or tilted helmets), and variability in helmet appearance.
💡
Conclusion & Impact:These findings inform the selection and deployment of robust, AI-driven compliance monitoring solutions and contribute to the design of safer, more effective occupational safety systems.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Derin öğrenme · YOLO · Nesne tespiti

Abstract

The construction industry consistently reports a high incidence of work-related injuries. Consequently, effective safety interventions—particularly the enforcement of protective helmet use—are essential for mitigating risk. Although contemporary deep learning methods can automatically detect helmet compliance, their performance may deteriorate under adverse conditions (e.g., low illumination), noncompliant wear (e.g., loosely fastened or tilted helmets), and variability in helmet appearance. To address these challenges, this study systematically benchmarks YOLO (You Only Look Once) object detection models. We develop a deep learning–based monitoring system for construction sites and evaluate four YOLO variants—YOLOv5s, YOLOv8s, YOLOv11s, and YOLOv12s—on the Hard Hat Workers and BAC_HIEN_CONSTRUCTION_SAFETY_2024 datasets. Models are compared using mean average precision (mAP), precision, recall, and F1-score, together with inference time and real-time detection capability. The results provide empirical guidance on the trade-off between accuracy and speed across YOLO versions, highlighting the models best suited for reliable helmet detection under realistic site conditions. These findings inform the selection and deployment of robust, AI-driven compliance monitoring solutions and contribute to the design of safer, more effective occupational safety systems.

Özet

İnşaat sektörü, iş kazalarının en fazla yaşandığı alanlardan biridir. Bu nedenle, işçilerin sağlığını korumak ve kazaları önlemek amacıyla alınacak güvenlik önlemleri, özellikle güvenlik kasklarının kullanımı, büyük önem taşımaktadır. Derin öğrenme yöntemleri kaskları tespit edebilse de düşük ışık koşulları, yanlış kullanım ve güvenlik kasklarının tasarımlarındaki farklılıklar gibi zorluklar performansı etkileyebilmektedir. Bu zorlukların üstesinden gelebilmek için YOLO (You Only Look Once) gibi hızla gelişen derin öğrenme tabanlı nesne tespit algoritmalarının performanslarını karşılaştırmak gereklidir. Bu çalışma, YOLO algoritmasının çeşitli sürümlerinin büyük veri kümeleri üzerindeki performans analizini sunmaktadır. İnşaat sahalarında güvenlik kasklarının kullanımını izlemek ve iş kazalarını önlemeye yardımcı olmak amacıyla derin öğrenme tabanlı bir nesne tespit sistemi geliştirilmiştir. Bu amaçla, Hard Hat Workers ve BAC_HIEN_CONSTRUCTION_SAFETY_2024 veri setlerinde YOLOv5s, YOLOv8s, YOLOv11s ve YOLOv12s sürümleri değerlendirilmiştir. Deneysel sonuçlar, her bir modeli ortalama doğruluk, kesinlik, duyarlılık ve F1 skoru açısından, ayrıca işleme süresi ve gerçek zamanlı tespit yeteneği bakımından incelemiştir. Bulgular, güvenlik ihlallerinin tespiti için en uygun YOLO sürümünün belirlenmesine dair değerli bilgiler sunmakta ve iş güvenliği sistemlerinde daha etkili yapay zekâ tabanlı nesne tespit çözümlerinin geliştirilmesine katkı sağlamaktadır.

🔖

Cite This Article

Sağlıyan, M., Özcan, A. (2026). Performance Analysis of YOLO Algorithms for Safety Helmet Detection in Construction Sites. Afyon Kocatepe Üniversitesi Fen ve Mühendislik Bilimleri Dergisi, Cilt: 26 Sayı: 3, 615–625. https://doi.org/10.35414/akufemubid.1799653

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (33)

