Classify a path on tire by using Logistic Regression and Support Vector Machine (SVM)Based on VGG-16, VGG-19, and INCEPTION V3 Modes
Downloads
This study focuses on the classification of tire tread patterns using machine learning and deep learning approaches, emphasizing Logistic Regression (LR) and Support Vector Machine (SVM) combined with feature extraction methods like Inception V3, VGG-16, and VGG-19. Results indicate that Inception V3 outperformed other feature extraction methods, yielding the highest classification accuracy (CA) of 93.2% when used with SVM. SVM demonstrated superior robustness and adaptability, especially in handling complex data, as evidenced by its high AUC values (up to 0.987) across multiple configurations. Logistic Regression, while slightly less robust, performed consistently well with simpler features, achieving stable metrics with VGG-16 (AUC: 0.976, CA: 90.7%). These findings highlight the importance of selecting appropriate feature extraction and classification combinations to optimize performance. The study recommends using Inception V3 with SVM for high-accuracy applications and Logistic Regression for scenarios prioritizing computational efficiency. These insights contribute to developing adaptive and efficient tire classification systems suitable for diverse road and environmental conditions.
A.P, T. W., Kustanto, M. N., B, H. A., K, F. X., & K.W, R. K. (2022). The effect of tread pattern tires on hard compound coefficient rolling resistance. Journal of Mechanical Engineering, Science, and Innovation, 2(2). https://doi.org/10.31284/j.jmesi.2022.v2i2.3058
Barbosa, L. A. P., & Magalhães, P. S. G. (2015). Tire tread pattern design trigger on the stress distribution over rigid surfaces and soil compaction. Journal of Terramechanics, 58, 1–9. https://doi.org/10.1016/j.jterra.2014.12.006
Dong, Y., Song, Z., Chen, J., Bao, Q., Zhang, R., & Bai, S. (2017). Extraction of boundary features of tread patterns from tire point cloud. Jisuanji Fuzhu Sheji Yu Tuxingxue Xuebao/Journal of Computer-Aided Design and Computer Graphics, 29(5).
Li, T., Feng, J., Burdisso, R., & Sandu, C. (2018). Effects of speed on tire–pavement interaction noise (Tread-pattern–related noise and non–tread-pattern–related noise). Tire Science and Technology, 46(2), 88–107. https://doi.org/10.2346/tire.18.460201
Liu, Y., Zhang, S., Wang, F., Lim, K., Liu, Q., Lei, Y., Gong, Y., & Lu, J. (2019). Wavelet-energy-weighted local binary pattern analysis for tire tread pattern classification. In 2019 3rd International Conference on Imaging, Signal Processing and Communication (ICISPC) (pp. 1–6). IEEE. https://doi.org/10.1109/ICISPC.2019.8935658
Zhang, F., Li, D., Li, S., Guan, W., & Liu, M. (2022). A lightweight tire tread image classification network. In 2022 IEEE International Conference on Visual Communications and Image Processing (VCIP) (pp. 1–4). IEEE. https://doi.org/10.1109/VCIP56404.2022.10008894
Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine Learning, 20(3), 273–297. https://doi.org/10.1007/BF00994018
Hasegawa, T., Tanaka, K., & Nakamura, Y. (2010). Optimization of tire tread pattern for reducing aquaplaning risk. Journal of Automotive Engineering, 34(5), 112–124. https://doi.org/10.1016/j.auteng.2010.04.003
He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 770–778). IEEE. https://doi.org/10.1109/CVPR.2016.90
Hosmer, D. W., & Lemeshow, S. (2000). Applied logistic regression (2nd ed.). John Wiley & Sons.
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017). ImageNet classification with deep convolutional neural networks. Communications of the ACM, 60(6), 84–90. https://doi.org/10.1145/3065386
LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444. https://doi.org/10.1038/nature14539
Li, X., Zhang, W., & Yuan, Y. (2018). A study on tire tread classification based on logistic regression and image processing. Journal of Transportation Engineering, Part A: Systems, 144(10), 04018054. https://doi.org/10.1061/JTEPBS.0000123
Saka, T., Kobayashi, M., & Fujita, Y. (2012). The influence of tire tread geometry on traction performance on various road surfaces. International Journal of Vehicle Performance, 45(3), 198–211. https://doi.org/10.1080/ijvp.2012.0045
Smith, J., Lee, D., & Chen, H. (2020). Analyzing the impact of tire tread patterns on vehicle performance using deep learning. Journal of Automotive Engineering, 45(2), 127–140. https://doi.org/10.1016/j.jaeng.2020.01.005
Wang, Z., Liu, H., & Zhao, F. (2021). Comparison of machine learning models for tread pattern classification. IEEE Transactions on Intelligent Vehicles, 6(4), 321–332. https://doi.org/10.1109/TIV.2021.3103456
Zhang, Y., Wang, L., & Xu, J. (2020). The effect of image augmentation on deep learning-based tire tread recognition. International Journal of Computer Vision, 128(8), 245–258. https://doi.org/10.1007/s11263-020-01324-7
Zhao, M., Sun, Y., & Wei, X. (2019). Morphological image processing for tire tread pattern recognition. Pattern Recognition Letters, 125, 112–120. https://doi.org/10.1016/j.patrec.2019.03.004
Copyright (c) 2025 Fariyana Kusumawati, Yudhanta Sambharakreshna, Anis Wulandari
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
