|
Title |
Development of Crack Detection System for Highway Tunnels using Imaging Device and Deep Learning
|
|
Authors |
김병현(Byung-Hyun Kim) ; 조수진(Soo-Jin Cho) ; 채홍제(Hong-Je Chae) ; 김홍기(Hong-Ki Kim) ; 강종하(Jong-Ha Kang) |
|
DOI |
https://doi.org/10.11112/jksmi.2021.25.4.65 |
|
Keywords |
터널 균열 탐지; 딥러닝; Cascade Mask R-CNN; 비균열 학습 Tunnel crack detection; Deep learning; Cascade mask R-CNN; Negative sample training |
|
Abstract |
In order to efficiently inspect rapidly increasing old tunnels in many well-developed countries, many inspection methodologies have beenproposed using imaging equipment and image processing. However, most of the existing methodologies evaluated their performance on a clean concretesurface with a limited area where other objects do not exist. Therefore, this paper proposes a 6-step framework for tunnel crack detection deep learningmodel development. The proposed method is mainly based on negative sample (non-crack object) training and Cascade Mask R-CNN. The proposedframework consists of six steps: searching for cracks in images captured from real tunnels, labeling cracks in pixel level, training a deep learning model,collecting non-crack objects, retraining the deep learning model with the collected non-crack objects, and constructing final training dataset. Toimplement the proposed framework, Cascade Mask R-CNN, an instance segmentation model, was trained with 1561 general crack images and 206non-crack images. In order to examine the applicability of the trained model to the real-world tunnel crack detection, field testing is conducted ontunnel spans with a length of about 200m where electric wires and lights are prevalent. In the experimental result, the trained model showed 99%precision and 92% recall, which shows the excellent field applicability of the proposed framework.
|