Road Crack Detection Using Deep Convolutional Neural Network and Adaptive Thresholding

By Rui Fan, Mohammud Junaid Bocus, Yilong Zhu, Jianhao Jiao, Li Wang, Fulong Ma, Shanshan Cheng, Meilin Liu. 2019 IEEE Intelligent Vehicles Symposium (IV), 2019

552 citations recorded by the source corpus.

Crack is one of the most common road distresses which may pose road safety hazards. Generally, crack detection is performed by either certified inspectors or structural engineers. This task is, however, time-consuming, subjective and labor-intensive. In this paper, a novel road crack detection algorithm which is based on deep learning and adaptive image segmentation is proposed. Firstly, a deep convolutional neural network is trained to determine whether an image contains cracks or not. The images containing cracks are then smoothed using bilateral filtering, which greatly minimizes the number of noisy pixels. Finally, cracks are extracted from the road surface using an adaptive thresholding method. The experimental results illustrate that our network can classify images with an accuracy of 99.92%, and the cracks can be successfully extracted from the images using our proposed thresholding algorithm.

Subfields: Civil and Structural Engineering

View Paper: https://www.semanticscholar.org/paper/a228ce656599fe1413ed0671c260b5dad1190e3e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Civil and Structural Engineering

Citations in the corpus

Surface Defect Detection Methods for Industrial Products: A Review (2021, background): “com/aniruddhsharma/structural-defects-network-concrete-crackimages (accessed on 21 June 2021) Surface Crack Detection Dataset [129,130] Crack on concrete surface https://www.”
How to get pavement distress detection ready for deep learning? A systematic approach (2017, methodology): “Surprisingly, the network with significantly more weights (ASINVOS net) performed only slightly better that the relatively small RCD net 99×99 [17].”
Deep Learning for Change Detection in Remote Sensing Images: Comprehensive Review and Meta-Analysis (2020, methodology): “The CNNs are hierarchical models which converts the input image into multiple layers of feature maps.”
A review of computer vision–based structural health monitoring at local and global levels (2020, methodology): “Continued Topics of structural condition assessment Methodology and examples Crack detection in steel structure Machine learning Boosting: Yeum and Dyke80 SVM: Chen et al.81 Boltzman machine-based ANN: Xu et al.82 Deep learning CNN for image classification: Xu et al.83 FCN for semantic segmentation:”
Data-Driven Structural Health Monitoring and Damage Detection through Deep Learning: State-of-the-Art Review (2020, methodology): “Dung and Anh [33] Concrete crack detection, segmentation, and density evaluation A public dataset of 40,000 concrete crack 227 × 227 pixel images”
Comparison of deep convolutional neural networks and edge detectors for image-based crack detection in concrete (2018, methodology): “DCNNs have been used in vision-based structural health monitoring in recent years for crack detection 197 [42], road pavement cracks [55, 56], corrosion detection [57, 58], multi-damage detection [41, 59] structural 198 health monitoring [62].”
Feature Pyramid and Hierarchical Boosting Network for Pavement Crack Detection (2019, methodology): “Moreover, Zhang et al [1] compare their method with hand crafted feature based methods to demonstrate the advantages of feature representation of deep”