Traffic Sign Recognition using Capsule Network
DOI:
https://doi.org/10.59395/ijadis.v7i2.1599Keywords:
Capsule network, Convolutional neural network, Recognition, Traffic signAbstract
Traffic sign recognition (TSR) is critical for systems that rely on its output to inform downstream decisions. This study investigates TSR using Capsule Network (CapsNet), an advancement over the convolutional neural network (CNN) that captures spatial relationships between image features, conferring robustness to affine transformations. The proposed architecture comprises a convolutional layer (256 filters, 9X9 kernel, stride 1, ReLU activation), a primary capsule layer (32 channels of 6X6 capsules, each an 8-dimensional vector), and a class capsule layer (one 16-dimensional capsule per target class). The model was evaluated on both original and augmented datasets, the latter incorporating rotations ranging from -5 degrees to +5 degrees. On the original dataset, CapsNet achieved 100% training and testing accuracy with a training loss of 0,0048 at epoch 20. On the augmented dataset, the model achieved 100% training accuracy (loss: 0,0056) and 98% testing accuracy (5 misclassifications). Compared to multi-scale CNN (MS-CNN), which produced 7 misclassifications on the augmented dataset, CapsNet demonstrated superior consistency and robustness under affine transformations. These findings suggest that CapsNet is a viable and effective architecture for real-world TSR applications.
Downloads
References
[1] K. Bengler, K. Dietmayer, B. Farber, M. Maurer, C. Stiller, and H. Winner, “Three Decades of Driver Assistance Systems: Review and Future Perspectives,” IEEE Intell. Transp. Syst. Mag., vol. 6, no. 4, pp. 6–22, 2014, doi: 10.1109/MITS.2014.2336271. DOI: https://doi.org/10.1109/MITS.2014.2336271
[2] T. Günthner and H. Proff, “On the way to autonomous driving: How age influences the acceptance of driver assistance systems,” Transp. Res. Part F Traffic Psychol. Behav., vol. 81, pp. 586–607, Aug. 2021, doi: 10.1016/j.trf.2021.07.006. DOI: https://doi.org/10.1016/j.trf.2021.07.006
[3] V. K. Kukkala, J. Tunnell, S. Pasricha, and T. Bradley, “Advanced Driver-Assistance Systems: A Path Toward Autonomous Vehicles,” IEEE Consum. Electron. Mag., vol. 7, no. 5, pp. 18–25, Sep. 2018, doi: 10.1109/MCE.2018.2828440. DOI: https://doi.org/10.1109/MCE.2018.2828440
[4] A. Ziebinski, R. Cupek, D. Grzechca, and L. Chruszczyk, “Review of advanced driver assistance systems (ADAS),” presented at the PROCEEDINGS OF THE INTERNATIONAL CONFERENCE OF COMPUTATIONAL METHODS IN SCIENCES AND ENGINEERING 2017 (ICCMSE-2017), Thessaloniki, Greece, 2017, p. 120002. doi: 10.1063/1.5012394. DOI: https://doi.org/10.1063/1.5012394
[5] M. M. Narkhede and N. B. Chopade, “Review of Advanced Driver Assistance Systems and Their Applications for Collision Avoidance in Urban Driving Scenario,” in Machine Learning and Big Data Analytics (Proceedings of International Conference on Machine Learning and Big Data Analytics (ICMLBDA) 2021), vol. 256, R. Misra, R. K. Shyamasundar, A. Chaturvedi, and R. Omer, Eds., in Lecture Notes in Networks and Systems, vol. 256. , Cham: Springer International Publishing, 2022, pp. 253–267. doi: 10.1007/978-3-030-82469-3_23. DOI: https://doi.org/10.1007/978-3-030-82469-3_23
