Author:
Shivayogi Ananya Belagodu,Matasagara Dharmendra Nehal Chakravarthy,Ramakrishna Anala Maddur,Subramanya Kolala Nagaraju
Abstract
Traffic Sign Recognition (TSR) is one of the most sought-after topics in computer vision, mostly due to the increasing scope and advancements in self-driving cars. In our study, we attempt to implement a TSR system that helps a driver stay alert during driving by providing information about the various traffic signs encountered. We will be looking at a working model that classifies the traffic signs and gives output in the form of an audio message. Our study will be focused on traffic sign detection and recognition on Indian roads. A dataset of Indian road traffic signs was created, based upon which our deep learning model will work. The developed model was deployed on NVIDIA Jetson Nano using YOLOv4 architecture, giving an accuracy in the range of 54.68–76.55% on YOLOv4 architecture. The YOLOv4-Tiny model with DeepStream implementation achieved an FPS of 32.5, which is on par with real-time detection requirements.
Publisher
Universiti Putra Malaysia
Subject
General Earth and Planetary Sciences,General Environmental Science
Reference16 articles.
1. Bochkovskiy, A., Wang, C. Y., & Liao, H. Y. M. (2020). YOLOv4: Optimal speed and accuracy of object detection. arXiv Preprint. http://arxiv.org/abs/2004.10934
2. Do, H. N, Vo, M. T., Luong, H. Q., Nguyen, A. H., Trang, K., & Vu, L. T. K. (2017). Speed limit traffic sign detection and recognition based on support vector machines. In 2017 International Conference on Advanced Technologies for Communications (ATC) (pp. 274-278). IEEE Publishing. https://doi.org/10.1109/ATC.2017.8167633
3. Ellahyani, A., El Ansari, M., El Jaafari, I., & Charfi, S. (2016). Traffic sign detection and recognition using features combination and random forests. International Journal of Advanced Computer Science and Applications, 7(1), 686-693. https://doi.org/10.14569/IJACSA.2016.070193
4. Hasegawa, R., Iwamoto, Y., & Chen, Y. W. (2019). Robust detection and recognition of japanese traffic sign in the complex scenes based on deep learning. In 2019 IEEE 8th Global Conference on Consumer Electronics (GCCE) (pp. 575-578). IEEE Publishing. https://doi.org/10.1109/GCCE46687.2019.9015419
5. Hegadi, R. S. (2011). Automatic traffic sign recognition. In Proceedings of International Conference on Communication, Computation, Management & Nanotechnology (pp. 1-5). REC Bhalki.
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献