Harnessing Synthetic Image Datasets for Enhanced Scene Understanding in Construction Work Zones
Publication: International Conference on Transportation and Development 2024
ABSTRACT
This study pioneers the use of synthetic image datasets, generated via the Unity game engine, to train deep learning models for construction work zone scene understanding. This innovative approach simplifies data acquisition and ensures a rich, diverse dataset that includes various construction scenarios, both hazardous and typical. We created 12,360 images with accurate bounding box annotations, ensuring high-quality, consistent data crucial for model training and validation, and effectively addressing the ambiguities often found in real-world datasets. An object detection benchmark was established using this dataset alongside eight state-of-the-art object detectors. This benchmark thoroughly evaluates the performance of these detectors on a wide range of construction site images, enabling comparisons and analyses of different models. It highlights their respective strengths and weaknesses in construction site applications. Notably, YOLOv8-L demonstrated exceptional performance, achieving a mean average precision (mAP) of 70.7% on the validation set and 69.8% on the testing set. These results underscore the efficacy of synthetic datasets in training models for complex scene understanding. This integration of synthetic and real-world imagery has the potential to revolutionize scene comprehension in construction zones, significantly enhancing safety and efficiency in the construction industry.
Get full access to this chapter
View all available purchase options and get full access to this chapter.
REFERENCES
Hu, P. 2022. “Application of Machine Learning Algorithms in the Development of Civil Engineering Software.” 2022 IEEE 2nd International Conference on Electronic Technology, Communication and Information (ICETCI), 975–978. Changchun, China: IEEE.
Li, C., et al. 2022. “YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications.”.
Lv, W., Y. Zhao, S. Xu, J. Wei, G. Wang, C. Cui, Y. Du, Q. Dang, and Y. Liu. 2023. “DETRs Beat YOLOs on Real-time Object Detection.”.
Mahmoud, A. S., M. H. Ahmad, and Y. M. Yatim. 2020. “Overview of Safety Performance in the Construction Industry.” Proceedings of International Structural Engineering and Construction, (Y. Vacanas, C. Danezis, A. Singh, and S. Yazdani, eds.), 7 (1). https://doi.org/10.14455/ISEC.res.2020.7(1).CSA-07.
Park, J., and Y. K. Cho. 2017. “Development and Evaluation of a Probabilistic Local Search Algorithm for Complex Dynamic Indoor Construction Sites.” J. Comput. Civ. Eng., 31 (4): 04017015. https://doi.org/10.1061/(ASCE)CP.1943-5487.0000658.
Song, Z., et al. 2023. “Synthetic Datasets for Autonomous Driving: A Survey.”. https://doi.org/10.48550/ARXIV.2304.12205.
Stone, J. R., D. I. Blockley, and B. W. Pilsworth. 1991. “Managing risk in civil engineering by machine learning from failures.” [1990] Proceedings. First International Symposium on Uncertainty Modeling and Analysis, 255–259. College Park, MD, USA: IEEE Comput. Soc. Press.
Timofeeva, S. S., D. V. Ulrikh, and N. V. Tsvetkun. 2017. “Professional Risks in Construction Industry.” Procedia Engineering, 206: 911–917. https://doi.org/10.1016/j.proeng.2017.10.571.
US Bureau of Labor Statistics. 2022. National Census of Fatal Occupational Injuries in 2021. 2022.
Wang, C.-Y., A. Bochkovskiy, and H.-Y. M. Liao. 2023a. YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors. 7464–7475.
Wang, M., D. Hu, J. Chen, and S. Li. 2023b. “Underground infrastructure detection and localization using deep learning enabled radargram inversion and vision based mapping.” Automation in Construction, 154: 105004. https://doi.org/10.1016/j.autcon.2023.105004.
Information & Authors
Information
Published In
History
Published online: Jun 13, 2024
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.