An efficient approach for automatic crack detection using deep learning
International Journal of Structural Integrity
ISSN: 1757-9864
Article publication date: 9 April 2024
Issue publication date: 13 May 2024
Abstract
Purpose
Automation of detecting cracked surfaces on buildings or in any industrially manufactured products is emerging nowadays. Detection of the cracked surface is a challenging task for inspectors. Image-based automatic inspection of cracks can be very effective when compared to human eye inspection. With the advancement in deep learning techniques, by utilizing these methods the authors can create automation of work in a particular sector of various industries.
Design/methodology/approach
In this study, an upgraded convolutional neural network-based crack detection method has been proposed. The dataset consists of 3,886 images which include cracked and non-cracked images. Further, these data have been split into training and validation data. To inspect the cracks more accurately, data augmentation was performed on the dataset, and regularization techniques have been utilized to reduce the overfitting problems. In this work, VGG19, Xception and Inception V3, along with Resnet50 V2 CNN architectures to train the data.
Findings
A comparison between the trained models has been performed and from the obtained results, Xception performs better than other algorithms with 99.54% test accuracy. The results show detecting cracked regions and firm non-cracked regions is very efficient by the Xception algorithm.
Originality/value
The proposed method can be way better back to an automatic inspection of cracks in buildings with different design patterns such as decorated historical monuments.
Keywords
Citation
Usharani, S., Gayathri, R., Kovvuri, U.S.D.R., Nivas, M., Md, A.Q., Tee, K.F. and Sivaraman, A.K. (2024), "An efficient approach for automatic crack detection using deep learning", International Journal of Structural Integrity, Vol. 15 No. 3, pp. 434-460. https://doi.org/10.1108/IJSI-10-2023-0102
Publisher
:Emerald Publishing Limited
Copyright © 2024, Emerald Publishing Limited