A Method of Concrete Surface Crack Detection Using an Improved Convolutional Neural Network (CNN) Model: Fid-Bau Portal

Fachinformationsdienst BAUdigital

A platform for research: civil engineering, architecture and urbanism

A Method of Concrete Surface Crack Detection Using an Improved Convolutional Neural Network (CNN) Model

He, Zhexin / Zhang, Huan

This essay spotlights concrete crack detection in infrastructure maintenance, highlighting its importance for structural integrity, cost-effectiveness, and eco-consciousness. It delves into various detection methods and introduces an improved VGG-16-based deep learning model with batch normalization, P-ReLU activation, and Adam optimization for better training outcomes. Through experiments on the MendeleyData-CrackDetection dataset, the enhanced model outperforms the original. This study underscores the significance of hyperparameter optimization and algorithm choice in deep learning.

Access

Page navigation

Document information

Export, share and cite

A Method of Concrete Surface Crack Detection Using an Improved Convolutional Neural Network (CNN) Model

He, Zhexin / Zhang, Huan

This essay spotlights concrete crack detection in infrastructure maintenance, highlighting its importance for structural integrity, cost-effectiveness, and eco-consciousness. It delves into various detection methods and introduces an improved VGG-16-based deep learning model with batch normalization, P-ReLU activation, and Adam optimization for better training outcomes. Through experiments on the MendeleyData-CrackDetection dataset, the enhanced model outperforms the original. This study underscores the significance of hyperparameter optimization and algorithm choice in deep learning.

A Method of Concrete Surface Crack Detection Using an Improved Convolutional Neural Network (CNN) Model

He, Zhexin / Zhang, Huan

This essay spotlights concrete crack detection in infrastructure maintenance, highlighting its importance for structural integrity, cost-effectiveness, and eco-consciousness. It delves into various detection methods and introduces an improved VGG-16-based deep learning model with batch normalization, P-ReLU activation, and Adam optimization for better training outcomes. Through experiments on the MendeleyData-CrackDetection dataset, the enhanced model outperforms the original. This study underscores the significance of hyperparameter optimization and algorithm choice in deep learning.

Access

Page navigation

Document information

Export, share and cite

Document information

Title:

A Method of Concrete Surface Crack Detection Using an Improved Convolutional Neural Network (CNN) Model

Additional title:

Lecture Notes in Civil Engineering

Contributors:

Xiang, Ping (editor) / Zuo, Liangdong (editor) / He, Zhexin (author) / Zhang, Huan (author)

Conference:

International Prefabricated Building Seminar on Frontier Technology and Talent Training ; 2023 ; Chongqin, China

Published in:

Novel Technology and Whole-Process Management in Prefabricated Building ; Chapter: 36 ; 335-345

Lecture Notes in Civil Engineering ; 382

Publication date:

2024-07-20

Size:

11 pages

ISBN:

978-981-97-5108-2 , 978-981-97-5107-5

ISSN:

2366-2565 , 2366-2557

DOI:

https://doi.org/10.1007/978-981-97-5108-2_36

Type of media:

Article/Chapter (Book)

Type of material:

Electronic Resource

Language:

English

Keywords:

concrete crack detection , an improved VGG-16 model , Adam optimziation Engineering , Building Construction and Design , Computer-Aided Engineering (CAD, CAE) and Design

Licence:

http://creativecommons.org/licenses/by/4.0/

Similar titles

Autonomous concrete crack detection using deep fully convolutional neural network

Dung, Cao Vu / Anh, Le Duc | British Library Online Contents | 2019

Autonomous concrete crack detection using deep fully convolutional neural network

Dung, Cao Vu / Anh, Le Duc | Elsevier | 2018

A vision-based active learning convolutional neural network model for concrete surface crack detection

Wang, Zhen / Xu, Guoshan / Ding, Yong et al. | SAGE Publications | 2020

Fast Crack Detection Using Convolutional Neural Network

Yang, Jiesheng / Lin, Fangzheng / Xiang, Yusheng et al. | TIBKAT | 2021

Fast Crack Detection Using Convolutional Neural Network

Yang, J. / Lin, F. / Xiang, Yusheng et al. | DataCite | 2021

Free access