Convolutional neural networks for the crack diagnostics in concrete structures
Open Access
- 10 June 2024
- conference paper
- Published by EDP Sciences
- Vol. 534, 01004
Abstract
New models of artificial neural networks are proposed for the identification and classification of cracks in concrete and reinforced concrete walls. The cloud tool Teachable Machine is used to develop a neural network with a pre-defined internal architecture. TensorFlow libraries allow us to develop a convolutional neural network with a tuneable architecture. The program code is written in Python and the learning was performed using the cloud environment Colab. The rational magnitudes of the learning parameters and the topology of the convolutional neural network are determined allowing to achieve the highest accuracy and the lowest losses of the model. The obtained results show a high efficiency of the artificial intelligence to solve problems of the health monitoring of building structures. The proposed models allow real-time automatic diagnostics by analysing photographs, images from smartphone or quadcopter webcams. The latter makes it possible to inspect buildings without the physical presence of humans at the site, which is especially important for working in dangerous places, such as tall buildings, partially destroyed buildings, mined areas, etc. The proposed methods can be further extended for the monitoring and classification of a wide range of defects in building structures.Keywords
This publication has 7 references indexed in Scilit:
- A deep learning‐based image captioning method to automatically generate comprehensive explanations of bridge damageComputer-Aided Civil and Infrastructure Engineering, 2021
- Image-Based Crack Detection Methods: A ReviewInfrastructures, 2021
- Automated Vision-Based Crack Detection on Concrete Surfaces Using Deep LearningApplied Sciences, 2021
- Deep Learning for Detecting Building Defects Using Convolutional Neural NetworksSensors, 2019
- Deep Learning Neural Networks for sUAS-Assisted Structural Inspections: Feasibility and Application2018
- ImageNet classification with deep convolutional neural networksCommunications of the ACM, 2017
- SDNET2018: A concrete crack image dataset for machine learning applications