  1. Badgujar, C. M., Poulose, A., & Gan, H. (2024). Agricultural object detection with You Only Look Once (YOLO) Algorithm: A bibliometric and systematic literature review. Computers and Electronics in Agriculture, 223, 109090. https://doi.org/10.1016/j.compag.2024.109090
  2. Baldini, C., Migliorelli, L., Berardini, D., Azam, M. A., Sampieri, C., Ioppi, A.,. & Mattos, L. S. (2025). Improving real-time detection of laryngeal lesions in endoscopic images using a decoupled super-resolution enhanced YOLO. Computer Methods and Programs in Biomedicine, 260, 108539. https://doi.org/10.1016/j.cmpb.2024.108539
  3. Chen, W., Liu, M., Zhou, X., Pan, J., & Tan, H. (2022). Safety Helmet Wearing Detection in Aerial Images Using Improved YOLOv4. Computers, Materials & Continua, 72(2). https://doi.org/10.32604/cmc.2022.026664
  4. Çiftçi, M., Türkdamar, M. U., & Öztürk, C. (2024). Leveraging YOLO Models for Safety Equipment Detection on Construction Sites. Journal of Computing Theories and Applications, 1(4), 492-506. https://doi.org/10.62411/jcta.10453
  5. Dogra, A. K., Sharma, V., & Sohal, H. (2024). A survey of deep learning techniques for detecting and recognizing objects in complex environments. Computer Science Review, 54, 100686. https://doi.org/10.1016/j.cosrev.2024.100686
  6. Gheorghe, C., Duguleana, M., Boboc, R. G., & Postelnicu, C. C. (2024). Analyzing Real-Time Object Detection with YOLO Algorithm in Automotive Applications: A Review. CMES-Computer Modeling in Engineering & Sciences, 141(3). https://doi.org/10.32604/cmes.2024.054735
  7. Han, G., Zhu, M., Zhao, X., & Gao, H. (2021). Method based on the cross-layer attention mechanism and multiscale perception for safety helmet-wearing detection. Computers and Electrical Engineering, 95, 107458. https://doi.org/10.1016/j.compeleceng.2021.107458
  8. Huang, Y., Qian, Y., Wei, H., Lu, Y., Ling, B., & Qin, Y. (2023). A survey of deep learning-based object detection methods in crop counting. Computers and Electronics in Agriculture, 215, 108425. https://doi.org/10.1016/j.compag.2023.108425
  9. Jalonen, T., Al-Sa'd, M., Mellanen, R., Kiranyaz, S., & Gabbouj, M. (2023). Real-time damage detection in fiber lifting ropes using convolutional neural networks. arXiv preprint arXiv:2302.11947. https://doi.org/10.1109/JSEN.2024.3521118
  10. Jamtsho, Y., Riyamongkol, P., & Waranusast, R. (2021). Real-time license plate detection for non-helmeted motorcyclist using YOLO. Ict Express, 7(1), 104-109. https://doi.org/10.1016/j.icte.2020.07.008
  11. Kwak, N., & Kim, D. (2023). Detection of worker’s safety helmet and mask and identification of worker using deeplearning. Computers, Materials and Continua, 75(1), 1671-1686. https://doi.org/10.32604/cmc.2023.035762
  12. Lee, J. Y., Choi, W. S., & Choi, S. H. (2023). Verification and performance comparison of CNN-based algorithms for two-step helmet-wearing detection. Expert Systems with Applications, 225, 120096. https://doi.org/10.1016/j.eswa.2023.120096
  13. Li, H., Wu, D., Zhang, W., & Xiao, C. (2024). YOLO-PL: Helmet wearing detection algorithm based on improved YOLOv4. Digital Signal Processing, 144, 104283. https://doi.org/10.1016/j.dsp.2023.104283
  14. Lin, H., Jiang, F., Jiang, Y., Luo, H., Yao, J., & Liu, J. (2023). A Model for Helmet-Wearing Detection of Non-Motor Drivers Based on YOLOv5s. Computers, Materials & Continua, 75(3). https://doi.org/10.32604/cmc.2023.036893
  15. Liu, Y., Zhang, J., Shi, L., Huang, M., Lin, L., Zhu, L.,. & Zhang, C. (2024). Detection method of the seat belt for workers at height based on UAV image and YOLO algorithm. Array, 22, 100340. https://doi.org/10.1016/j.array.2024.100340
  16. Manorat, P., Tuarob, S., & Pongpaichet, S. (2025). Artificial intelligence in computer programming education: A systematic literature review. Computers and Education: Artificial Intelligence, 100403. https://doi.org/10.1016/j.caeai.2025.100403
