Haolan Wang
ABSTRACT
Traffic sign recognition is an integral part of future autonomous driving systems. Deep learning has been applied in this task, while the performance of the recent vision Transformers is unexplored. In this study, eight different vision Transformers are validated in three real-world traffic sign datasets for the first time. The experimental results demonstrate that the best vision Transformer has a performance between the pre-trained DenseNet and the DenseNet trained from scratch. Besides, the best vision Transformers model has less training time compared to DenseNet.
- Mathias M, Timofte R, Benenson R, Traffic sign recognition—How far are we from the solution?[C]//The 2013 international joint conference on Neural networks (IJCNN). IEEE, 2013: 1-8.Google Scholar
- Jiang W, Zhang L. Edge-siamnet and edge-triplenet: New deep learning models for handwritten numeral recognition[J]. IEICE Transactions on Information and Systems, 2020, 103(3): 720-723.Google ScholarCross Ref
- Jiang W. Evaluation of deep learning models for Urdu handwritten characters recognition[C]//Journal of Physics: Conference Series. IOP Publishing, 2020, 1544(1): 012016.Google ScholarCross Ref
- Zaibi A, Ladgham A, Sakly A. A Lightweight Model for Traffic Sign Classification Based on Enhanced LeNet-5 Network[J]. Journal of Sensors, 2021, 2021.Google Scholar
- Loukmane A, Graña M, Mestari M. A Model for Classification of Traffic Signs Using Improved Convolutional Neural Network and Image Enhancement[C]//2020 Fourth International Conference On Intelligent Computing in Data Sciences (ICDS). IEEE, 2020: 1-8.Google Scholar
- Haque W A, Arefin S, Shihavuddin A S M, DeepThin: A novel lightweight CNN architecture for traffic sign recognition without GPU requirements[J]. Expert Systems with Applications, 2021, 168: 114481.Google ScholarCross Ref
- Luo X, Zhu J, Yu Q. Efficient convNets for fast traffic sign recognition[J]. IET Intelligent Transport Systems, 2019, 13(6): 1011-1015.Google ScholarCross Ref
- Satti S K, Devi K S, Dhar P, Enhancing and classifying traffic signs using computer vision and deep convolutional neural network[C]//International Conference on Machine Learning, Image Processing, Network Security and Data Sciences. Springer, Singapore, 2020: 243-253.Google Scholar
- Arcos-Garcia A, Alvarez-Garcia J A, Soria-Morillo L M. Evaluation of deep neural networks for traffic sign detection systems[J]. Neurocomputing, 2018, 316: 332-344.Google ScholarCross Ref
- Cao J, Song C, Peng S, Improved traffic sign detection and recognition algorithm for intelligent vehicles[J]. Sensors, 2019, 19(18): 4021.Google ScholarCross Ref
- Yazdan R, Varshosaz M. Improving traffic sign recognition results in urban areas by overcoming the impact of scale and rotation[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2021, 171: 18-35.Google ScholarCross Ref
- Zhang J, Wang W, Lu C, Lightweight deep network for traffic sign classification[J]. Annals of Telecommunications, 2020, 75(7): 369-379.Google ScholarCross Ref
- Gao Y, Xiao G. Real-time chinese traffic warning signs recognition based on cascade and CNN[J]. Journal of Real-Time Image Processing, 2021, 18(3): 669-680.Google ScholarDigital Library
- Stallkamp J, Schlipsing M, Salmen J, The German traffic sign recognition benchmark: a multi-class classification competition[C]//The 2011 international joint conference on neural networks. IEEE, 2011: 1453-1460.Google Scholar
- Zhang J, Xie Z, Sun J, A cascaded R-CNN with multiscale attention and imbalanced samples for traffic sign detection[J]. IEEE Access, 2020, 8: 29742-29754.Google ScholarCross Ref
- Dosovitskiy A, Beyer L, Kolesnikov A, An image is worth 16x16 words: Transformers for image recognition at scale[J]. arXiv preprint arXiv:2010.11929, 2020.Google Scholar
- Zhou D, Kang B, Jin X, Deepvit: Towards deeper vision transformer[J]. arXiv preprint arXiv:2103.11886, 2021.Google Scholar
- Touvron H, Cord M, Sablayrolles A, Going deeper with image transformers[J]. arXiv preprint arXiv:2103.17239, 2021.Google Scholar
- Hassani A, Walton S, Shah N, Escaping the big data paradigm with compact transformers[J]. arXiv preprint arXiv:2104.05704, 2021.Google Scholar
- Chen C F, Fan Q, Panda R. Crossvit: Cross-attention multi-scale vision transformer for image classification[J]. arXiv preprint arXiv:2103.14899, 2021.Google Scholar
- Heo B, Yun S, Han D, Rethinking spatial dimensions of vision transformers[J]. arXiv preprint arXiv:2103.16302, 2021.Google Scholar
- Chu X, Tian Z, Wang Y, Twins: Revisiting the design of spatial attention in vision transformers[J]. arXiv preprint arXiv:2104.13840, 2021, 1(2): 3.Google Scholar
- Caron M, Touvron H, Misra I, Emerging properties in self-supervised vision transformers[J]. arXiv preprint arXiv:2104.14294, 2021.Google Scholar
- Touvron H, Cord M, Douze M, Training data-efficient image transformers & distillation through attention[C]//International Conference on Machine Learning. PMLR, 2021: 10347-10357.Google Scholar
- Wu H, Xiao B, Codella N, Cvt: Introducing convolutions to vision transformers[J]. arXiv preprint arXiv:2103.15808, 2021.Google Scholar
- Tolstikhin I, Houlsby N, Kolesnikov A, Mlp-mixer: An all-mlp architecture for vision[J]. arXiv preprint arXiv:2105.01601, 2021.Google Scholar
Index Terms
- Traffic Sign Recognition with Vision Transformers
Recommendations
Traffic Sign Recognition Based on Deep Learning Technique
ICISDM '22: Proceedings of the 6th International Conference on Information System and Data MiningTraffic sign recognition plays a significant role in intelligent transportation system. Therefore, in this paper, I propose a traffic sign recognition algorithm based on Convolutional Neural Network (CNN). The dataset collected to train and test in ...
Deep Learning Approach for U.S. Traffic Sign Recognition
ICDLT '19: Proceedings of the 2019 3rd International Conference on Deep Learning TechnologiesAdvanced Driver Assistant Systems (ADAS) have seen massive improvements in recent years; from detecting pedestrians, road lanes, traffic signs and signals, and vehicles to recognizing and tracking traffic signs. Traffic sign recognition systems are used ...
Traffic sign detection and recognition using fully convolutional network guided proposals
Detecting and recognizing traffic signs is a hot topic in the field of computer vision with lots of applications, e.g., safe driving, path planning, robot navigation etc. We propose a novel framework with two deep learning components including fully ...
Comments