Utilizing Mean Teacher Semi-Supervised Learning for Robust Pothole Image Classification (제한된 레이블 데이터에 대한 Mean Teacher 반지도학습을 활용한 견고한 포트홀 이미지 분류)

Inki Kim (김인기); Beomjun Kim (김범준); Jeonghwan Gwak (곽정환)

doi:10.9708/jksci.2023.28.05.017

Utilizing Mean Teacher Semi-Supervised Learning for Robust Pothole Image Classification

Journal of The Korea Society of Computer and Information
Abbr : JKSCI
2023, 28(5), pp.17-28
DOI : 10.9708/jksci.2023.28.05.017
Publisher : The Korean Society Of Computer And Information
Research Area : Engineering > Computer Science
Received : April 24, 2023
Accepted : May 15, 2023
Published : May 31, 2023

Inki Kim ¹, Beomjun Kim ¹, Jeonghwan Gwak ¹

¹한국교통대학교

Accredited

ABSTRACT

Potholes that occur on paved roads can have fatal consequences for vehicles traveling at high speeds and may even lead to fatalities. While manual detection of potholes using human labor is commonly used to prevent pothole-related accidents, it is economically and temporally inefficient due to the exposure of workers on the road and the difficulty in predicting potholes in certain categories. Therefore, completely preventing potholes is nearly impossible, and even preventing their formation is limited due to the influence of ground conditions closely related to road environments. Additionally, labeling work guided by experts is required for dataset construction. Thus, in this paper, we utilized the Mean Teacher technique, one of the semi-supervised learning-based knowledge distillation methods, to achieve robust performance in pothole image classification even with limited labeled data. We demonstrated this using performance metrics and GradCAM, showing that when using semi-supervised learning, 15 pre-trained CNN models achieved an average accuracy of 90.41%, with a minimum of 2% and a maximum of 9% performance difference compared to supervised learning.

KEYWORDS

Pothole detection, Semi-supervised learning, Knowledge distillation, Mean Teacher technique, pre-trained models, transfer learning

Citation status

* References for papers published after 2023 are currently being built.

[book] A. Tarvainen / 2017 / Mean Teachers Are Better Role Models:Weight-Averaged Consistency Targets Improve Semi-Supervised Deep Learning Results / Advances in Neural Information Processing System (NIPS 2017)

[confproc] R. R. Selvaraju / 2017 / Grad-CAM: Visual Explanations from Deep Networks Via Gradient-Based Localization / Proceedings of the IEEE International Conference on Computer Vision (ICCV) : 618~626

[confproc] K. K. Vupparaboina / 2015 / Laser-Based Detection and Depth Estimation of Dry and Water-Filled Potholes: A Geometric Approach / Twenty First National Conference on Communications (NCC) : 1~6

[confproc] X. Yu / 2011 / Pavement Pothole Detection And Severity Measurement Using Laser Imaging / IEEE International Conference on Electro/Information Technology(EIT) : 1~5

[confproc] S. Nienaber / 2015 / Detecting Potholes Using Simple Image Processing Techniques and Real-World Footage / Proceedings of the 34th Southern African Transport Conference (SATC 2015) : 6~9

[confproc] K. Vigneshwar / 2016 / Detection and Counting of Pothole Using Image Processing Techniques / IEEE International Conference on Computational Intelligence and Computing Research (ICCIC) : 1~4

[web] A. A. Shaghouri / 2021 / Real-Time Pothole Detection Using Deep Learning / arXiv / arXiv:2107.06356

[journal] M. H. Asad / 2022 / Pothole Detection Using Deep Learning: A Real-Time and AI-on-the-Edge Perspective / Advances in Civil Engineering

[confproc] N. Çınar / 2022 / An automated pothole detection via transfer learning / 2022 International Conference on Decision Aid Sciences and Applications (DASA) : 1355~1358

[confproc] I. K. Kim / 2022 / Pothole Detection Method based on an Ensemble of Pretrained Deep Learning Models / 2022 Korean Society for Internet Information (KSII) SPRING Conference

[web] S. Laine / 2016 / Temporal Ensembling for Semi-Supervised Learning / arXiv / arXiv:1610.02242

[confproc] A. Krizhevsky / 2017 / ImageNet Classification with Deep Convolutional Neural Networks / Proceedings of the 25th International Conference on Neural Information Processing Systems(NIPS) 60(6) : 84~90

[confproc] G. Huang / 2017 / Densely Connected Convolutional Networks / Proceedings of the IEEE conference on computer vision and pattern recognition(CVPR) : 4700~4708

[confproc] M. Tan / 2019 / EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks / International conference on machine learning(ICML) : 6105~6114

[confproc] C. Szegedy / 2016 / Rethinking the Inception Architecture for Computer Vision / 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) : 2818~2826

[confproc] M. Tan / 2019 / MnasNet : Platform-Aware Neural Architecture Search for Mobile / Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR) : 2820~2828

[web] G. Andrew / 2017 / MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications / arXiv / 10.48550/arXiv.1704.04861

[journal] Jing Xu / 2022 / RegNet: Self-Regulated Network for Image Classification / IEEE Transactions on Neural Networks and Learning Systems / Institute of Electrical and Electronics Engineers (IEEE) 34(11) : 1~6

[confproc] K. He / 2016 / Deep Residual Learning for Image Recognition / 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) : 770~778

[confproc] S. Xie / 2017 / Aggregated Residual Transformations for Deep Neural Networks / Proceddings of the IEEE conference on computer vision and pattern recognition(CVPR) : 1492~1500

[confproc] X. Zhang / 2018 / ShuffleNet: An Extremely Efficient Convolutional Nerual Network for Mobile Devices / Proceddings of the IEE conference on computer vision and pattern recognition(CVPR) : 6848~6856

[web] K. Simonyan / 2014 / Very Deep Convolutional Networks for Large-Scale Image Recognition / arXiv / arXiv:1409.1556

This paper was written with support from the National Research Foundation of Korea.

KJCKorea
Journal Central

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65

Utilizing Mean Teacher Semi-Supervised Learning for Robust Pothole Image Classification

ABSTRACT

KEYWORDS

Citation status

* References for papers published after 2023 are currently being built.

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65

Utilizing Mean Teacher Semi-Supervised Learning for Robust Pothole Image Classification

ABSTRACT

KEYWORDS

Statistics

Tools

Issue List

Citation status

KCI Citation Counts (3)

REFERENCES (22) * References for papers published after 2023 are currently being built.

Search PDF

Citation

* References for papers published after 2023 are currently being built.