Interpretability Analysis of Convolutional Neural Networks for Crack Detection

被引：4

作者：

Wu, Jie ^{[1
]}

He, Yongjin ^{[2
]}

Xu, Chengyu ^{[3
]}

Jia, Xiaoping ^{[3
]}

Huang, Yule ^{[2
]}

Chen, Qianru ^{[4
]}

Huang, Chuyue ^{[1
]}

Eslamlou, Armin Dadras ^{[2
]}

Huang, Shiping ^{[2
]}

机构：

[1] Wuhan Polytech Univ, Sch Civil Engn & Architecture, Wuhan 430023, Peoples R China

[2] South China Univ Technol, Sch Civil Engn & Transportat, Guangzhou 510640, Peoples R China

[3] China Railway 17th Bur Grp Guangzhou Co Ltd, Guangzhou 510799, Peoples R China

[4] South China Normal Univ, Sch Elect & Informat Engn, Foshan 528225, Peoples R China

来源：

BUILDINGS | 2023年 / 13卷 / 12期

关键词：

structural health monitoring; crack detection; interpretability analysis; convolutional neural network;

D O I：

10.3390/buildings13123095

中图分类号：

TU [建筑科学];

学科分类号：

0813 ;

摘要：

Crack detection is an important task in bridge health monitoring, and related detection methods have gradually shifted from traditional manual methods to intelligent approaches with convolutional neural networks (CNNs) in recent years. Due to the opaque process of training and operating CNNs, if the learned features for identifying cracks in the network are not evaluated, it may lead to safety risks. In this study, to evaluate the recognition basis of different crack detection networks; several crack detection CNNs are trained using the same training conditions. Afterwards, several crack images are used to construct a dataset, which are used to interpret and analyze the trained networks and obtain the learned features for identifying cracks. Additionally, a crack identification performance criterion based on interpretability analysis is proposed. Finally, a training framework is introduced based on the issues reflected in the interpretability analysis.

引用

页数：14

共 50 条

[11] Investigations on Plate Crack Damage Detection Using Convolutional Neural Networks
Ma, Dongliang
Wang, Deyu
[J]. INTERNATIONAL JOURNAL OF OFFSHORE AND POLAR ENGINEERING, 2021, 31 (02) : 220 - 229
[12] UNCERTAINTY MODELING AND INTERPRETABILITY IN CONVOLUTIONAL NEURAL NETWORKS FOR POLYP SEGMENTATION
Wickstrom, Kristoffer
Kampffmeyer, Michael
Jenssen, Robert
[J]. 2018 IEEE 28TH INTERNATIONAL WORKSHOP ON MACHINE LEARNING FOR SIGNAL PROCESSING (MLSP), 2018,
[13] Automated Video Interpretability Assessment using Convolutional Neural Networks
Kalukin, Andrew R.
[J]. 2018 IEEE APPLIED IMAGERY PATTERN RECOGNITION WORKSHOP (AIPR), 2018,
[14] Improve the interpretability of convolutional neural networks with probability density function
[J]. Yang, Jie (yangjiee@dlut.edu.cn), 2025, 699
[15] Ensemble of Deep Convolutional Neural Networks for Automatic Pavement Crack Detection and Measurement
Fan, Zhun
Li, Chong
Chen, Ying
Di Mascio, Paola
Chen, Xiaopeng
Zhu, Guijie
Loprencipe, Giuseppe
[J]. COATINGS, 2020, 10 (02)
[16] Convolutional neural networks-based crack detection for real concrete surface
Li, Shengyuan
Zhao, Xuefeng
[J]. SENSORS AND SMART STRUCTURES TECHNOLOGIES FOR CIVIL, MECHANICAL, AND AEROSPACE SYSTEMS 2018, 2018, 10598
[17] Deep Convolutional Neural Networks for Road Crack Detection: Qualitative and Quantitative Comparisons
Fan, Jiahe
Bocus, Mohammud Junaid
Wang, Li
Fan, Rui
[J]. 2021 IEEE INTERNATIONAL CONFERENCE ON IMAGING SYSTEMS AND TECHNIQUES (IST), 2021,
[18] Enhancing Interpretability in Medical Image Classification by Integrating Formal Concept Analysis with Convolutional Neural Networks
Khatri, Minal
Yin, Yanbin
Deogun, Jitender
[J]. BIOMIMETICS, 2024, 9 (07)
[19] Uncertainty and interpretability in convolutional neural networks for semantic segmentation of colorectal polyps
Wickstrom, Kristoffer
Kampffmeyer, Michael
Jenssen, Robert
[J]. MEDICAL IMAGE ANALYSIS, 2020, 60
[20] Interpretability and Optimisation of Convolutional Neural Networks Based on Sinc-Convolution
Habib, Ahsan
Karmakar, Chandan
Yearwood, John
[J]. IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, 2023, 27 (04) : 1758 - 1769

← 1 2 3 4 5 →