Обнаружение средств индивидуальной защиты с помощью моделей глубокого обучения на основе многоуровневого объединения признаков и пространственного внимания
Просмотры: 267 / Загрузок PDF: 154
DOI:
https://doi.org/10.32523/bulmathenu.2025/2.2Ключевые слова:
глубокое обучение, средства индивидуальной защиты, промышленная безопасность, интеллектуальная система, компьютерное зрениеАннотация
В предлагаемой исследовательской работе рассматривается эффективность моделей глубокого обучения для автоматического распознавания средств индивидуальной защиты в сложных условиях, встречающихся на различных промышленных рабочих местах. Поскольку средства индивидуальной защиты предназначены для защиты работников от различных травм и опасных воздействий, мониторинг их использования с применением современных моделей глубокого обучения является важным решением. Учитывая разнообразие этих опасностей, современный, но неисследованный набор данных с многоклассовой структурой, содержащий различные сцены и полностью аннотированный, был обучен на сети глубокого обучения Yolov11 для оценки их надежности и потенциала в реальных приложениях. Результаты показывают, что архитектурно улучшенные модели могут эффективно работать даже со сложными и несбалансированными данными. Кроме того, в ходе исследования был проведен сравнительный анализ с моделями YOLOv8 и YOLOv10, а также был выполнен анализ для каждого варианта архитектуры. Данное исследование вносит вклад в развитие интеллектуальных систем в области промышленной безопасности и предлагает практическое решение для контроля соблюдения работниками правил безопасности на рабочем месте, тем самым спасая жизни.
Библиографические ссылки
Қазақстандағы өндірістік жарақаттану деңгейі 4{,}5%-ға азайған [Электронды ресурс]. - URL: https://www.gov.kz/memleket/entities/enbek/press/news/details/924686?lang=ru (Қаралған күні: 09.06.2025).
Статистика бойынша медициналық көрсеткіштер – 2022 жыл [Электронды ресурс]. - URL: https://stat.gov.kz/ru/industries/social-statistics/stat-medicine/publications/6411/ (Қаралған күні: 09.06.2025).
Статистика бойынша медициналық көрсеткіштер – 2023 жыл [Электронды ресурс]. - URL: https://stat.gov.kz/ru/industries/social-statistics/stat-medicine/publications/158509/ (Қаралған күні: 09.06.2025).
жылдың басынан бері Қазақстанда өндірісте 40 адам қайтыс болған [Электронды ресурс]. - URL: https://24.kz/ru/news/social/705980-40-pogibshikh-na-proizvodstve-v-kazakhstane-s-nachala-goda (Қаралған күні: 09.06.2025).
Barlybayev A., Amangeldy N., Kurmetbek B., Krak I., Razakhova B., Tursynova N., Turebayeva R. Personal protective equipment detection using YOLOv8 architecture on object detection benchmark datasets: a comparative study, Cogent Engineering. – 2024. – Vol. 11, No. 1. – P. 2333209. – https://doi.org/10.1080/23311916.2024.2333209. DOI: https://doi.org/10.1080/23311916.2024.2333209
López L., Suárez-Ramírez J., Alemán-Flores M., Monzón N. Automated PPE compliance monitoring in industrial environments using deep learning-based detection and pose estimation, SSRN [Preprint]. – 2024 Nov 28. – https://doi.org/10.2139/ssrn.5037705. DOI: https://doi.org/10.2139/ssrn.5037705
Ferdous M., Ahsan S.M. PPE detector: a YOLO-based architecture to detect personal protective equipment (PPE) for construction sites, PeerJ Comput. Sci. – 2022. – Vol. 8. – Article e999. – https://doi.org/10.7717/peerj-cs.999. DOI: https://doi.org/10.7717/peerj-cs.999
Kim D., Xiong S. Enhancing worker safety: real-time automated detection of personal protective equipment to prevent falls from heights at construction sites using improved YOLOv8 and edge devices, J. Constr. Eng. Manag. – 2025. – Vol. 151, No. 1. – P. 04024187. – https://doi.org/10.1061/(ASCE)CO.1943-7862.0002418. DOI: https://doi.org/10.1061/JCEMD4.COENG-14985
Amangeldy N., Barlybayev A., Gazizova N., Kurmetbek B. Evaluation of YOLOv8 and YOLOv10 models in PPE recognition tasks, J. Electr. Syst. – 2024. – Vol. 20, No. 10. – P. 8141–8148. – Available from: https://journal.esrgroups.org/jes/article/view/7051.
