Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Deep semantic segmentation for visual understanding on construction sites
Visual understanding on construction sites by deep learning, such as semantic segmentation, is hardly mentioned in the literature due to the severe lack of labeled data sets. To resolve this issue, we collect and label 859 images, including 12 classes of objects in construction activities, from different construction sites. We then adopt DeepLabV3+ on this data set with modifications. We leverage the Cityscape data set to pretrain the model, and then fine‐tune it on our collected data set. Moreover, multiple data augmentation techniques are utilized to expand the training data set. Our model reaches 0.6467 mean intersection over union (mIoU) and 92.62% mean pixel accuracy (mPA) in the out‐of‐sample test with the capability of processing over 45 frames per second with a resolution of pixels. In addition, we develop a synthetic robotic system integrated with red–green–blue (RGB)‐depth camera for visual understanding on sites. It can detect the depth information of objects and has high potential in automated construction and visual surveillance.
Deep semantic segmentation for visual understanding on construction sites
Visual understanding on construction sites by deep learning, such as semantic segmentation, is hardly mentioned in the literature due to the severe lack of labeled data sets. To resolve this issue, we collect and label 859 images, including 12 classes of objects in construction activities, from different construction sites. We then adopt DeepLabV3+ on this data set with modifications. We leverage the Cityscape data set to pretrain the model, and then fine‐tune it on our collected data set. Moreover, multiple data augmentation techniques are utilized to expand the training data set. Our model reaches 0.6467 mean intersection over union (mIoU) and 92.62% mean pixel accuracy (mPA) in the out‐of‐sample test with the capability of processing over 45 frames per second with a resolution of pixels. In addition, we develop a synthetic robotic system integrated with red–green–blue (RGB)‐depth camera for visual understanding on sites. It can detect the depth information of objects and has high potential in automated construction and visual surveillance.
Deep semantic segmentation for visual understanding on construction sites
Wang, Zifeng (Autor:in) / Zhang, Yuyang (Autor:in) / Mosalam, Khalid M. (Autor:in) / Gao, Yuqing (Autor:in) / Huang, Shao‐Lun (Autor:in)
Computer‐Aided Civil and Infrastructure Engineering ; 37 ; 145-162
01.02.2022
18 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Deep semantic segmentation for visual scene understanding of soil types
| Elsevier | 2022
Semantic Segmentation-based Visual Detection of Construction Objects on Oversized Excavation Sites
| Springer Verlag | 2024