A Fast Target Tracking Method Based on BACF

doi:10.12382/bgxb.2022.0708

Abstract

Abstract:

Image tracking algorithm is the key technology of imaging terminal guidance of missile and optoelectronic reconnaissance equipment. Its high efficiency and low latency extremely improve the accurate and stable tracking performance. In order to realize the high efficiency of imaging terminal guidance tracking algorithm of missile limited by volume, power, and so on, a data-driven multi-task allocation strategy is proposed based on the background-aware correlation filters (BACF)-based tracking algorithm. And a Cache-optimized fast Fourier transform (FFT) efficient method is proposed, considering that the correlation filtering calculation is key path to reduce the real-time capability of algorithm. The real-time capability of multi-core processing is improved by about 56.3% by mapping the algorithm to multi-core DSP processor platform for parallel processing. It is verified by UAV flight tests that the optimized method can meet the requirements of target tracking real-time capability under space constraints, which provides a strong technical support for engineering applications of smart munitions.

Key words: target tracking algorithm, background-aware correlation filter, task allocation, fast Fourier transform

CLC Number:

MA Xiao, LI Xinqi, LIU Zhenyuan, DI Chao. A Fast Target Tracking Method Based on BACF[J]. Acta Armamentarii, 2024, 45(2): 497-503.

Figures/Tables 7

Fig.1 Implementation flow chart of BACF tracking algorithm

Fig.2 Simulated results based on BACF target tracking algorithm

Fig.3 Block diagram of multi-task allocation of target tracking algorithm

Fig.4 Hardware diagram of tracking system

Table 1 Calculation time of FFT

点数	计算时间/μs
128	0.825
256	1.490
512	3.525
1024	6.862
2048	16.35
4096	27.243
8192	34.845

