통합 이미지 처리 기술을 이용한 콘크리트 교량 균열 탐지 및 매핑 = Crack Inspection and Mapping of Concrete Bridges using Integrated Image Processing Techniques

다국어 초록 (Multilingual Abstract)

In many developed countries, such as South Korea, efficiently maintaining the aging infrastructures is an important issue. Currently, inspectors visually inspect the infrastructure for maintenance needs, but this method is inefficient due to its high costs, long logistic times, and hazards to the inspectors. Thus, in this paper, a novel crack inspection approach for concrete bridges is proposed using integrated image processing techniques. The proposed approach consists of four steps: (1) training a deep learning model to automatically detect cracks on concrete bridges, (2) acquiring in-situ images using a drone, (3) generating orthomosaic images based on 3D modeling, and (4) detecting cracks on the orthmosaic image using the trained deep learning model. Cascade Mask R-CNN, a state-of-the-art instance segmentation deep learning model, was trained with 3235 crack images that included 2415 hard negative images. We selected the Tancheon overpass, located in Seoul, South Korea, as a testbed for the proposed approach, and we captured images of pier 34-37 and slab 34-36 using a commercial drone. Agisoft Metashape was utilized as a 3D model generation program to generate an orthomosaic of the captured images. We applied the proposed approach to four orthomosaic images that displayed the front, back, left, and right sides of pier 37. Using pixel-level precision referencing visual inspection of the captured images, we evaluated the trained Cascade Mask R-CNN's crack detection performance. At the coping of the front side of pier 37, the model obtained its best precision: 94.34%. It achieved an average precision of 72.93% for the orthomosaics of the four sides of the pier. The test results show that this proposed approach for crack detection can be a suitable alternative to the conventional visual inspection method.

참고문헌 (Reference)

1 J. Kiefer, "Stochastic Estimation of the Maximum of a Regression Function" 23 (23): 462-466, 1952

2 G. Leonardi, "Road Degradation Survey through Images by Drone" 101 : 222-228, 2019

3 R. Fan, "Real-Time Dense Stereo Embedded in a UAV for Road Inspection" 2019

4 K. Chen, "MMDetection: Open MMLab Detection Toolbox and Benchmark"

5 J. Deng, "ImageNet: A Large-Scale Hierarchical Image Database" Institute of Electrical and Electronics Engineers (IEEE) 248-255, 2009

6 B. Kim, "Image-Based Concrete Crack Assessment Using Mask and Region-Based Convolutional Neural Network" 26 (26): e2381-, 2019

7 T. Y. Lin, "Feature Pyramid Networks for Object Detection" 2017 : 936-944, 2017

8 S. Goessens, "Feasibility Study for Drone-Based Masonry Construction of Real-Scale Structures" 94 : 458-480, 2018

9 S. Ren, "Faster R-CNN : Towards Real-Time Object Detection with Region Proposal Networks" 39 (39): 1137-1149, 2017

10 Byunghyun Kim, "Efflorescence assessment using hyperspectral imaging for concrete structures" 국제구조공학회 22 (22): 209-221, 2018

11 B. Hariharan, "Discriminative Decorrelation for Clustering and Classification" 7575 (7575): 459-472, 2012

12 B. Kim, "Development of Automatic Input Negative Sample Algorithm for Improving Crack Detection Deep Learning Model" 23 (23): 125-, 2019

13 Y. Liu, "Deep Learning-Based Enhancement of Motion Blurred UAV Concrete Crack Images" 34 (34): 04020028-, 2020

14 H. Kim, "Concrete Crack Identification Using a UAV Incorporating Hybrid Image Processing" 17 (17): 2052-, 2017

15 T. Nishikawa, "Concrete Crack Detection by Multiple Sequential Image Filtering" 27 (27): 29-47, 2012

16 Z. Cai, "Cascade R-CNN: High Quality Object Detection and Instance Segmentation" 2019

17 Yufei Liu, "Automated assessment of cracks on concrete surfaces using adaptive digital image processing" 국제구조공학회 14 (14): 719-741, 2014

18 B. Kim, "Automated Vision-based Detection of Cracks on Concrete Structures Using a Deep Learning Technique" 18 : 3452-, 2018

19 Y. Li, "Applications of Multirotor Drone Technologies in Construction Management" 19 (19): 401-412, 2019

20 Agisoft, "Agisoft Metashape"

21 S. Xie, "Aggregated Residual Transformations for Deep Neural Networks" 2017 : 5987-5995, 2017

22 E. Ridolfi, "Accuracy Analysis of a Dam Model from Drone Surveys" 17 (17): 1777-, 2017