Journal article 426 views 81 downloads
Automatic pavement texture recognition using lightweight few-shot learning
,
Hai Yan ,
Zhuo Liu ,
Ning Chen,
Yinghao Miao,
Yue Hou
Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, Volume: 381, Issue: 2254
Swansea University Author: Yue Hou
-
PDF | Version of Record
© 2023 The Authors. Published by the Royal Society. Distributed under the terms of a Creative Commons Attribution 4.0 Licence (CC BY 4.0).
Download (1004.02KB)
DOI (Published version): 10.1098/rsta.2022.0166
Abstract
Texture is a crucial characteristic of roads, closely related to their performance. The recognition of pavement texture is of great significance for road maintenance professionals to detect potential safety hazards and carry out necessary countermeasures. Although deep learning models have been appl...
| Published in: | Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences |
|---|---|
| ISSN: | 1364-503X 1471-2962 |
| Published: |
The Royal Society
2023
|
| Online Access: |
Check full text
|
| URI: | https://cronfa.swan.ac.uk/Record/cronfa63895 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract: |
Texture is a crucial characteristic of roads, closely related to their performance. The recognition of pavement texture is of great significance for road maintenance professionals to detect potential safety hazards and carry out necessary countermeasures. Although deep learning models have been applied for recognition, the scarcity of data has always been a limitation. To address this issue, this paper proposes a few-shot learning model based on the Siamese network for pavement texture recognition with a limited dataset. The model achieved 89.8% accuracy in a four-way five-shot task classifying the pavement textures of dense asphalt concrete, micro surface, open-graded friction course and stone matrix asphalt. To align with engineering practice, global average pooling (GAP) and one-dimensional convolution are implemented, creating lightweight models that save storage and training time. Comparative experiments show that the lightweight model with GAP implemented on dense layers and one-dimensional convolution on convolutional layers reduced storage volume by 94% and training time by 99%, despite a 2.9% decrease in classification accuracy. Moreover, the model with only GAP implemented on dense layers achieved the highest accuracy at 93.5%, while reducing storage volume and training time by 83% and 6%, respectively. |
|---|---|
| Keywords: |
Pavement detection, deep learning, convolutional neural network, Siamese network, one-dimensional convolution |
| College: |
Faculty of Science and Engineering |
| Funders: |
This work was supported by Key Science and Technology Projects in the Transportation Industry in 2021 (grant no. 2021-ZD2-047), Key Science and Technology Projects in the Transportation Industry in 2022 (grant no. 2022-ZD6-116), Plan Project of Shandong Transportation S&T (grant no. 2021B49), Opening Project Fund of Materials Service Safety Assessment Facilities (grant no. MSAF-2021-005) and National Natural Science Foundation of China (grant no. 51978048). |
| Issue: |
2254 |