Machine learning using synthetic images for detecting dust emissions on construction sites

Abstract

PurposeAutomated dust monitoring in workplaces helps provide timely alerts to over-exposed workers and effective mitigation measures for proactive dust control. However, the cluttered nature of construction sites poses a practical challenge to obtain enough high-quality images in the real world. The study aims to establish a framework that overcomes the challenges of lacking sufficient imagery data (“data-hungry problem”) for training computer vision algorithms to monitor construction dust.Design/methodology/approachThis study develops a synthetic image generation method that incorporates virtual environments of construction dust for producing training samples. Three state-of-the-art object detection algorithms, including Faster-RCNN, you only look once (YOLO) and single shot detection (SSD), are trained using solely synthetic images. Finally, this research provides a comparative analysis of object detection algorithms for real-world dust monitoring regarding the accuracy and computational efficiency.FindingsThis study creates a construction dust emission (CDE) dataset consisting of 3,860 synthetic dust images as the training dataset and 1,015 real-world images as the testing dataset. The YOLO-v3 model achieves the best performance with a 0.93 F1 score and 31.44 fps among all three object detection models. The experimental results indicate that training dust detection algorithms with only synthetic images can achieve acceptable performance on real-world images.Originality/valueThis study provides insights into two questions: (1) how synthetic images could help train dust detection models to overcome data-hungry problems and (2) how well state-of-the-art deep learning algorithms can detect nonrigid construction dust.

Dust detection and cleanliness assessment based on S-YOLOv5s for NPP reactor containment wall-climbing cleaning robot

“…In Ref. [ 13 ], the results of Faster Region-Convolutional Neural Network (Faster RCNN), You Only Look Once (YOLO) and Single Shot MultiBox Detector (SSD) were compared for recognizing images of fugitive dust, and YOLOv3 gave the best results. In Ref.…”

Research on a quality detection system of dust suppression spraying for railway coal transportation using laser backscattering technology

“…Visual dust suppression methods and air particle detection methods are two prevailed methods, mainly applied to the dust suppression quality detection of large outdoor mines or workshops [9,10]. Visual dust suppressor can be mixed with some inorganic powder pigments and sprayed to form a colored shell.…”

Deep learning for detecting distresses in buildings and pavements: a critical gap analysis

“…Recently, deep learning has greatly endorsed computer vision development, which offers a feasible method for automated crack detection (Ogunseiju et al , 2021). Deep learning allows computers to learn from experience by using artificial neural networks (ANNs) and other machine learning algorithms (Xiong and Tang, 2021). This technique is “deep” as it contains many layers that are used for feature extraction, transformation and pattern analysis using supervised or unsupervised learning (Ongsulee, 2018).…”