[6] J. Levinson et al., “Towards fully autonomous driving: Systems and algorithms,” in 2011 IEEE Intelligent Vehicles Symposium (IV), Baden-Baden, Germany: IEEE, Jun. 2011, pp. 163–168. doi: 10.1109/IVS.2011.5940562. DOI: https://doi.org/10.1109/IVS.2011.5940562
[7] A. Suriya Prakash, D. Vigneshwaran, R. Seenivasaga Ayyalu, and S. Jayanthi Sree, “Traffic Sign Recognition using Deeplearning for Autonomous Driverless Vehicles,” in 2021 5th International Conference on Computing Methodologies and Communication (ICCMC), Erode, India: IEEE, Apr. 2021, pp. 1569–1572. doi: 10.1109/ICCMC51019.2021.9418437. DOI: https://doi.org/10.1109/ICCMC51019.2021.9418437
[8] Y. Zhu and W. Q. Yan, “Traffic sign recognition based on deep learning,” Multimed. Tools Appl., vol. 81, no. 13, pp. 17779–17791, May 2022, doi: 10.1007/s11042-022-12163-0. DOI: https://doi.org/10.1007/s11042-022-12163-0
[9] X. R. Lim, C. P. Lee, K. M. Lim, T. S. Ong, A. Alqahtani, and M. Ali, “Recent Advances in Traffic Sign Recognition: Approaches and Datasets,” Sensors, vol. 23, no. 10, p. 4674, May 2023, doi: 10.3390/s23104674. DOI: https://doi.org/10.3390/s23104674
[10] A. Hamza and S. Nawal, “Traffic Sign Classification using Deep Learning Comparative Study,” Procedia Comput. Sci., vol. 233, pp. 939–949, 2024, doi: 10.1016/j.procs.2024.03.283. DOI: https://doi.org/10.1016/j.procs.2024.03.283
[11] G. Sapijaszko, T. Alobaidi, and W. B. Mikhael, “Traffic Sign Recognition Based on Multilayer Perceptron Using DWT and DCT,” in 2019 IEEE 62nd International Midwest Symposium on Circuits and Systems (MWSCAS), Dallas, TX, USA: IEEE, Aug. 2019, pp. 440–443. doi: 10.1109/MWSCAS.2019.8884897. DOI: https://doi.org/10.1109/MWSCAS.2019.8884897
[12] A. Kerim and M. O. Efe, “Recognition of Traffic Signs with Artificial Neural Networks: A Novel Dataset and Algorithm,” in 2021 International Conference on Artificial Intelligence in Information and Communication (ICAIIC), Jeju Island, Korea (South): IEEE, Apr. 2021, pp. 171–176. doi: 10.1109/ICAIIC51459.2021.9415238. DOI: https://doi.org/10.1109/ICAIIC51459.2021.9415238
[13] W. Li, H. Song, and P. Wang, “Finely Crafted Features for Traffic Sign Recognition,” Int. J. Circuits Syst. Signal Process., vol. 16, pp. 159–170, Jan. 2022, doi: 10.46300/9106.2022.16.20. DOI: https://doi.org/10.46300/9106.2022.16.20
[14] B. Wang, “Research on the Optimal Machine Learning Classifier for Traffic Signs,” SHS Web Conf., vol. 144, p. 03014, 2022, doi: 10.1051/shsconf/202214403014. DOI: https://doi.org/10.1051/shsconf/202214403014
[15] Y. Le Cun et al., “Handwritten Digit Recognition: Applications of Neural Net Chips and Automatic Learning,” in Neurocomputing, F. F. Soulié and J. Hérault, Eds., Berlin, Heidelberg: Springer Berlin Heidelberg, 1990, pp. 303–318. doi: 10.1007/978-3-642-76153-9_35. DOI: https://doi.org/10.1007/978-3-642-76153-9_35
[16] M. Akbar, “Traffic sign recognition using convolutional neural networks,” J. Teknol. Dan Sist. Komput., vol. 9, no. 2, pp. 120–125, Apr. 2021, doi: 10.14710/jtsiskom.2021.13959. DOI: https://doi.org/10.14710/jtsiskom.2021.13959
[17] M. Akbar, I. Susilawati, B. S. Jati, and N. Alamsyah, “Multi-Task Learning for Traffic Sign Recognition using Multi-Scale Convolutional Neural Networks,” Int. J. Adv. Data Inf. Syst., vol. 6, no. 2, pp. 391–402, Aug. 2025, doi: 10.59395/ijadis.v6i2.1406. DOI: https://doi.org/10.59395/ijadis.v6i2.1406
[18] S. Sabour, N. Frosst, and G. E. Hinton, “Dynamic Routing Between Capsules,” 2017, doi: 10.48550/ARXIV.1710.09829.