Shi C., Zhu D., Shen J., Zheng Y., Zhou C. GBSG-YOLOv8n: a model for enhanced personal protective equipment detection in industrial environments, Electronics. – 2023. – Vol. 12, No. 22. – P. 4628. – https://doi.org/10.3390/electronics12224628. DOI: https://doi.org/10.3390/electronics12224628
Majumder A., Chatterjee S. YoloGA: an evolutionary computation based YOLO algorithm to detect personal protective equipment, J. Intell. Fuzzy Syst. – 2025 May 15. – [Epub ahead of print]. DOI: https://doi.org/10.1177/18758967251338695
Bento J., Paixão T., Alvarez A.B. Performance evaluation of YOLOv8, YOLOv9, YOLOv10, and YOLOv11 for stamp detection in scanned documents, Appl. Sci. – 2025. – Vol. 15, No. 6. – P. 3154. DOI: https://doi.org/10.3390/app15063154. DOI: https://doi.org/10.3390/app15063154
Barro-Torres S., Fernández-Caramés T.M., Pérez-Iglesias H.J., Escudero C.J. Real-time personal protective equipment monitoring system, Comput. Commun. – 2012. – Vol. 36, No. 1. – P. 42–50. DOI: https://doi.org/10.1016/j.comcom.2012.01.005. DOI: https://doi.org/10.1016/j.comcom.2012.01.005
Nikulin A., Ikonnikov D., Dolzhikov I. Smart personal protective equipment in the coal mining industry, Int. J. Civ. Eng. Technol. – 2019. – Vol. 10, No. 4. – P. 852–863.
Park S.H. Personal protective equipment for healthcare workers during the COVID-19 pandemic, Infect. Chemother. – 2020. – Vol. 52, No. 2. – P. 165. DOI: https://doi.org/10.3947/ic.2020.52.2.165. DOI: https://doi.org/10.3947/ic.2020.52.2.165
Wong T.K.M., Man S.S., Chan A.H.S. Critical factors for the use or non-use of personal protective equipment amongst construction workers, Saf. Sci. – 2020. – Vol. 126. – Article 104663. DOI: https://doi.org/10.1016/j.ssci.2020.104663. DOI: https://doi.org/10.1016/j.ssci.2020.104663
Zhang H., Mu C., Ma X., Guo X., Hu C. MEAG-YOLO: A novel approach for the accurate detection of personal protective equipment in substations, Appl. Sci. – 2024. – Vol. 14, No. 11. – P. 4766. DOI: https://doi.org/10.3390/app14114766. DOI: https://doi.org/10.3390/app14114766
Ragab M.G., Abdulkader S.J., Muneer A., et al. A comprehensive systematic review of YOLO for medical object detection (2018 to 2023), IEEE Access. – 2024. DOI: https://doi.org/10.1109/access.2024.3386826. DOI: https://doi.org/10.1109/ACCESS.2024.3386826
Chen J., Zhu J., Li Z., Yang X. YOLOv7-WFD: A novel convolutional neural network model for helmet detection in high-risk workplaces, IEEE Access. – 2023. – Vol. 11. DOI: https://doi.org/10.1109/ACCESS.2023.3323588. DOI: https://doi.org/10.1109/ACCESS.2023.3323588
Elesawy A., Abdelkader E.M., Osman H. A detailed comparative analysis of You Only Look Once-based architectures for the detection of personal protective equipment on construction sites, Eng. – 2024. – Vol. 5, No. 1. DOI: https://doi.org/10.3390/eng5010019. DOI: https://doi.org/10.3390/eng5010019
