Loading [MathJax]/extensions/tex2jax.js

Surrounding rock classification from onsite images with deep transfer learning based on EfficientNet

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autorschaft

  • Xiaoying Zhuang
  • Wenjie Fan
  • Hongwei Guo
  • Xuefeng Chen
  • Qimin Wang

Organisationseinheiten

Externe Organisationen

  • Tongji University
  • Guizhou Xingyi Huancheng Expressway Co., Ltd.

Details

OriginalspracheEnglisch
Seiten (von - bis)1311-1320
Seitenumfang10
FachzeitschriftFrontiers of Structural and Civil Engineering
Jahrgang18
Ausgabenummer9
Frühes Online-Datum13 Aug. 2024
PublikationsstatusVeröffentlicht - Sept. 2024

Abstract

This paper proposes an accurate, efficient and explainable method for the classification of the surrounding rock based on a convolutional neural network (CNN). The state-of-the-art robust CNN model (EfficientNet) is applied to tunnel wall image recognition. Gaussian filtering, data augmentation and other data pre-processing techniques are used to improve the data quality and quantity. Combined with transfer learning, the generality, accuracy and efficiency of the deep learning (DL) model are further improved, and finally we achieve 89.96% accuracy. Compared with other state-of-the-art CNN architectures, such as ResNet and Inception-ResNet-V2 (IRV2), the presented deep transfer learning model is more stable, accurate and efficient. To reveal the rock classification mechanism of the proposed model, Gradient-weight Class Activation Map (Grad-CAM) visualizations are integrated into the model to enable its explainability and accountability. The developed deep transfer learning model has been applied to support the tunneling of the Xingyi City Bypass in the high mountain area of Guizhou, China, with great results.

ASJC Scopus Sachgebiete

Zitieren

Surrounding rock classification from onsite images with deep transfer learning based on EfficientNet. / Zhuang, Xiaoying; Fan, Wenjie; Guo, Hongwei et al.
in: Frontiers of Structural and Civil Engineering, Jahrgang 18, Nr. 9, 09.2024, S. 1311-1320.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Zhuang, X, Fan, W, Guo, H, Chen, X & Wang, Q 2024, 'Surrounding rock classification from onsite images with deep transfer learning based on EfficientNet', Frontiers of Structural and Civil Engineering, Jg. 18, Nr. 9, S. 1311-1320. https://doi.org/10.1007/s11709-024-1134-7
Zhuang, X., Fan, W., Guo, H., Chen, X., & Wang, Q. (2024). Surrounding rock classification from onsite images with deep transfer learning based on EfficientNet. Frontiers of Structural and Civil Engineering, 18(9), 1311-1320. https://doi.org/10.1007/s11709-024-1134-7
Zhuang X, Fan W, Guo H, Chen X, Wang Q. Surrounding rock classification from onsite images with deep transfer learning based on EfficientNet. Frontiers of Structural and Civil Engineering. 2024 Sep;18(9):1311-1320. Epub 2024 Aug 13. doi: 10.1007/s11709-024-1134-7
Zhuang, Xiaoying ; Fan, Wenjie ; Guo, Hongwei et al. / Surrounding rock classification from onsite images with deep transfer learning based on EfficientNet. in: Frontiers of Structural and Civil Engineering. 2024 ; Jahrgang 18, Nr. 9. S. 1311-1320.
Download
@article{6869e6450606456d8b84ec47aa839668,
title = "Surrounding rock classification from onsite images with deep transfer learning based on EfficientNet",
abstract = "This paper proposes an accurate, efficient and explainable method for the classification of the surrounding rock based on a convolutional neural network (CNN). The state-of-the-art robust CNN model (EfficientNet) is applied to tunnel wall image recognition. Gaussian filtering, data augmentation and other data pre-processing techniques are used to improve the data quality and quantity. Combined with transfer learning, the generality, accuracy and efficiency of the deep learning (DL) model are further improved, and finally we achieve 89.96% accuracy. Compared with other state-of-the-art CNN architectures, such as ResNet and Inception-ResNet-V2 (IRV2), the presented deep transfer learning model is more stable, accurate and efficient. To reveal the rock classification mechanism of the proposed model, Gradient-weight Class Activation Map (Grad-CAM) visualizations are integrated into the model to enable its explainability and accountability. The developed deep transfer learning model has been applied to support the tunneling of the Xingyi City Bypass in the high mountain area of Guizhou, China, with great results.",
keywords = "convolutional neural network, EfficientNet, Gradient-weight Class Activation Map, surrounding rock classification",
author = "Xiaoying Zhuang and Wenjie Fan and Hongwei Guo and Xuefeng Chen and Qimin Wang",
note = "Publisher Copyright: {\textcopyright} Higher Education Press 2024.",
year = "2024",
month = sep,
doi = "10.1007/s11709-024-1134-7",
language = "English",
volume = "18",
pages = "1311--1320",
number = "9",

