Simulator with Computer Vision for Detection, Tracking, and Distance Calculation of Moving Objects
DOI:
https://doi.org/10.31637/epsir-2024-812Keywords:
Artificial Vision, Collision Detection, Deep Learning, Artificial Intelligence, Road Safety, You Only Look Once, Object Tracking, Traffic SimulatorAbstract
Introduction: In the framework of a research on computer vision systems for motorcycle collision prevention, a digital simulator has been developed that evaluates relevant traffic scenarios. Methodology: The simulator analyzes synthetic video sequences of various traffic environments using computer vision models. It uses the YOLO algorithm, known for its speed and accuracy in object detection, to identify, classify and track vehicles, pedestrians and other moving objects. Results: The system is able to estimate Euclidean distance and project the trajectory of items from the rider's perspective, replicating what would be captured by a vision system on a real motorcycle. The adaptability of YOLO allows its use in multiple contexts without the need for intensive retraining. Discussion: The simulator provides a controlled environment to evaluate the performance of collision detection algorithms in critical scenarios, allowing repeatable testing without real risks. Conclusions: This simulator facilitates the validation of collision avoidance algorithms, providing a safe and efficient environment to test their performance in critical traffic situations.
Downloads
References
APEX Simulación y Tecnología. (18 de agosto de 2020). Simulador de motocicleta – Software [Archivo de Video]. YouTube. https://www.youtube.com/watch?v=MyaN1J5q6V8&t=1s
Bharadwaj, R., Gajbhiye, P., Rathi, A., Sonawane, A. y Uplenchwar, R. (2023). Lane, car, traffic sign and collision detection in simulated environment using GTA-V. En J. S. Raj, I. Perikos y V. E. Balas (Eds.), Intelligent sustainable systems. ICoISS 2023. Lecture notes in networks and systems (Vol. 665, pp. 465-476). Springer. https://doi.org/10.1007/978-981-99-1726-6_36 DOI: https://doi.org/10.1007/978-981-99-1726-6_36
Garg, R., Wadhwa, N., Ansari, S. y Barron, J. T. (2019). Learning single camera depth estimation using dual-pixels. En Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) (pp. 7628-7637). https://acortar.link/09sWYo DOI: https://doi.org/10.1109/ICCV.2019.00772
Lee, J. H. y Kim, C. S. (2019). Monocular depth estimation using relative depth maps. En Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 9729-9738). https://bit.ly/4cSqPyZ DOI: https://doi.org/10.1109/CVPR.2019.00996
Makris, S. y Aivaliotis, P. (2022). AI-based vision system for collision detection in HRC applications. Procedia CIRP, 104, 1-6. https://doi.org/10.1016/j.procir.2022.01.021 DOI: https://doi.org/10.1016/j.procir.2022.02.171
Müller, U., Ben, J., Cosatto, E., Flepp, B. y Cun, Y. L. (2018). Off-road obstacle avoidance through end-to-end learning. En Advances in Neural Information Processing Systems (pp. 4278-4287). https://acortar.link/hyy9hd
Redmon, J. y Farhadi, A. (2018). YOLOv3: an incremental improvement. arXiv preprint arXiv:1804.02767. https://arxiv.org/abs/1804.02767
Redmon, J., Divvala, M., Girshick, R. y Farhadi, A. (2016). You only look once: Unified, real-time object detection. En Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 1-9). https://doi.org/10.1109/CVPR.2016.91 DOI: https://doi.org/10.1109/CVPR.2016.91
Ros, G., Sellart, L., Materzynska, J., Vázquez, D. y López, A. M. (2016). The SYNTHIA dataset: A large collection of synthetic images for semantic segmentation of urban scenes. En Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 3234-3243). https://acortar.link/4Ym1Rw DOI: https://doi.org/10.1109/CVPR.2016.352
Shim, K., Kim, J., Lee, G. y Shim, B. (2023). Depth-relative self attention for monocular depth estimation. En Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence. IJCAI-23 (pp. 1395-1403). https://www.ijcai.org/proceedings/2023/0155.pdf DOI: https://doi.org/10.24963/ijcai.2023/155
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 Leonardo Valderrama García
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Authors who publish with this journal agree to the following terms:- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under Creative Commons Non Commercial, No Derivatives Attribution 4.0. International (CC BY-NC-ND 4.0.), that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
