Automatic intelligent recognition of pavement distresses with limited dataset using generative adversarial networks

Liu, Z., Pan, S., Gao, Z. , Chen, N., Li, F., Wang, L. and Hou, Y. (2023) Automatic intelligent recognition of pavement distresses with limited dataset using generative adversarial networks. Automation in Construction, 146, 104674. (doi: 10.1016/j.autcon.2022.104674)

[img] Text
286102.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

2MB

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%).

Item Type:Articles
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Gao, Dr Zhiwei
Authors: Liu, Z., Pan, S., Gao, Z., Chen, N., Li, F., Wang, L., and Hou, Y.
College/School:College of Science and Engineering > School of Engineering > Infrastructure and Environment
Journal Name:Automation in Construction
Publisher:Elsevier
ISSN:0926-5805
ISSN (Online):1872-7891
Published Online:26 November 2022
Copyright Holders:Copyright © 2022 Elsevier B.V.
First Published:First published in Automation in Construction 146:104674
Publisher Policy:Reproduced in accordance with the publisher copyright policy

University Staff: Request a correction | Enlighten Editors: Update this record