Lightweight Road Crack Detection Algorithm Integrating Knowledge Distillation and Fractal Dimension

DOI

10.2991/978-94-6239-721-7_23How to use a DOI?

Keywords

Fractal dimension; Knowledge distillation; Residual network; Road crack detection

Abstract

Accurate road crack detection plays a vital role in traffic safety and infrastructure maintenance. However, mainstream deep learning-based detection models suffer from high computational costs and are difficult to deploy on resource-constrained edge devices. This paper proposes FDKDNet, a lightweight road crack detection network that integrates knowledge distillation and fractal dimension analysis. The ResNet101-based teacher model is enhanced with asymmetric convolutions to effectively capture slender crack features. The ResNet18-based student model utilizes Gaussian process-guided Bayesian optimization to adaptively optimize the distillation temperature and realize efficient knowledge transfer. Furthermore, a novel fractal feature module is designed to quantify the morphological complexity of cracks within anchor boxes. Experimental results on the Crack500 dataset show that FDKDNet achieves 90.3% detection accuracy with only one-fourth the parameters of the teacher model, and outperforms existing state-of-the-art methods. The proposed method provides an efficient and practical solution for road crack detection on resource-limited platforms.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Zihao Zhao
AU  - Jun Cheng
AU  - Yuanyuan Li
PY  - 2026
DA  - 2026/07/06
TI  - Lightweight Road Crack Detection Algorithm Integrating Knowledge Distillation and Fractal Dimension
BT  - Proceedings of the 2026 6th International Conference on Public Management and Intelligent Society (PMIS 2026)
PB  - Atlantis Press
SP  - 239
EP  - 246
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6239-721-7_23
DO  - 10.2991/978-94-6239-721-7_23
ID  - Zhao2026
ER  -