Journal article 175 views
Automatic intelligent recognition of pavement distresses with limited dataset using generative adversarial networks
Zhuo Liu,
Shuo Pan,
Zhiwei Gao,
Ning Chen,
Feng Li,
Linbing Wang,
Yue Hou
Automation in Construction, Volume: 146, Start page: 104674
Swansea University Author: Yue Hou
DOI (Published version): 10.1016/j.autcon.2022.104674
Abstract
Automatic monitoring of pavement structure health has always been a significant problem for transportation engineers. Although the generative adversarial network (GAN) has proven to be an effective tool for improving pavement distress recognition accuracy, it may lead to increased computational cost...
| Published in: | Automation in Construction |
|---|---|
| ISSN: | 0926-5805 |
| Published: |
Elsevier BV
2023
|
| Online Access: |
Check full text
|
| URI: | https://cronfa.swan.ac.uk/Record/cronfa62096 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract: |
Automatic monitoring of pavement structure health has always been a significant problem for transportation engineers. Although the generative adversarial network (GAN) has proven to be an effective tool for improving pavement distress recognition accuracy, it may lead to increased computational cost, which inconsistent with the requirements of engineering practice. This paper describes a lightweight GAN structure for automatic pavement distress identification with high computation efficiency and low computation cost. Squeeze and expand (SE), multiscale convolution (MC), and depthwise separable convolution (DSC) were selected as alternative lightweight methods, and two series of comparative experiments were conducted. The results showed that the GAN-based model with SE implemented on its fully connected layer, MC&DSC implemented on its transpose convolution layers in the generator, and MC implemented on its convolution layers in the discriminator could reduce the largest proportion of model parameters (94.8%) while achieving satisfactory classification accuracy (85.4%). |
|---|---|
| Keywords: |
Automatic intelligent recognition; Pavement distresses; Lightweight GAN; Multiscale convolution; Depthwise separable convolution |
| College: |
Faculty of Science and Engineering |
| Funders: |
This work was supported by the Opening Project Fund of Materials Service Safety Assessment Facilities (MSAF- 2021-109), Key Science and Technology Projects in the Transportation Industry in 2021 (2021-ZD2-047), 2021 Science and Technology Innovation Project of Shandong Hi-Speed Group (SDGS-2021-0472-2), and the National Natural Science Foundation of China (Grant no. 51708026). |
| Start Page: |
104674 |