Automatic Recognition of Highway Tunnel Defects Based on an Improved U-Net Model

It is critically important for the safe operation of highway tunnels that defects in tunnel linings be promptly identified. Recently, many effective identification models based on deep learning technology have emerged in the field of tunnel defect recognition. However, the results of previous studies indicate that imperfections exist in these models in which the boundaries of the defects can only be generally located in the complex visual environment of tunnel linings, which include noisy patterns and uneven light sources. In order to more accurately identify defect details in such complex environments, an improved U-net model including a combined Squeeze-and-Excitation (SE) and ResNet block is proposed and evaluated in this paper. The accuracy of the conventional U-net model is found to be considerably improved by the inclusion of the SE and ResNet blocks. An under-sampling strategy is utilized to address the problem of imbalanced crack image data and determine an appropriate ratio of negative to positive samples for the crack dataset. The effectiveness of the proposed model is then demonstrated using actual highway tunnel lining images to compare its results with those of the Gabor Filter, Multiscale DCNN, FCN, and conventional U-net defect recognition models. The accuracy of the proposed model was found to be quite competitive with extant models.

[1]  Naoki Tanaka,et al.  A Crack Detection Method in Road Surface Images Using Morphology , 1998, MVA.

[2]  Zhou He,et al.  3D Deep Affine-Invariant Shape Learning for Brain MR Image Segmentation , 2018, DLMIA/ML-CDS@MICCAI.

[3]  Sylvie Chambon,et al.  Automatic Road Defect Detection by Textural Pattern Recognition Based on AdaBoost , 2012, Comput. Aided Civ. Infrastructure Eng..

[4]  Yun Liu,et al.  Automatic Crack Detection and Classification Method for Subway Tunnel Safety Monitoring , 2014, Sensors.

[5]  Shivprakash Iyer,et al.  Segmentation of Pipe Images for Crack Detection in Buried Sewers , 2006, Comput. Aided Civ. Infrastructure Eng..

[6]  Tarek Zayed,et al.  Machine vision-based model for spalling detection and quantification in subway networks , 2017 .

[7]  Lijun Wu,et al.  Automated Pixel-Level Surface Crack Detection Using U-Net , 2018, MIWAI.

[8]  Ikhlas Abdel-Qader,et al.  ANALYSIS OF EDGE-DETECTION TECHNIQUES FOR CRACK IDENTIFICATION IN BRIDGES , 2003 .

[9]  Xian Liu,et al.  Predictive maintenance of shield tunnels , 2013 .

[10]  Chang-Soo Han,et al.  Auto inspection system using a mobile robot for detecting concrete cracks in a tunnel , 2007 .

[11]  Maria Q. Feng,et al.  Vision-Based Displacement Sensor for Monitoring Dynamic Response Using Robust Object Search Algorithm , 2013 .

[12]  Mieczysław Mastyło,et al.  Bilinear interpolation theorems and applications , 2013 .

[13]  David Lattanzi,et al.  Robust Automated Concrete Damage Detection Algorithms for Field Applications , 2014, J. Comput. Civ. Eng..

[14]  Nikolaos Doulamis,et al.  Deep Convolutional Neural Networks for efficient vision based tunnel inspection , 2015, 2015 IEEE International Conference on Intelligent Computer Communication and Processing (ICCP).

[15]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[16]  Jaime Gómez García-Bermejo,et al.  Crack Detection in Concrete Tunnels Using a Gabor Filter Invariant to Rotation , 2017, Sensors.

[17]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[18]  Wei Zhang,et al.  Unified Vision‐Based Methodology for Simultaneous Concrete Defect Detection and Geolocalization , 2018, Comput. Aided Civ. Infrastructure Eng..

[19]  Enhua Wu,et al.  Squeeze-and-Excitation Networks , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[20]  Hojjat Adeli,et al.  Supervised Deep Restricted Boltzmann Machine for Estimation of Concrete , 2017 .

[21]  Hojjat Adeli,et al.  A novel unsupervised deep learning model for global and local health condition assessment of structures , 2018 .

[22]  Hao Feng,et al.  A SVM-based pipeline leakage detection and pre-warning system , 2010 .

[23]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[24]  Xuefeng Zhao,et al.  Automatic pixel‐level multiple damage detection of concrete structure using fully convolutional network , 2019, Comput. Aided Civ. Infrastructure Eng..

[25]  Kenichi Soga,et al.  Failure mechanism of joint waterproofing in precast segmental tunnel linings , 2019, Tunnelling and Underground Space Technology.

[26]  B. F. Spencer,et al.  Vision-based Structural Inspection using Multiscale Deep Convolutional Neural Networks , 2018, ArXiv.

[27]  Yicheng Li,et al.  A Fast Detection Method via Region‐Based Fully Convolutional Neural Networks for Shield Tunnel Lining Defects , 2018, Comput. Aided Civ. Infrastructure Eng..

[28]  Chul Min Yeum,et al.  Vision‐Based Automated Crack Detection for Bridge Inspection , 2015, Comput. Aided Civ. Infrastructure Eng..

[29]  Shuai Guo,et al.  Sewer damage detection from imbalanced CCTV inspection data using deep convolutional neural networks with hierarchical classification , 2019, Automation in Construction.

[30]  Dongming Zhang,et al.  Deep learning based image recognition for crack and leakage defects of metro shield tunnel , 2018, Tunnelling and Underground Space Technology.

[31]  Haibo He,et al.  Learning from Imbalanced Data , 2009, IEEE Transactions on Knowledge and Data Engineering.

[32]  Oral Büyüköztürk,et al.  Deep Learning‐Based Crack Damage Detection Using Convolutional Neural Networks , 2017, Comput. Aided Civ. Infrastructure Eng..

[33]  Xiuling Wang,et al.  Statistical analysis of fire accidents in Chinese highway tunnels 2000–2016 , 2019, Tunnelling and Underground Space Technology.

[34]  José García Rodríguez,et al.  A Review on Deep Learning Techniques Applied to Semantic Segmentation , 2017, ArXiv.

[35]  Angelos Amditis,et al.  AUTONOMOUS ROBOTIC INSPECTION IN TUNNELS , 2016 .

[36]  Jack Chin Pang Cheng,et al.  Automated detection of sewer pipe defects in closed-circuit television images using deep learning techniques , 2018, Automation in Construction.