Deep learning-based visual crack detection using Google Street View images

• Published in: Neural computing & applications 2021-11, Vol.33 (21), p.14565-14582
• Main Authors: Maniat, Mohsen, Camp, Charles V., Kashani, Ali R.
• Format: Article
• Language: English
• Subjects: Artificial Intelligence; Artificial neural networks; Asphalt pavements; Comparative studies; Computational Biology/Bioinformatics; Computational Science and Engineering; Computer Science; Data Mining and Knowledge Discovery; Datasets; Evaluation; Image classification; Image Processing and Computer Vision; Inspection; Machine learning; Neural networks; Original Article; Pavements; Probability and Statistics in Computer Science; Quality assessment; Smartphones
• ISSN: 0941-0643; 1433-3058
• DOI: 10.1007/s00521-021-06098-0

In this study, the utility of using Google Street View (GSV) for evaluating the quality of pavement is investigated. A convolutional neural network (CNN) is developed to perform image classification on GSV pavement images. Pavement images are extracted from GSV and then divided into smaller image patches to form data sets. Each image patch is visually classified into different categories of pavement cracks based on the standard practice. A comparative study of pavement quality assessment is conducted between the results of the CNN classified image patches obtained from GSV and those from a sophisticated commercial visual inspection company. The result of the comparison indicates the feasibility and effectiveness of using GSV images for pavement evaluation. The trained network is then tested on a new data set. This study shows that the designed CNN helps classify the pavement images into different defined crack categories.