Leakage Identification of Underground Structures Using Classification Deep Neural Networks and Transfer Learning

被引:1
|
作者
Wang, Wenyang [1 ,2 ]
Chen, Qingwei [1 ,2 ]
Shen, Yongjiang [3 ,4 ]
Xiang, Zhengliang [3 ,4 ]
机构
[1] Shandong Zhiyuan Elect Power Design Consulting Co, Jinan 250021, Peoples R China
[2] State Grid Shandong Elect Power Co, Econ & Technol Res Inst, Jinan 250021, Peoples R China
[3] Cent South Univ, Hunan Prov Key Lab Disaster Prevent & Mitigat Rail, Changsha 410075, Peoples R China
[4] Cent South Univ, Sch Civil Engn, Changsha 410075, Peoples R China
基金
中国博士后科学基金; 中国国家自然科学基金;
关键词
underground structures; water leakage defect; computer vision; transfer learning; deep learning; WATER LEAKAGE; RECOGNITION; CRACK;
D O I
10.3390/s24175569
中图分类号
O65 [分析化学];
学科分类号
070302 ; 081704 ;
摘要
Water leakage defects often occur in underground structures, leading to accelerated structural aging and threatening structural safety. Leakage identification can detect early diseases of underground structures and provide important guidance for reinforcement and maintenance. Deep learning-based computer vision methods have been rapidly developed and widely used in many fields. However, establishing a deep learning model for underground structure leakage identification usually requires a lot of training data on leakage defects, which is very expensive. To overcome the data shortage, a deep neural network method for leakage identification is developed based on transfer learning in this paper. For comparison, four famous classification models, including VGG16, AlexNet, SqueezeNet, and ResNet18, are constructed. To train the classification models, a transfer learning strategy is developed, and a dataset of underground structure leakage is created. Finally, the classification performance on the leakage dataset of different deep learning models is comparatively studied under different sizes of training data. The results showed that the VGG16, AlexNet, and SqueezeNet models with transfer learning can overall provide higher and more stable classification performance on the leakage dataset than those without transfer learning. The ResNet18 model with transfer learning can overall provide a similar value of classification performance on the leakage dataset than that without transfer learning, but its classification performance is more stable than that without transfer learning. In addition, the SqueezeNet model obtains an overall higher and more stable performance than the comparative models on the leakage dataset for all classification metrics.
引用
收藏
页数:23
相关论文
共 50 条
  • [41] Transfer Learning of Deep Neural Networks for Visual Collaborative Maritime Asset Identification
    Warren, Nicholas
    Garrard, Benjamin
    Staudt, Elliot
    Tesic, Jelena
    2018 4TH IEEE INTERNATIONAL CONFERENCE ON COLLABORATION AND INTERNET COMPUTING (CIC 2018), 2018, : 246 - 255
  • [42] Deep convolutional neural networks for construction and demolition waste classification: VGGNet structures, cyclical learning rate, and knowledge transfer
    Lin, Kunsen
    Zhou, Tao
    Gao, Xiaofeng
    Li, Zongshen
    Duan, Huabo
    Wu, Huanyu
    Lu, Guanyou
    Zhao, Youcai
    JOURNAL OF ENVIRONMENTAL MANAGEMENT, 2022, 318
  • [43] Deep convolutional neural networks for construction and demolition waste classification: VGGNet structures, cyclical learning rate, and knowledge transfer
    Lin, Kunsen
    Zhou, Tao
    Gao, Xiaofeng
    Li, Zongshen
    Duan, Huabo
    Wu, Huanyu
    Lu, Guanyou
    Zhao, Youcai
    JOURNAL OF ENVIRONMENTAL MANAGEMENT, 2022, 318
  • [44] Using Convolutional Neural Networks and Transfer Learning for Bone Age Classification
    Zhou, Jianlong
    Li, Zelin
    Zhi, Weiming
    Liang, Bin
    Moses, Daniel
    Dawes, Laughlin
    2017 INTERNATIONAL CONFERENCE ON DIGITAL IMAGE COMPUTING - TECHNIQUES AND APPLICATIONS (DICTA), 2017, : 17 - 22
  • [45] Coffee Maturity Classification Using Convolutional Neural Networks and Transfer Learning
    Tamayo-Monsalve, Manuel Alejandro
    Mercado-Ruiz, Esteban
    Villa-Pulgarin, Juan Pablo
    Bravo-Ortiz, Mario Alejandro
    Arteaga-Arteaga, Harold Brayan
    Mora-Rubio, Alejandro
    Alzate-Grisales, Jesus Alejandro
    Arias-Garzon, Daniel
    Romero-Cano, Victor
    Orozco-Arias, Simon
    Gustavo-Osorio, Gustavo
    Tabares-Soto, Reinel
    IEEE ACCESS, 2022, 10 : 42971 - 42982
  • [46] Compact Convolutional Neural Networks for Pterygium Classification using Transfer Learning
    Abdani, Siti Raihanah
    Zulkifley, Mohd Asyraf
    Hussain, Aini
    PROCEEDINGS OF THE 2019 IEEE INTERNATIONAL CONFERENCE ON SIGNAL AND IMAGE PROCESSING APPLICATIONS (IEEE ICSIPA 2019), 2019, : 140 - 143
  • [47] Lesion classification in mammograms using convolutional neural networks and transfer learning
    Perre, Ana C.
    Alexandre, Luis A.
    Freire, Luis C.
    COMPUTER METHODS IN BIOMECHANICS AND BIOMEDICAL ENGINEERING-IMAGING AND VISUALIZATION, 2019, 7 (5-6): : 550 - 556
  • [48] Lesion Classification in Mammograms Using Convolutional Neural Networks and Transfer Learning
    Perre, Ana
    Alexandre, Luis A.
    Freire, Luis C.
    VIPIMAGE 2017, 2018, 27 : 360 - 368
  • [49] Water Leakage and Crack Identification in Tunnels Based on Transfer-Learning and Convolutional Neural Networks
    Man, Ke
    Liu, Ruilin
    Liu, Xiaoli
    Song, Zhifei
    Liu, Zongxu
    Cao, Zixiang
    Wu, Liwen
    WATER, 2022, 14 (09)
  • [50] Transfer Learning for Maritime Vessel Detection using Deep Neural Networks
    Farahnakian, Fahimeh
    Zelioli, Luca
    Heikkonen, Jukka
    2021 IEEE INTELLIGENT TRANSPORTATION SYSTEMS CONFERENCE (ITSC), 2021, : 1 - 6