Abstract
The increasing exploitation of underground space has resulted in numerous oversized excavation projects which entail widely distributed risk factors and therefore demand a real-time risk management without blind spots. To facilitate vision-based risk identification and management, this study develops a comprehensive image dataset labeled in semantic mask level and an enhanced semantic segmentation algorithm to enable visual detection of the precise boundaries of construction objects on oversized excavation sites. Taking an oversized deep excavation project in Ningbo, Zhejiang Province, China as the case study, we first created an image dataset comprising real images, synthetic images generated from BIM, and web images crawled from the Internet. Ten classes of objects in construction activities of excavation sites, including worker, machine, and structure were selected as the target objects and finely annotated. A DeepLabv3+ algorithm modified with a lightweight MobileNetV2 backbone network and sub-pixel convolution (MobileNetV2-s) was employed on the developed dataset. Moreover, a compound loss function and transfer learning technique were leveraged for better algorithm training. Results demonstrate that MobileNetV2 achieves a mIoU of 69.67% and a mPA of 83.81% with an inference speed of 37.17 frames per second at a resolution of 1280 × 720 pixels, which strikes an optimal balance between performance and efficiency. The present study offers a promising solution for more efficient and reliable vision-based risk identification and management in oversized excavation projects.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Ding L, Fang W, Luo H et al (2018) A deep hybrid learning model to detect unsafe behavior: Integrating convolution neural networks and long short-term memory. Autom Constr 86:118–124
Yan X, Zhang H, Li H (2020) Computer vision-based recognition of 3D relationship between construction entities for monitoring struck-by accidents. Comput-Aided Civil Infrastruct Eng 35(9):1023–1038
Russell BC, Torralba A, Murphy KP et al (2008) LabelMe: a database and web-based tool for image annotation. Int J Comput Vision 77(1):157–173
Chen L-C, Zhu Y, Papandreou G, et al (2018) Encoder-decoder with atrous separable convolution for semantic image segmentation. In: Ferrari V, Hebert M, Sminchisescu C, et al (eds) Computer vision—ECCV 2018. Springer, Heidelberg, pp 833–851
Sandler M, Howard A, Zhu M, et al (2018) MobileNetV2: inverted residuals and linear bottlenecks. In: 2018 IEEE/CVF conference on computer vision and pattern recognition, pp 4510–4520
Shi W, Caballero J, Huszár F, et al (2016) Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network. In: 2016 IEEE conference on computer vision and pattern recognition (CVPR), pp 1874–1883
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Yang, YF., Liao, SM., Wang, W. (2024). Semantic Segmentation-based Visual Detection of Construction Objects on Oversized Excavation Sites. In: Wu, W., Leung, C.F., Zhou, Y., Li, X. (eds) Proceedings of the 18th Conference of the Associated Research Centers for the Urban Underground Space. ACUUS 2023. Lecture Notes in Civil Engineering, vol 471. Springer, Singapore. https://doi.org/10.1007/978-981-97-1257-1_94
Download citation
DOI: https://doi.org/10.1007/978-981-97-1257-1_94
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-97-1256-4
Online ISBN: 978-981-97-1257-1
eBook Packages: EngineeringEngineering (R0)