  17. Mostafa, S. A., Ravi, S., Zebari, D. A., Zebari, N. A., Mohammed, M. A., Nedoma, J.,. & Ding, W. (2024). A YOLO-based deep learning model for Real-Time face mask detection via drone surveillance in public spaces. Information Sciences, 676, 120865. https://doi.org/10.1016/j.ins.2024.120865
  18. Nazli, N. A. N. M., Sabri, N., Aminuddin, R., Ibrahim, S., Yusof, S., & Nasir, S. D. N. M. (2024). A real-time system for detecting personal protective equipment compliance using deep learning model YOLOv5. Procedia Computer Science, 245, 647-656. https://doi.org/10.1016/j.procs.2024.10.291
  19. Raj, G. D., & Prabadevi, B. (2025). ENHANCING SURFACE DETECTION: A COMPREHENSIVE ANALYSIS OF VARIOUS YOLO MODELS. Heliyon, 11(3), e42433. https://doi.org/10.1016/j.heliyon.2025.e42433
  20. Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 779-788). https://doi.org/10.48550/arXiv.1506.02640
  21. Said, Y., Alassaf, Y., Ghodhbani, R., Alsariera, Y. A., Saidani, T., Rhaiem, O. B.,. & Hleili, M. (2024). AI-Based Helmet Violation Detection for Traffic Management System. CMES-Computer Modeling in Engineering & Sciences, 141(1), 733-749. https://doi.org/10.32604/cmes.2024.052369
  22. Son, H., & Kim, C. (2021). Integrated worker detection and tracking for the safe operation of construction machinery. Automation in Construction, 126, 103670. https://doi.org/10.1016/j.autcon.2021.103670
  23. Vourkos, E. G., Agathokleous, R. A., Panayides, A. S., Kalogirou, S. A., & Christoforou, E. G. (2025). Improving photovoltaics hotspot and dirt accumulation detection with computer vision: A systematic YOLO model comparison based on field aerial images. Sustainable Energy Technologies and Assessments, 78, 104355. https://doi.org/10.1016/j.seta.2025.104355
  24. Vukicevic, A. M., Djapan, M., Isailovic, V., Milasinovic, D., Savkovic, M., & Milosevic, P. (2022). Generic compliance of industrial PPE by using deep learning techniques. Safety science, 148, 105646. https://doi.org/10.1016/j.ssci.2021.105646
  25. Zhang, Y., Huang, S., Qin, J., Li, X., Zhang, Z., Fan, Q., & Tan, Q. (2025). Detection of helmet use among construction workers via helmet-head region matching and state tracking. Automation in Construction, 171, 105987. https://doi.org/10.1016/j.autcon.2025.105987
  26. Zhou, Q., Qin, J., Xiang, X., & Tan, Y. (2021). Algorithm of Helmet Wearing Detection Based on AT-YOLO Deep Mode. Computers, Materials & Continua, 69(1), 159-174. https://doi.org/10.32604/cmc.2021.017480
  27. Bac Hien Construction Safety 2024 Veri Seti, https://universe.roboflow.com/constructionsatefy/bac_hien_construction_safety_2024, (10.02.2026)
  28. Hard Hat Workers Veri Seti, https://universe.roboflow.com/joseph-nelson/hard-hat-workers, (10.02.2026)
  29. Baldini, C., Migliorelli, L., Berardini, D., Azam, M. A., Sampieri, C., Ioppi, A.,. & Mattos, L. S. (2025). Improving real-time detection of laryngeal lesions in endoscopic images using a decoupled super-resolution enhanced YOLO. Computer Methods and Programs in Biomedicine, 260, 108539. https://doi.org/10.1016/j.cmpb.2024.108539
  30. Liu, Y., Zhang, J., Shi, L., Huang, M., Lin, L., Zhu, L.,. & Zhang, C. (2024). Detection method of the seat belt for workers at height based on UAV image and YOLO algorithm. Array, 22, 100340. https://doi.org/10.1016/j.array.2024.100340
  31. Mostafa, S. A., Ravi, S., Zebari, D. A., Zebari, N. A., Mohammed, M. A., Nedoma, J.,. & Ding, W. (2024). A YOLO-based deep learning model for Real-Time face mask detection via drone surveillance in public spaces. Information Sciences, 676, 120865. https://doi.org/10.1016/j.ins.2024.120865
  32. Said, Y., Alassaf, Y., Ghodhbani, R., Alsariera, Y. A., Saidani, T., Rhaiem, O. B.,. & Hleili, M. (2024). AI-Based Helmet Violation Detection for Traffic Management System. CMES-Computer Modeling in Engineering & Sciences, 141(1), 733-749. https://doi.org/10.32604/cmes.2024.052369
  33. Bac Hien Construction Safety 2024 Veri Seti, https://universe.roboflow.com/constructionsatefy/bac_hien_construction_safety_2024, (10.02.2026)