Optimized YOLOv4 Algorithm for Car Detection in Traffic Flow

Abstract

The vehicle detection accuracy and actual in images and videos appear to be very tough and critical duties in a key technology traffic system. Specifically, under convoluted traffic conditions. As a result, the presented study proposes single-stage deep neural networks YOLOv4-3L, YOLOv4-2L, YOLOv4-GB, and YOLOv3-GB. After optimizing the network structure by adding more layers in the right positions with the right amount of filters, the dataset will be repaired and the noise reduced before being sent to the mentoring. This research will be applied to YOLOv3 and YOLOv4. In this study the OA-Dataset is collect and used, the data set is manually labeled with the care of different weathers and scenarios, as well as for end-to-end training of the network. Around the same time, optimized YOLOv4 and YOLOv3 demonstrate a significant degree of accuracy with 99.68 % and precision of 91 %. The speed and detection accuracy of this algorithm are found to be higher than that of previous algorithms.

Keywords

• Car Detection
• The Traffic Flow
• YOLOv4
• Deep learning

References

1. [1] Y. Q. Huang, J. C. Zheng, S. D. Sun, C. F. Yang, and J. Liu, “Optimized YOLOv3 algorithm and its application in traffic flow detections,” Appl. Sci., vol. 10, no. 9, May 2020, doi: 10.3390/app10093079.
2. [2] Y. Xu, G. Yu, Y. Wang, X. Wu, and Y. Ma, “A hybrid vehicle detection method based on viola-jones and HOG + SVM from UAV images,” Sensors (Switzerland), vol. 16, no. 8, 2016, doi: 10.3390/s16081325.
3. [3] Q. J. Qiu, L. Yong, and D. W. Cai, “Vehicle detection based on LBP features of the Haar-like Characteristics,” Proc. World Congr. Intell. Control Autom., vol. 2015-March, no. March, pp. 1050–1055, 2015, doi: 10.1109/WCICA.2014.7052862.
4. [4] P. F. Felzenszwalb, R. B. Girshick, D. Mcallester, and D. Ramanan, “Object Detection With Partbase,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 32, no. 9, pp. 1627–1645, 2010.
5. [5] K. Mu, F. Hui, X. Zhao, and C. Prehofer, “Multiscale edge fusion for vehicle detection based on difference of Gaussian,” Optik (Stuttg)., vol. 127, no. 11, pp. 4794–4798, 2016, doi: 10.1016/j.ijleo.2016.01.017.
6. [6] K. S. Choi, J. S. Shin, J. J. Lee, Y. S. Kim, S. B. Kim, and C. W. Kim, “In vitro trans-differentiation of rat mesenchymal cells into insulin-producing cells by rat pancreatic extract,” Biochem. Biophys. Res. Commun., vol. 330, no. 4, pp. 1299–1305, 2005, doi: 10.1016/j.bbrc.2005.03.111.
7. [7] K. He, X. Zhang, S. Ren, and J. Sun, “Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 37, no. 9, pp. 1904–1916, 2015, doi: 10.1109/TPAMI.2015.2389824.
8. [8] R. Girshick, J. Donahue, T. Darrell, J. Malik, U. C. Berkeley, and J. Malik, “1043.0690,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 1, p. 5000, 2014, doi: 10.1109/CVPR.2014.81.

9. [9] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You Only Look Once: Unified, Real-Time Object Detection.” [Online]. Available: https://goo.gl/bEs6Cj.
10. [10] J. Redmon and A. Farhadi, “YOLO9000: Better, Faster, Stronger.” [Online]. Available: http://pjreddie.com/yolo9000/.
11. [11] J. Redmon and A. Farhadi, “YOLOv3: An Incremental Improvement.” [Online]. Available: https://pjreddie.com/yolo/.
12. [12] A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, “YOLOv4: Optimal Speed and Accuracy of Object Detection,” 2020, [Online]. Available: http://arxiv.org/abs/2004.10934.
13. [13] V. Thakar, H. Saini, W. Ahmed, M. M. Soltani, A. Aly, and J. Y. Yu, “Efficient Single-Shot Multibox Detector for Construction Site Monitoring,” 2018 IEEE Int. Smart Cities Conf. ISC2 2018, no. 1, p. 77, 2019, doi: 10.1109/ISC2.2018.8656929.
14. [14] J. Liu, Y. Huang, J. Peng, J. Yao, and L. Wang, “Fast Object Detection at Constrained Energy,” IEEE Trans. Emerg. Top. Comput., vol. 6, no. 3, pp. 409–416, 2018, doi: 10.1109/TETC.2016.2577538.
15. [15] S. Ren, K. He, and R. Girshick, “Faster R-CNN : Towards Real-Time Object Detection with Region Proposal Networks,” pp. 1–9.
16. [16] F. B. Tesema, J. Lin, J. Ou, H. Wu, and W. Zhu, “Feature Fusing of Feature Pyramid Network for Multi-Scale Pedestrian Detection,” 2018 15th Int. Comput. Conf. Wavelet Act. Media Technol. Inf. Process. ICCWAMTIP 2018, no. 1, pp. 10–13, 2019, doi: 10.1109/ICCWAMTIP.2018.8632614.
17. [17] V. Sangeetha and K. J. R. Prasad, “Syntheses of novel derivatives of 2-acetylfuro[2,3-a]carbazoles, benzo[1,2-b]-1,4-thiazepino[2,3-a]carbazoles and 1-acetyloxycarbazole-2- carbaldehydes,” Indian J. Chem. - Sect. B Org. Med. Chem., vol. 45, no. 8, pp. 1951–1954, 2006, doi: 10.1002/chin.200650130.
18. [18] Y. Liu, “Big Data Technology and Its Analysis of Application in Urban Intelligent Transportation System,” Proc. - 3rd Int. Conf. Intell. Transp. Big Data Smart City, ICITBS 2018, vol. 2018-Janua, pp. 17–19, 2018, doi: 10.1109/ICITBS.2018.00012.
19. [19] E. Toropov, L. Gui, S. Zhang, S. Kottur, and J. M. F. Moura, “TRAFFIC FLOW FROM A LOW FRAME RATE CITY CAMERA Electrical and Computer Engineering Pittsburgh , PA , USA Instituto Superior Técnico Instituto de Sistemas e Robótica Lisbon , Portugal,” Int. Conf. Image Process., pp. 3802–3806, 2015.
20. [20] “Gaussian filtering • Significant values,” pp. 18–32, 2010.