Abstract
Aiming at fast moving targets, such as ships, high-speed vehicles and athletes, this paper discusses a series of target detection algorithms based on neural network, YOLOv3 and background modeling. Compared KCF tracking with SSD tracking, Gaussian filter was applied to remove noise from pictures, and edge preserving filter was used to preserve edge features. Moreover, the algorithm combining deepsort tracking algorithm with YOLOv3 detection algorithm can improve the accuracy of YOLOv3 target detection, solve the problem of target loss during target tracking, adjust the frame size in real time, and improve the fit with the target position. Experiments show that the proposed algorithm based on detection before tracking has strong learning ability and robustness to unknown environment.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Wang, C.: Overview of moving object detection methods in intelligent video surveillance system. Autom. Instrum. 03, 1–3 (2017)
He, Y.: Moving target detection based on improved frame difference and background subtraction. Yanshan University (2016)
Zhang, J., Xu, C., Tang, M., Lu, W., Bian, Z.: Research on improved target detection method based on SSD [J]. Laser and Infrared, 2019, 49(08):1019-1025
Redmon, J., Farhadi, A.: YOLOv3: An Incremental Improvement (2018)
Jiang, X., Xiang, C., Liang, D.: An improved Mean Shift moving target tracking algorithm. Inf. Technol. 01, 127–130 (2017)
Henriques, J.F., Caseiro, R., Martins, P., et al.: High-speed tracking with kernelized correlation filters. IEEE Trans. Pattern Anal. Mach. Intell. 37(3), 583–596 (2015)
Bewley, A., Ge, Z., Ott, L., et al.: Simple Online and Realtime Tracking (2016)
Wojke, N., Bewley, A., Paulus, D.: Simple online and realtime tracking with a deep association metric. In: IEEE International Conference on Image Processing (ICIP). IEEE (2017)
Liu, C., Song, P., Di, K.: Multi-target tracking based on deep learning and information fusion. Ind. Control Comput. 32(10), 108–109+112 (2019)
Josefina, T., et al.: Do children with overweight respond faster to food-related words? Appetite 161, 105134 (prepublish) (2021)
Bisht, M., Gupta, R.: Offline handwritten Devanagari modified character recognition using convolutional neural network. Sādhanā 46(1), 20 (2021)
Cai, J., et al.: To explore the changes and differences of microstructure of vocal fold in vocal fold paralysis and cricoarytenoid joint dislocation by diffusion tensor imaging. J. Voice (prepublish) (2020)
Fonseca, E., Santos, J.F., Paisana, F., DaSilva, L.A.: Radio access technology characterisation through object detection. Comput. Commun. 168, 12–19 (prepublish) (2020)
Fan, L., Zhang, T., Du, W.: Optical-flow-based framework to boost video object detection performance with object enhancement. Expert Syst. Appl. 170, (prepublish) (2020)
Ding, L., Xu, X., Cao, Y., Zhai, G., Yang, F., Qian, L.: Detection and tracking of infrared small target by jointly using SSD and pipeline filter. Digital Sig. Process. 110, 102949 (prepublish) (2020)
Zhao, D., Yuan, B., Shi, Z., Jiang, Z.: Selective focus saliency model driven by object class‐awareness. IET Image Process. 15(6) (2020)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Wang, Y., Liang, Z., Cheng, X. (2021). Fast Target Tracking Based on Improved Deep Sort and YOLOv3 Fusion Algorithm. In: Zeng, J., Qin, P., Jing, W., Song, X., Lu, Z. (eds) Data Science. ICPCSEE 2021. Communications in Computer and Information Science, vol 1451. Springer, Singapore. https://doi.org/10.1007/978-981-16-5940-9_27
Download citation
DOI: https://doi.org/10.1007/978-981-16-5940-9_27
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-5939-3
Online ISBN: 978-981-16-5940-9
eBook Packages: Computer ScienceComputer Science (R0)