Multi-scenario Vehicle Detection Based on the YOLOv5 Algorithm

Liang Zhou

doi:10.62051/mbgg8t65

Authors

Liang Zhou

DOI:

https://doi.org/10.62051/mbgg8t65

Keywords:

YOLOv5, vehicle detection, intelligent transportation system, real-time detection.

Abstract

This paper addresses the issues of increasing traffic volume leading to frequent traffic congestion and accidents by studying the effectiveness of occluded vehicle detection based on the YOLOv5 algorithm. The aim is to explore the performance of this algorithm in various occlusion scenarios. Firstly, we introduce the occluded vehicle images used for testing and describe different types of occlusions in detail. Then, the basic principles and construction process of the YOLOv5 algorithm are explained. Subsequently, we propose a vehicle detection and recognition model based on YOLOv5 and validate the model's effectiveness through convergence analysis of the training loss curve and accuracy analysis of the test set. Finally, experimental results demonstrate that the YOLOv5 algorithm can efficiently and accurately detect vehicles in complex scenarios and partially occluded conditions, showcasing its application potential in intelligent transportation systems.

Downloads

Download data is not yet available.

References

[1] Chen, Z., et al. (2021). A Review on Recent Advances in Vehicle Detection Algorithms. IEEE Access, 9, 23212-23229.

[2] Zhang, H., et al. (2020). A Survey on Recent Advances in Vehicle Detection from Remote Sensing Images. Remote Sensing, 12(10), 1644.

[3] Duan, K., et al. (2020). CenterTrack: Tracking Objects as Points. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 9416-9426).

[4] Barnich, O., & Van Droogenbroeck, M. (2011). ViBe: A Universal Background Subtraction Algorithm for Video Sequences. IEEE Transactions on Image Processing, 20(6), 1709-1724.

[5] Hofmann, M., Tiefenbacher, P., & Rigoll, G. (2012). Background Segmentation with Feedback: The Pixel-Based Adaptive Segmenter. In 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (pp. 97-104). IEEE.

[6] Zach, C., Pock, T., & Bischof, H. (2007). A Duality Based Approach for Realtime TV-L1 Optical Flow. In Proceedings of the Joint Pattern Recognition Symposium (pp. 214-223). Springer, Berlin, Heidelberg.

[7] Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). ImageNet Classification with Deep Convolutional Neural Networks. In Advances in Neural Information Processing Systems (NIPS) (pp. 1097-1105).

[8] Girshick, R. (2015). Fast R-CNN. In Proceedings of the IEEE International Conference on Computer Vision (ICCV) (pp. 1440-1448).

[9] John. (2021). YOLOV5-DeepSORT-Vehicle-Tracking-Master. Retrieved from [GitHub Repository URL]

[10] Redmon, Joseph and Farhadi.YOLOv3:An Incremental Improvement[D].arXiv.2018.