Fig.5 Block diagram of multi-threaded computing of multi-core DSP

Fig.6 Verification of tracking algorithm

References 23

[1]	李大维, 莫波, 高可, 等. 针对弹载平台红外小目标的检测与跟踪算法[J]. 宇航学报, 2020, 41(11):1440-1448.
	LI D W, MO B, GAO K, et al. Detection and tracking algorithm for small infrared target on missile-borne platform[J]. Journal of Astronautics, 2020, 41(11):1440-1448. (in Chinese)
[2]	孟琭, 李诚新. 近年目标跟踪算法短评——相关滤波与深度学习[J]. 中国图象图形学报, 2019, 24(7):1011-1016.
	MENG L, LI C X. Brief review of object tracking algorithms in recent years: correlated filtering and deep learning[J]. Journal of Image and Graphics, 2019, 24(7):1011-1016. (in Chinese)
[3]	姜珊, 底晓强, 韩成. 融合时空特性的孪生网络视觉跟踪[J]. 兵工学报, 2021, 42(9):1940-1950. doi: 10.3969/j.issn.1000-1093.2021.09.015
	JIANG S, DI X Q, HAN C. Siamese network for visual tracking with temporal-spatial property[J]. Acta Armamentarii, 2021, 42(9):1940-1950. (in Chinese) doi: 10.3969/j.issn.1000-1093.2021.09.015
[4]	王红涛, 邓淼磊, 赵文君, 等. 基于深度学习的单目标跟踪算法综述[J]. 计算机系统应用, 2022, 31(5):40-51.
	WANG H T, DENG M L, ZHAO W J, et al. Survey on single object tracking algorithms based on deep learning[J]. Computer Systems & Applications, 2022, 31(5):40-51. (in Chinese)
[5]	孟晓燕, 段建民. 基于相关滤波的目标跟踪算法研究综述[J]. 北京工业大学学报, 2020, 46(12):1393-1416.
	MENG X Y, DUAN J M. Advances in correlation filter-based object tracking algorithms: a review[J]. Journal of Beijing University of Technology, 2020, 46(12):1393-1416. (in Chinese)
[6]	王少博, 张成, 苏迪, 等. 基于改进YOLOv3和核相关滤波算法的旋转弹目标探测算法[J]. 兵工学报, 2022, 43(5):1032-1045. doi: 10.12382/bgxb.2021.0283
	WANG S B, ZHANG C, SU D, et al. A target detecting algorithm for spinning projectile based on improved YOLOv3 and KCF[J]. Acta Armamentarii, 2022, 43(5):1032-1045. (in Chinese) doi: 10.12382/bgxb.2021.0283
[7]	LI B, WU W, WANG Q,et a1. SiamRPN++: evolution of Siamese visual tracking with very deep networks[C]// Proceedings of 2019 IEEE Conference on Computer Vision and Pattern Recognition. Long Beach, CA, US:IEEE, 2019: 4277-4286.
[8]	ZHANG Z P, PENG H W. Deeper and wider Siamese networks for real-time visual tracking[C]// Proceedings of 2019 IEEE Conferenee on Computer Vision and Pattern Recognition. Long Beach, CA, US:IEEE, 2019: 4586-4595.
[9]	ZHENG L Y, CHEN Y Y, TANG M, et al. Siamese deformable cross-correlation network for real-time visual tracking[J]. Neurocomputing, 2020, 401 (11): 36-47. doi: 10.1016/j.neucom.2020.02.080 URL
[10]	BOLME D S, BEVERIDGE J R, DRAPER B A, et al. Visual object tracking using adaptive correlation falters[C]//Proceedings of 2010 IEEE Conference on Computer Vision and Pauern Recognition. San Francisco, CA, US: IEEE, 2010: 2544-2550.
[11]	HENRIQUES J F, CASEIRO R, MARTINS P, et al. Exploiting the circulant structure of tracking-by-detection with kernels[C]//Proceedings of the 12th European Conference on Computer Vision. Florence, Italy:Springer, 2012, Part IV:702-715.
[12]	HENRIQUES J F, CASEIRO R, MARTINS P, et al. Highspeed tracking with kernelized correlation filters[J]. IEEE Transactionson Pattern Analysis and Machine Intelligence, 2015, 37(3): 583-596
[13]	刘珊, 杨卫, 邵星灵. 基于尺度自适应的背景感知目标跟踪算法[J]. 电光与控制, 2020, 27(9):19-23.
	LIU S, YANG W, SHAO X L. A background-aware target tracking algorithm based on scale adaptation[J]. Electronics Optics & Control, 2020, 27(9):19-23. (in Chinese)
[14]	GALOOGAHI H K, FAGG A, LUCEY S. Learning background-aware correlation filters for visual tracking[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision. Venice, Italy:IEEE, 2017:1144-1152.
[15]	马潇, 高立宁, 刘腾飞, 等. 基于Cache优化的大点数FFT在TS201上的实现[J]. 电子与信息学报, 2013(7): 1774-1778.
	MA X, GAO L N, LIU T F, et al. Cache-optimized implementation of long sequences FFT on TS201[J]. Journal of Electronics & Information Technology, 2013(7): 1774-1778. (in Chinese)
[16]	高立宁, 朱亮, 刘腾飞, 等. 基于超标量处理器的高效FFT映射方法[J]. 北京理工大学学报, 2016, 36(9):940-946.
	GAO L N, ZHU L, LIU T F, et al. An efficient FFT-mapping method based on superscalar processor[J]. Transactions of Beijing Institute of Technology, 2016, 9: 940-946. (in Chinese)
[17]	仇祝令, 查宇飞, 吴敏, 等. 基于注意力学习的正则化相关滤波跟踪算法[J]. 电子学报, 2020, 48(9):1762-1768. doi: 10.3969/j.issn.0372-2112.2020.09.014
	QIU Z L, ZHA Y F, WU M, et al. Learning attentional regularized correlation filter for visual tracking[J]. Journal Article, 2020, 48(9):1762-1768. (in Chinese)
[18]	陈丹, 姚伯羽. 运动模型引导的自适应核相关目标跟踪方法[J]. 电子学报, 2021, 49(3):550-558. doi: 10.12263/DZXB.20200433
	CHEN D, YAO B Y. Adaptive response kernel correlation target tracking method guided by motion model[J]. Acta Electronica Sinica, 2021, 49(3):550-558. (in Chinese) doi: 10.12263/DZXB.20200433
[19]	王凌, 吴楚格, 范文慧. 边缘计算资源分配与任务调度优化综述[J]. 系统仿真学报, 2021, 33(3):509-520. doi: 10.16182/j.issn1004731x.joss.20-0584
	WANG L, WU C G, FAN W H. A survey of edge computing resource allocation and task scheduling optimization[J]. Journal of System Simulation, 2021, 33(3):509-520. (in Chinese) doi: 10.16182/j.issn1004731x.joss.20-0584
[20]	刘嘉琦, 张亚文, 张瀚文, 等. 一种基于两级缓存的协同缓存机制[J]. 软件学报, 2021, 32(9):2963-2976.
	LIU J Q, ZHANG Y W, ZHANG H W, et al. Cache coordination scheme based on two-level Cache[J]. Journal of Software, 2021, 32(9):2963-2976. (in Chinese)
[21]	李浩, 谢伦国. 片上多处理器末级Cache优化技术研究[J]. 计算机研究与发展, 2012, 49(增刊1):172-179.
	LI H, XIE L G. Research development of optimization technology on last level Cache in chip multi-processors[J]. Journal of Computer Research and Development, 2012, 49(S1):172-179. (in Chinese)
[22]	王世一. 数字信号处理[M]. 北京: 北京理工大学出版社, 1997: 123-131.
	WANG S Y. Digital signal processing[M]. Beijing: Beijing Institute of Technology Press, 1997:123-131. (in Chinese)
[23]	Texas Instruments.TMS320C6678 multicore fixed and floating-point digital signal processor[EB/OL]. (2017-10-11). http://www.ti.com.cn/lsds/ti_zh/processors/dsp/c6000_dsp/c66x/overview.page.