Academic Journal of Computing & Information Science, 2022, 5(5); doi: 10.25236/AJCIS.2022.050501.
Yichi Zhang1, Boyu Hu2, Xinyi Yuan3, Yupeng Li4
1University of Nottingham, Ningbo, Zhejiang, China
2Beijing Information Science and Technology University, Beijing, China
3Xihua University, Chengdu, Sichuan, China
4The Woodlands, Mississauga, Ontario, Canada
Based on the emergence and development of autonomous driving technology, the identification of obstacles on the road is a very important and challenging task. And there are many difficulties in the realization of this task, for example, there are many types of road targets, and the scale span of the road obstacles is large. In view of these problems, this experiment proposes three improvement directions for the YOLOv3 algorithm to perform the task of road target prediction: one is to improve the up-sampling multiple to use more shallow spatial information to improve the accuracy of small target detection. The second is to change the way of feature fusion of the feature pyramid. Thirdly, the convergence direction of the model is changed by clustering learning. Experiments on the BDD100K data set show that the yolov3_10cls_tiny algorithm proposed in this paper has the best detection performance and better average accuracy than YOLOv3.
Object recognition, YOLOv3, Clustering learning
Yichi Zhang, Boyu Hu, Xinyi Yuan, Yupeng Li. Road object recognition method based on improved YOLOv3. Academic Journal of Computing & Information Science (2022), Vol. 5, Issue 5: 1-9. https://doi.org/10.25236/AJCIS.2022.050501.
[1] Li, F. (2016) Talks About the Development and Future of Automatic Driving Technology. Heilongjiang Science and Technology Information, 16, 59.
[2] Redmon, J., Divvala, S., Girshick, R. and Farhadi, A. (2016) You Only Look Once: Unified, Real-Time Object Detection. In Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (pp. 779-788). IEEE.
[3] Xin, Y. (2014) Research on Methods in Dynamic Obstacles Detection, Prediction and Avoidance of Autonomous Vehicles [Unpublished doctoral dissertation]. University of Science and Technology of China.
[4] Girshick, R.B. (2015) Fast R-CNN. arXiv. https://arxiv.org/abs/1504.08083.
[5] Redmon, J. and Farhadi, A. (2016) YOLO9000: Better, Faster, Stronger. arXiv. https://arxiv.org/ abs/1612.08242.
[6] Redmon, J. and Farhadi, A. (2018) YOLOv3: An Incremental Improvement. arXiv. https://arxiv.org /abs/1804.02767.
[7] Chen, K., Zhu, Z., Deng, X., Ma, C. and Wang, H. (2021) Overview of Deep Learning Research on Multi-scale Target Detection. Journal of Software, 32(4), 1201-1227.
[8] Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2013). Rich feature hierarchies for accurate object detection and semantic segmentation. IEEE Computer Society.
[9] Girshick, R. (2015). Fast R-CNN. Computer Science.
[10] REN, S., HE, K., & GIRSHICK, R. (2017). Faster R⁃CNN: towards real⁃time object detection with region proposal networks. IEEE transactions on pattern analysis and machine intelligence.
[11] Uijlings, Rr, J., Sande, V. D., Ea, K., Gevers, & Smeulders, et al.(2013) Selective search for object recognition. International Journal of Computer Vision, 104(2):154-171.
[12] Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: unified, real-time object detection. IEEE.
[13] Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., & Fu, C. Y., et al. (2016). Ssd: single shot multibox detector. Springer, Cham.
[14] Redmon, J., & Farhadi, A. (2018). Yolov3: an incremental improvement. arXiv e-prints.
[15] Zou, H. and Hou, J., 2021. Research on road small target detection based on improved SSD algorithm. Computer Engineering.
[16] Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv.
[17] Luo, J., Huang, J. and Bai, X., 2022. Improved YOLOv3 Road Small Target Detection Method. Mini-Micro Systems, 43(03), pp.449-455.
[18] Liu, Y., Zhang, G., Xu, Q. and Zhang, Y., 2021. Vehicle Detection Method Based on RF-YOLOV3 Algorithm. Modern Electric Technique, 44(13), pp.153-158.
[19] Redmon, J., & Farhadi, A. (2017). YOLO9000: better, faster, stronger. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 7263-7271).
[20] Rezatofighi, H., Tsoi, N., Gwak, J., Sadeghian, A., Reid, I., & Savarese, S. (2019). Generalized intersection over union: A metric and a loss for bounding box regression. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 658-666).
[21] Yu, F., Chen, H., Wang, X., Xian, W., Chen, Y., Liu, F., ... & Darrell, T. (2020). Bdd100k: A diverse driving dataset for heterogeneous multitask learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 2636-2645).