Automatic classification of pavement crack using deep convolutional neural network
- 20 June 2018
- journal article
- research article
- Published by Taylor & Francis Ltd in International Journal of Pavement Engineering
- Vol. 21 (4), 457-463
- https://doi.org/10.1080/10298436.2018.1485917
Abstract
The classification of pavement crack heavily relies on the engineers’ experience or the hand-crafted algorithms. Convolutional Neural Network (CNN) has demonstrated to be useful for image classification, which provides an alternative to traditional imaging classification algorithms. This paper proposes a novel method using deep CNN to automatically classify image patches cropped from 3D pavement images. In all, four supervised CNNs with different sizes of receptive field are successfully trained. The experimental results demonstrate that all the proposed CNNs can perform the classification with a high accuracy. Overall classification accuracy of each proposed CNN is above 94%. Upon the evaluation of these neural networks with respect to accuracy and training time, we find that the size of receptive field has a slight effect on the classification accuracy. However, the CNNs with smaller size of receptive field require more training times than others.Keywords
Funding Information
- National Natural Science Foundation of China (U1534203, 51478398)
This publication has 22 references indexed in Scilit:
- CrackTree: Automatic crack detection from pavement imagesPattern Recognition Letters, 2012
- Wavelet-based pavement distress detection and evaluationOptical Engineering, 2006
- Automatic inspection of pavement cracking distressJournal of Electronic Imaging, 2006
- Convolutional face finder: a neural architecture for fast and robust face detectionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2004
- Novel Approach to Pavement Cracking Detection Based on Neural NetworkTransportation Research Record: Journal of the Transportation Research Board, 2001
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998
- Primitive‐Based Classification of Pavement Cracking ImagesJournal of Transportation Engineering, 1993
- Histogram‐Based Approach for Automated Pavement‐Crack SensingJournal of Transportation Engineering, 1992
- Pavement Distress Analysis Using Image Processing TechniquesComputer-Aided Civil and Infrastructure Engineering, 1991
- Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in positionBiological Cybernetics, 1980