應用影像語意分割技術於鋼筋間距查驗

10.6653/MoCICHE.202202_49(1).0005

紀乃文；莊仕杰；陳翊翔；陳鵬元；陳俊杉

土木水利

49卷1期（2022 / 02 / 01）

27 - 31

繁體中文

鋼筋檢驗在鋼筋混凝土構造物的施工過程當中是一個重要的環節，其檢查要項包含鋼筋之號數、間距、形式（含彎鈎與搭接）等等。傳統鋼筋檢驗仰賴人力完成，並且以局部抽驗的方式為主。但許多鋼筋籠必須在施工現場組立，使得查驗人員無法避免在危險的工作場域對鋼筋進行查驗。在人工智慧蓬勃發展的當下，本研究提出一種基於影像語意分割的技術，配合景深攝影機所取得的RGB-D深度影像，能夠以鋼筋籠上縱橫交錯的結點為辨識目標，並因此計算出結點與結點之間的距離來代表鋼筋間距。結果顯示其誤差可達工地現場實用之等級，並提供鋼筋檢驗自動化的可行參考方向及示範。

1. Badrinarayanan, V.,Kendall, A.,Cipolla, R.(2017).SegNet: A deep convolutional encoder-decoder architecture for image segmentation.IEEE Transactions on Pattern Analysis and Machine Intelligence,39,2481-2495.
2. Chen, L. C.,Papandreou, G.,Kokkinos, I.,Murphy, K.,Yuille, A.(2014).Semantic image segmentation with deep convolutional nets and fully connected CRFs.a conference paper at ICLR 2015
3. Chen, L. C.,Papandreou, G.,Kokkinos, I.,Murphy, K.,Yuille, A.(2016).DeepLab: Semantic image segmentation with deep convo-lutional nets, atrous convolution, and fully connected CRFs.IEEE Transactions on Pattern Analysis and Machine Intelligence
4. Chen, L. C.,Papandreou, G.,Schroff, F.,Adam, H.(2017).,未出版
5. Chen, L. C.,Zhu, Y.,Papandreou, G.,Schroff, F.,Adam, H.(2018).Encoder-decoder with atrous separable convolution for semantic image segmentation.ECCV,7
6. Crous, M.(2018).University of Amsterdam.
7. Gehrig, M.,Morris, D.,Bryant, J.(2004).Technical Presentation Paper for Performance Foundation AssociationTechnical Presentation Paper for Performance Foundation Association,未出版
8. Han, K.,Gwak, J.,Golparvar-Fard, M.,Saidi, K.,Cheok, G.,Franaszek, M.,Lipman, R.(2013).Vision-based field inspection of concrete reinforcing bars.13th International Conference on Construc-tion Applications of Virtual Reality,London, UK:
9. Hugenschmidt, J.,Mastrangelo, R.(2007).The inspection of large retaining walls using GPR.4th International Workshop on Advanced Ground Penetrating Radar
10. Long, J.,Shelhamer, E.,Darrell, T.(2015).Fully convolutional networks for semantic segmentation.2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
11. OSHA (Occupational Safety and Health Administration) (2021). Construction focus four training. https://www.osha.gov/dte/outreach/construction/focus_four/index.html (last access: 2021/01/25).
12. Yu, F.,Koltun,V.(2015).Multi-scale context aggregation by dilated convolutions.a conference paper at ICLR 2016
13. Zaid, M.,Gaydecki, P.,Quek, S.,Miller, G.,Fernandes, B.(2004).Extracting dimensional information from steel reinforcing bars in concrete using neural networks trained on data from an inductive sensor.NDT & E International,37,551-558.
14. Zhan, R.,Xie, H.(2009).GPR measurement of the diameter of steel bars in concrete specimens based on the stationary wavelet transform.Insight,51,151-155.
15. 行政院勞動部職業安全衛生署（2020）。108 年勞動檢查年報。https://www.osha.gov.tw/1106/1164/1165/1168/29804/ (last access: 2021/01/25)。