This research introduces several innovative models designed to enhance PPE compliance among construction workers using convolutional neural networks (CNNs) and transfer learning principles to build upon the advanced YOLO-v5 and YOLO-v8 architectures.
Abstract
Construction safety remains a critical focus for both practitioners and researchers due to the sector’s high rates of accidents and fatalities. Among the most effective measures to mitigate these safety risks is the consistent use of personal protective equipment (PPE), which provides a vital defense against workplace hazards. However, ensuring compliance with PPE usage on dynamic construction sites is a persistent challenge. This research introduces several innovative models designed to enhance PPE compliance among construction workers. The study leverages convolutional neural networks (CNNs) and transfer learning principles to build upon the advanced YOLO-v5 and YOLO-v8 architectures. These models are specifically designed to predict six critical categories related to construction safety: person, vest, and four distinct helmet colors. Additionally, You only look once (YOLO) results were integrated with a safety Power BI dashboard providing stakeholders with a comprehensive overview of the site’s safety status, enabling them to monitor compliance trends and to take prompt corrective actions. Validation of the models was conducted using two datasets: (CHV benchmark dataset and an original dataset collected from Egyptian construction. YOLO-v5 × 6 model was noted for its superior speed in analysis compared to the YOLO-v5l model. However, the YOLO-v8m model outperformed all others in terms of precision and accuracy. Specifically, YOLO-v8m achieved the highest mean average precision (mAP) score of 92.30% and the best F1 score of 0.89. It can be argued that the developed model could significantly contribute to endorsing the ability to prevent accidents and improving safety measures in construction environment.
This paper presents a comprehensive study on deep learning-based personal protective equipment (PPE) detection for real-time construction safety monitoring by applying four recent YOLO-based object detection architectures on a unified dataset with 43,986 images and 14 PPE-related classes. The investigate the detection of not only PPE-compliance classes (Hardhat, Safety Vest, Gloves, Mask, Goggles) but also PPE-violation classes (NO-Hardhat, NO-Safety Vest, NO-Gloves, NO-Mask, NO-Goggles). Experimental results showed that minority PPE-violation classes such as NO-Safety Vest are consistently under-detected, owing to severe class imbalance and high visual resemblance to compliance classes. A class-specific weighted Binary Cross-Entropy (BCE) loss function is then proposed and applied under identical training conditions across the four evaluated architectures. Models are evaluated using precision, recall, mAP@0.50, mAP@0.50:0.95, F1-Score and inference latency. Our experiments show the weighted loss improved NO-Safety Vest performance by up to 35.5% (YOLO26-W: 0.6485 vs. YOLO26 baseline: 0.4787). The overall mAP decreased marginally. Among all considered models, YOLO11m exhibited the best trade-off between detection performance and inference speed. The results demonstrate that the proposed class-specific weighted loss strategy consistently improved minority-class detection performance across multiple YOLO architectures to improve minority PPE violation detection without the need for new data annotation and model redesign.
Ajay Ramasamy J, Jeevaraj S, R. Laxmi et al.· 2026 6th International Confe...· 0 citations
Workplaces in construction and industry suffer from a significant number of workplace accidents because of a lack of safety mechanisms like helmets and high-visibility vests. In order to solve this issue, an automatic detection system for detecting the presence of the worker's helmet & safety vest using a deep learning model is developed. You Only Look Once – Neural Architecture Search (YOLO-NAS) algorithm was chosen to be used in the model because of its fast and high-quality detection process. A dataset with images containing healthy workers wearing helmets & safety vests (with some workers appearing in different safety vests) is collected and used as the training set. The resulting model is then saved for future usage in making predictions with new images. Then, this model is connected to Streamlit, which provides a convenient Web-based user interface through which a user can provide an image as an input to the model. The trained neural network is applied to the input image, and all detected objects that correspond to helmet & safety vest are put in bounding boxes on the input image. Moreover, each of the bounding boxes is labeled with the name of the object and confidence score of the detection. Thus, a clear visualization of the safety equipment worn by the worker is provided. Overall, the developed detection system significantly decreases the human intervention needed for visual verification of the safety helmet or vest. The described system is a user-friendly and cost-effective way of evaluating workplace safety compliance.
S. Vijayakumar, Loganathan Nachimuthu, Balasubramaniam C et al.· 2026 International Conferenc...· 0 citations
This review has organized improvements in SHM along the lines of vibration-based anomaly detection, vision-based defect recognition, and multi-modal data fusion along the lines of vibration-based anomaly detection, vision-based defect recognition, and multi-modal data fusion.
Bellal Mia, Md Umar Faruk, M. Hasan et al.· Scientia. Technology, Scienc...· 0 citations