Abstract:Traditional methods of detecting concrete damage, such as manual inspection, are typically slow, labor‐intensive, and subjective. Integrating deep learning algorithms has automated this process, representing a significant advance in building damage inspection. Semantic segmentation, a technique within deep learning, is increasingly recognized for its capability to identify the location and shape of concrete damage accurately. Unlike basic deep learning approaches like classification and object detection, semantic segmentation not only recognizes damage but also delineates its boundaries, and has great potential in facilitating dimension measurement. However, multi‐level, high‐precision detection for various structural damage assessments remains an area requiring further research. This study created a database of reinforced concrete surfaces with pixel‐level, multi‐category semantic segmentation annotations for various levels of component damage, including cracks and spalling of concrete, exposure, buckling, fracture of reinforcing bars, etc. The performance of advanced deep learning segmentation algorithms, including U‐Net, DeepLab, K‐Net, and FastSCNN, was consequently trained and evaluated in detecting various types of concrete damage. All models show good accuracy of more than 98%, but lower F1 scores around 70%. U‐Net and K‐Net demonstrate relatively stable performance, indicating a certain degree of consistency and a high‐performance peak. In contrast, DeepLabv3's F1 score fluctuates significantly, suggesting the model may suffer from overfitting or other stability issues during training, resulting in a relatively low average performance. FastSCNN exhibits the highest potential, achieving the highest F1 score. In addition, this study also tested the performance of each model on new damage images and explored the impact of the dataset proportion ratio (training set vs. validation set), contributing to providing insights into each algorithm's suitability for various inspection scenarios. Finally, perspectives on current challenges and future directions in the field have also been given.

Deep Learning-Based Damage Inspection for Concrete Structures

Model-informed Deep Learning Strategy with Vision Measurement for Damage Identification of Truss Structures

A real-time multi-defect automatic identification framework for concrete dams via improved YOLOv5 and knowledge distillation

Image-based Reinforced Concrete Component Mechanical Damage Recognition and Structural Safety Rapid Assessment Using Deep Learning with Frequency Information

A review of the research and application of deep learning-based computer vision in structural damage detection

Point Cloud-Based Dimensional Quality Assessment of Precast Concrete Components Using Deep Learning

Computer vision and deep learning-based post-earthquake intelligent assessment of engineering structures: Technological status and challenges

Automatic damage detection and segmentation using deep learning algorithms in reinforced concrete structure inspections

Automatic assessment of freeze-thaw damage in concrete structures using piezoelectric-based active sensing approach and deep learning technique

Structural Crack Detection Using Deep Learning–based Fully Convolutional Networks

Engineering deep learning methods on automatic detection of damage in infrastructure due to extreme events

Deep Learning-Based Concrete Surface Damage Monitoring Method Using Structured Lights and Depth Camera

Multi-scale Image-Based Damage Recognition and Assessment for Reinforced Concrete Structures in Post-Earthquake Emergency Response

Deep Learning for Concrete Crack Detection and Measurement

Deep learning-based concrete defects classification and detection using semantic segmentation

Deep Learning for Structural Health Monitoring: Data, Algorithms, Applications, Challenges, and Trends

Vibration-based damage detection for bridges by deep convolutional denoising autoencoder

A novel deep learning-based method for damage identification of smart building structures

A novel structural damage identification method based on the acceleration responses under ambient vibration and an optimized deep residual algorithm

Deep learning-based structural health monitoring through the infusion of optical photos and vibration data

A novel intelligent inspection robot with deep stereo vision for three-dimensional concrete damage detection and quantification