[19] S. J. Pawan and J. Rajan, “Capsule networks for image classification: A review,” Neurocomputing, vol. 509, pp. 102–120, Oct. 2022, doi: 10.1016/j.neucom.2022.08.073. DOI: https://doi.org/10.1016/j.neucom.2022.08.073
[20] R. LaLonde, Z. Xu, I. Irmakci, S. Jain, and U. Bagci, “Capsules for Biomedical Image Segmentation,” 2020, arXiv. doi: 10.48550/ARXIV.2004.04736. DOI: https://doi.org/10.1016/j.media.2020.101889
[21] S. Bonheur, D. Štern, C. Payer, M. Pienn, H. Olschewski, and M. Urschler, “Matwo-CapsNet: A Multi-label Semantic Segmentation Capsules Network,” in Medical Image Computing and Computer Assisted Intervention – MICCAI 2019, vol. 11768, D. Shen, T. Liu, T. M. Peters, L. H. Staib, C. Essert, S. Zhou, P.-T. Yap, and A. Khan, Eds., in Lecture Notes in Computer Science, vol. 11768. , Cham: Springer International Publishing, 2019, pp. 664–672. doi: 10.1007/978-3-030-32254-0_74. DOI: https://doi.org/10.1007/978-3-030-32254-0_74
[22] S. J. Pawan et al., “Capsule Network–based architectures for the segmentation of sub-retinal serous fluid in optical coherence tomography images of central serous chorioretinopathy,” Med. Biol. Eng. Comput., vol. 59, no. 6, pp. 1245–1259, Jun. 2021, doi: 10.1007/s11517-021-02364-4. DOI: https://doi.org/10.1007/s11517-021-02364-4
[23] T. Zeng, H. K.-H. So, and E. Y. Lam, “RedCap: residual encoder-decoder capsule network for holographic image reconstruction,” Opt. Express, vol. 28, no. 4, p. 4876, Feb. 2020, doi: 10.1364/OE.383350. DOI: https://doi.org/10.1364/OE.383350
[24] A. Jaiswal, W. AbdAlmageed, Y. Wu, and P. Natarajan, “CapsuleGAN: Generative Adversarial Capsule Network,” 2018, arXiv. doi: 10.48550/ARXIV.1802.06167. DOI: https://doi.org/10.1007/978-3-030-11015-4_38
[25] K. Duarte, Y. S. Rawat, and M. Shah, “VideoCapsuleNet: A Simplified Network for Action Detection,” 2018, arXiv. doi: 10.48550/ARXIV.1805.08162.
[26] K. Duarte, Y. S. Rawat, and M. Shah, “CapsuleVOS: Semi-Supervised Video Object Segmentation Using Capsule Routing,” 2019, arXiv. doi: 10.48550/ARXIV.1910.00132. DOI: https://doi.org/10.1109/ICCV.2019.00857
[27] J. Bae and D.-S. Kim, “End-to-End Speech Command Recognition with Capsule Network,” in Interspeech, 2018. [Online]. Available: https://api.semanticscholar.org/CorpusID:52192557 DOI: https://doi.org/10.21437/Interspeech.2018-1888
[28] J. Poncelet, V. Renkens, and H. Van Hamme, “Low resource end-to-end spoken language understanding with capsule networks,” Comput. Speech Lang., vol. 66, p. 101142, Mar. 2021, doi: 10.1016/j.csl.2020.101142. DOI: https://doi.org/10.1016/j.csl.2020.101142
[29] J. Kim, S. Jang, E. Park, and S. Choi, “Text classification using capsules,” Neurocomputing, vol. 376, pp. 214–221, Feb. 2020, doi: 10.1016/j.neucom.2019.10.033. DOI: https://doi.org/10.1016/j.neucom.2019.10.033
[30] M. U. Haq, M. A. J. Sethi, and A. U. Rehman, “Capsule Network with Its Limitation, Modification, and Applications—A Survey,” Mach. Learn. Knowl. Extr., vol. 5, no. 3, pp. 891–921, Aug. 2023, doi: 10.3390/make5030047. DOI: https://doi.org/10.3390/make5030047
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Mutaqin Akbar, Agus Sidiq Purnomo, Supatman Supatman, Bernadete Deta
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
How to Cite
Share
Plum Analytics