A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Automatic Detection and Analysis of Concrete Cracks Using YOLO
This research introduces a novel approach for the automated detection and analysis of concrete cracks using deep learning techniques, with a focus on predicting the lifespan of structures. Concrete infrastructure is susceptible to various forms of deterioration, and cracks are early indicators of potential structural issues. Traditional methods for crack detection and analysis are often time-consuming and subjective. In this study, we utilize deep learning calculations, explicitly convolutional neural networks (CNNs), to distinguish and order breaks in substantial surfaces naturally. The proposed model is trained on a diverse dataset of concrete images, enabling it to generalize across different environmental conditions and crack patterns. Furthermore, the research extends beyond crack detection to predict the potential lifespan of concrete structures. By leveraging historical data on concrete deterioration and utilizing advanced predictive modeling. Based on detected cracks, the system seeks to offer insights into the anticipated durability of structures. This predictive capability serves as a proactive tool for maintenance planning and resource allocation, contributing to the long-term sustainability of civil infrastructure. Our experiments’ outcomes show how well the deep learning model works to identify concrete cracks and forecast possible structural degradation.
Automatic Detection and Analysis of Concrete Cracks Using YOLO
This research introduces a novel approach for the automated detection and analysis of concrete cracks using deep learning techniques, with a focus on predicting the lifespan of structures. Concrete infrastructure is susceptible to various forms of deterioration, and cracks are early indicators of potential structural issues. Traditional methods for crack detection and analysis are often time-consuming and subjective. In this study, we utilize deep learning calculations, explicitly convolutional neural networks (CNNs), to distinguish and order breaks in substantial surfaces naturally. The proposed model is trained on a diverse dataset of concrete images, enabling it to generalize across different environmental conditions and crack patterns. Furthermore, the research extends beyond crack detection to predict the potential lifespan of concrete structures. By leveraging historical data on concrete deterioration and utilizing advanced predictive modeling. Based on detected cracks, the system seeks to offer insights into the anticipated durability of structures. This predictive capability serves as a proactive tool for maintenance planning and resource allocation, contributing to the long-term sustainability of civil infrastructure. Our experiments’ outcomes show how well the deep learning model works to identify concrete cracks and forecast possible structural degradation.
Automatic Detection and Analysis of Concrete Cracks Using YOLO
Hota, C P Pavan Kumar (author) / Sowjanya, Vvnls (author) / Sree, G Ramya (author) / Neeharika, K (author) / Supritha, G (author)
2024-06-21
1243003 byte
Conference paper
Electronic Resource
English
Evaluating YOLO Models for Efficient Crack Detection in Concrete Structures Using Transfer Learning
| DOAJ | 2024
The reinforced concrete trestle over Yolo basin
Engineering Index Backfile | 1914
Building Envelope Object Detection Using YOLO Models
| IEEE | 2022