Attention‐Based Convolutional Neural Network for Pavement Crack Detection

Wan, Haifeng; Gao, Lei; Su, Manman; Sun, Qirun; Huang, Lei

doi:10.1155/2021/5520515

Cited by 32 publications

(14 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From the actual situation and previous studies, it can be analyzed that the pressure and position of the vehicles on the road are constantly changing due to the different sizes and types of vehicles [15,16]. erefore, it is necessary to simplify the vehicle into four forms: wave load, impact load, horizontal movement vertical load, and random load according to the usual method [17,18]. e random load describes the driving trajectory more objectively and is close to reality.…”

Section: Moving Loadsmentioning

confidence: 99%

Viscoelastic-Thermodynamic Model and Crack Propagation in Asphalt Concrete Pavement

Yuan

Zhao

2022

Security and Communication Networks

View full text Add to dashboard Cite

Asphalt concrete pavement has a wide range of applications in the construction of high-grade highways. Asphalt concrete pavement has elastic properties, and its viscoelasticity and viscoplasticity are obviously affected by temperature. To determine whether a structure can continue to be used safely, it is most important to determine whether microscopic or macroscopic cracks present in the structure continue to propagate and cause structural failure. The purpose of this study was to study the effects of different ambient temperatures and mobile vehicles on the stress around the crack tip and the propagation path of different types of asphalt pavements. Based on the theory of fracture mechanics and viscoelasticity, this work studies the problem of crack propagation from a macroscopic perspective. It proposes to use the finite element method to analyze the crack problem in the pavement structure. With the help of ABAQUS software, a curved beam model with cracks was established for calculation, and moving loads and ambient temperature fields were added for analysis. The experimental results in this study show that with the decrease in the ambient temperature, the stress at the crack tip increases continuously. As the analysis time increases, the stress also increases. When the crack is smaller, the stress value at the tip is larger. When a = 15 mm, the stress value at the crack tip is the largest. However, when a = 20 mm and a = 25 mm, the stress at the crack tip is significantly reduced. It shows that the crack is easier to crack when the crack value is smaller. After the crack has grown to a certain length, the growth rate will slow down.

show abstract

Section: Moving Loadsmentioning

confidence: 99%

Viscoelastic-Thermodynamic Model and Crack Propagation in Asphalt Concrete Pavement

Yuan

Zhao

2022

Security and Communication Networks

View full text Add to dashboard Cite

show abstract

“…erefore, the CNN model does not require the feature computation process that extracts numerical input data Although the CNN is a highly capable image classification method [129][130][131][132][133], its performance considerably depends on the size of the collected training samples [53].…”

Section: Resultsmentioning

confidence: 99%

A Novel Approach for Detection of Pavement Crack and Sealed Crack Using Image Processing and Salp Swarm Algorithm Optimized Machine Learning

Hoang

Huynh

Tran

et al. 2022

Advances in Civil Engineering

View full text Add to dashboard Cite

During the phase of periodic survey, sealed crack and crack in asphalt pavement surface should be detected accurately. Moreover, the capability of identifying these two defects can help reduce the false-positive rate for pavement crack detection. Because crack and sealed crack are both line-based defects and may resemble each other in shape, this study puts forward an innovative method based on computer vision for detecting sealed crack and crack. This method is an integration of feature extraction based on image processing and metaheuristic optimized machine learning. Image processing is used to compute features that characterize visual appearance and texture of the pavement image. Subsequently, Salp Swarm Algorithm integrated with multiclass support vector machine is employed for pattern recognition. Based on experimental results, the newly developed method has achieved the most desired predictive performance with an accuracy rate = 91.33% for crack detection and 92.83% for sealed crack detection.

show abstract

“…CRAtt network is based on the architecture of an encoder‐decoder network, and the position attention module and channel attention module are connected after each encoder to summarize remote contextual information [62].…”

Section: Methodsmentioning

confidence: 99%

“…CRAtt network is based on the architecture of an encoderdecoder network, and the position attention module and channel attention module are connected after each encoder to summarize remote contextual information [62]. CrackAttention: The authors build the network using the encoder-decoder architecture and adopt a pyramid module to exploit global context information for the complex topology structures of cracks.…”

Section: Comparison Methodsmentioning

confidence: 99%

CurSeg: A pavement crack detector based on a deep hierarchical feature learning segmentation framework

Yuan

Meng

et al. 2022

IET Intelligent Trans Sys

View full text Add to dashboard Cite

Automatic crack detection is challenging due to the poor continuity of cracks, the different widths of cracks, and the low contrast between cracks and the surrounding pavement. In this paper, a deep convolutional neural network called CurSeg is proposed, which achieves pixelwise segmentation of cracks in an end-to-end manner. In this approach, features at different scales are fused together to attain the context information from the cracks. The elaborately designed model can effectively suppress the propagation of noise and further refine the crack features by aggregating multiscale and multilevel features from low-level to high-level. Residual detail attention (RDA) is also introduced to better capture the line structure and the ability to accurately locate the crack position in a complex context to make the network more discriminative and robust. CurSeg is evaluated on four datasets to validate the effectiveness of the approach. The experimental results demonstrate that this method achieves state-of-the-art performance on the four challenging datasets.

show abstract

Attention‐Based Convolutional Neural Network for Pavement Crack Detection

Cited by 32 publications

References 41 publications

Viscoelastic-Thermodynamic Model and Crack Propagation in Asphalt Concrete Pavement

Viscoelastic-Thermodynamic Model and Crack Propagation in Asphalt Concrete Pavement

A Novel Approach for Detection of Pavement Crack and Sealed Crack Using Image Processing and Salp Swarm Algorithm Optimized Machine Learning

CurSeg: A pavement crack detector based on a deep hierarchical feature learning segmentation framework

Contact Info

Product

Resources

About