Kwon Y.H., Park S., Minh Luan T., Oh S., Heo J. Training data sensitivity analysis of deep neural network for differentiating construction laborers with/without safety helmets, Computing in Civil Engineering 2023, ASCE. – 2024. DOI: https://doi.org/10.1061/9780784485248.062. DOI: https://doi.org/10.1061/9780784485248.062
Otgonbold M.E., Gochoo M., Alnajjar F., et al. SHEL5K: An extended dataset and benchmarking for safety helmet detection, Sensors. – 2022. – Vol. 22, No. 6. DOI: https://doi.org/10.3390/s22062315. DOI: https://doi.org/10.3390/s22062315
Samma H., Al-Azani S., Luqman H., Alfarraj M. Contrastive-based YOLOv7 for personal protective equipment detection, Neural Comput. Appl. – 2024. – Vol. 36, No. 5. DOI: https://doi.org/10.1007/s00521-023-09212-6. DOI: https://doi.org/10.1007/s00521-023-09212-6
Wang Z., Wu Y., Yang L., et al. Fast personal protective equipment detection for real construction sites using deep learning approaches, Sensors. – 2021. – Vol. 21, No. 10. – Article 3478. DOI: https://doi.org/10.3390/s21103478. DOI: https://doi.org/10.3390/s21103478
Nguyen N.T., Tran Q., Dao C.H., et al. Automatic detection of personal protective equipment in construction sites using metaheuristic optimized YOLOv5, Arab J. Sci. Eng. – 2024. – Vol. 49, No. 10. DOI: https://doi.org/10.1007/s13369-023-08700-0. DOI: https://doi.org/10.1007/s13369-023-08700-0
Zhang Q., Pei Z., Guo R., et al. An automated detection approach of protective equipment donning for medical staff under COVID-19 using deep learning, Comput. Model. Eng. Sci. – 2022. – Vol. 132, No. 3. DOI: https://doi.org/10.32604/cmes.2022.019085. DOI: https://doi.org/10.32604/cmes.2022.019085
Alateeq M.M., Fathimathul F.R., Ali M.A.S. Construction site hazards identification using deep learning and computer vision, Sustainability. – 2023. – Vol. 15, No. 3. DOI: https://doi.org/10.3390/su15032358. DOI: https://doi.org/10.3390/su15032358
Wang Z., Cai Z., Wu Y. An improved YOLOX approach for low-light and small object detection: PPE on tunnel construction sites, J. Comput. Des. Eng. – 2023. – Vol. 10, No. 3. DOI: https://doi.org/10.1093/jcde/qwad042. DOI: https://doi.org/10.1093/jcde/qwad042
Ahmad H.M., Rahimi A. SH17: A dataset for human safety and personal protective equipment detection in manufacturing industry, J. Saf. Sci. Resil. – 2024. – Vol. 6, No. 2. – P. 175–185. DOI: https://doi.org/10.1016/j.jnlssr.2024.09.002. DOI: https://doi.org/10.1016/j.jnlssr.2024.09.002
Nath N.D., Behzadan A.H., Paal S.G. Deep learning for site safety: Real-time detection of personal protective equipment, Automation in Construction. – 2020. – Vol. 112. – Article 103085. DOI: https://doi.org/10.1016/j.autcon.2020.103085. DOI: https://doi.org/10.1016/j.autcon.2020.103085
SH17 dataset for PPE detection [Electronic resource] / Mughees Ahmad. – URL: https://www.kaggle.com/datasets/mugheesahmad/sh17-dataset-for-ppe-detection, free. – Accessed: 26.06.2025.