}

Download

TY - JOUR

T1 - Surrounding rock classification from onsite images with deep transfer learning based on EfficientNet

AU - Zhuang, Xiaoying

AU - Fan, Wenjie

AU - Guo, Hongwei

AU - Chen, Xuefeng

AU - Wang, Qimin

N1 - Publisher Copyright: © Higher Education Press 2024.

PY - 2024/9

Y1 - 2024/9

N2 - This paper proposes an accurate, efficient and explainable method for the classification of the surrounding rock based on a convolutional neural network (CNN). The state-of-the-art robust CNN model (EfficientNet) is applied to tunnel wall image recognition. Gaussian filtering, data augmentation and other data pre-processing techniques are used to improve the data quality and quantity. Combined with transfer learning, the generality, accuracy and efficiency of the deep learning (DL) model are further improved, and finally we achieve 89.96% accuracy. Compared with other state-of-the-art CNN architectures, such as ResNet and Inception-ResNet-V2 (IRV2), the presented deep transfer learning model is more stable, accurate and efficient. To reveal the rock classification mechanism of the proposed model, Gradient-weight Class Activation Map (Grad-CAM) visualizations are integrated into the model to enable its explainability and accountability. The developed deep transfer learning model has been applied to support the tunneling of the Xingyi City Bypass in the high mountain area of Guizhou, China, with great results.

AB - This paper proposes an accurate, efficient and explainable method for the classification of the surrounding rock based on a convolutional neural network (CNN). The state-of-the-art robust CNN model (EfficientNet) is applied to tunnel wall image recognition. Gaussian filtering, data augmentation and other data pre-processing techniques are used to improve the data quality and quantity. Combined with transfer learning, the generality, accuracy and efficiency of the deep learning (DL) model are further improved, and finally we achieve 89.96% accuracy. Compared with other state-of-the-art CNN architectures, such as ResNet and Inception-ResNet-V2 (IRV2), the presented deep transfer learning model is more stable, accurate and efficient. To reveal the rock classification mechanism of the proposed model, Gradient-weight Class Activation Map (Grad-CAM) visualizations are integrated into the model to enable its explainability and accountability. The developed deep transfer learning model has been applied to support the tunneling of the Xingyi City Bypass in the high mountain area of Guizhou, China, with great results.

KW - convolutional neural network

KW - EfficientNet

KW - Gradient-weight Class Activation Map

KW - surrounding rock classification

UR - http://www.scopus.com/inward/record.url?scp=85201303623&partnerID=8YFLogxK

U2 - 10.1007/s11709-024-1134-7

DO - 10.1007/s11709-024-1134-7

M3 - Article

AN - SCOPUS:85201303623

VL - 18

SP - 1311

EP - 1320

JO - Frontiers of Structural and Civil Engineering

JF - Frontiers of Structural and Civil Engineering

SN - 2095-2430

IS - 9

ER -