基于时间信息加权的ViSAR目标阴影跟踪算法

doi:10.12305/j.issn.1001-506X.2023.10.09

摘要/Abstract

摘要：

基于目标阴影的跟踪技术是视频合成孔径雷达(video synthetic aperture radar, ViSAR)目标探测的重要手段, 但ViSAR数据存在目标特征不明显且随时间不规则变化、相干斑噪声干扰强等问题, 使得ViSAR目标阴影跟踪精度较低。为此, 提出了一种鲁棒的基于时间信息加权的ViSAR目标阴影跟踪算法。针对目标特征不明显且随时间不规则变化的问题, 将尺度自适应均值偏移(adaptive scale mean shift, ASMS)跟踪算法引入到ViSAR目标阴影跟踪中, 同时在ASMS算法的背景比例加权(background ratio weighted, BRW)技术中添加历史帧的特征, 并对尺度正则项进行时间信息加权修正, 来对目标特征进行整合。针对相干斑噪声干扰强的问题, 对ASMS算法加入局部中值滤波操作的预处理步骤, 在不增加计算量的同时平滑了噪声。在对两类ViSAR数据集上不同尺寸、不同运动状态的目标阴影的跟踪实验结果表明, 与现有高性能跟踪算法相比, 所提算法在保证了实时性的基础上提高了跟踪精度, 且不需要额外的训练样本, 具备较好的工程应用价值。

关键词: 视频合成孔径雷达, 目标跟踪, 尺度自适应均值偏移, 时间信息加权, 中值滤波

Abstract:

Target shadow-based tracking technology is an important method for target detection in video synthetic aperture radar (ViSAR). However, there are some problems in ViSAR data, e.g., unclear target characteristics, irregular change of target characteristics with time, strong speckle noise interference, etc., which makes ViSAR target shadow tracking less accurate. Therefore, a robust ViSAR target shadow tracking algorithm based on time information weighting is proposed in this paper. Aiming at the problem that the target characteristics are not obvious and change irregularly with time, the adaptive scale mean shift (ASMS) tracking algorithm is introduced to ViSAR target shadow tracking. Meanwhile, the characteristics of historical frames are added to the background ratio weighted (BRW) technology of the ASMS algorithm. Target features are integrated with the time information weighted correction to the scale regularity term. Aiming at the strong interference of speckle noise, the preprocessing step of local median filtering operation is added to the ASMS algorithm to smooth the noise without increasing the amount of calculation. The experimental results of tracking target shadows with different sizes and different motion states on two kinds of ViSAR dataset show that, compared with the existing high-performance tracking algorithms, the proposed algorithm improves the tracking accuracy while ensuring real-time performance, and does not require additional training samples, which has good engineering application value.

Key words: video synthetic aperture radar (ViSAR), object tracking, adaptive scale mean shift (ASMS), time information weighting, median filtering

中图分类号:

TP753

黄艳堃, 邓玉辉, 孙光才, 邢孟道. 基于时间信息加权的ViSAR目标阴影跟踪算法[J]. 系统工程与电子技术, 2023, 45(10): 3065-3075.

Yankun HUANG, Yuhui DENG, Guangcai SUN, Mengdao XING. Shadow tracking algorithm for ViSAR target based on time information weighting[J]. Systems Engineering and Electronics, 2023, 45(10): 3065-3075.

图/表 17

图1

图2

图3

图4

图5

图6

表1

图7

图8

表2

图9

图10

图11

表3

图12

图13

表4

参考文献 30

1	WELLS L, SORENSEN K, DOERRY A, et al. Developments in SAR and IFSAR systems and technologies at Sandia National Laboratories[C]//Proc. of the 2003 IEEE Aerospace Conference Proceedings, 2003: 1085-1095.
2	DAMINI A, BALAJI B, PARRY C, et al. A videoSAR mode for the X-band wideband experimental airborne radar[C]. Proc. of the International Society for Optical Engineering, 2010: 135-145.
3	SONG X S , YU W D . Processing Video-SAR data with the fast backprojection method[J]. IEEE Trans.on Aerospace and Electronic Systems, 2016, 52 (6): 2838- 2848. doi: 10.1109/TAES.2016.150581
4	PALM S, WAHLEN A, STANKO S, et al. Real-time onboard processing and ground based monitoring of FMCW-SAR videos[C]// Proc. of the EUSAR 10th European Conference on Synthetic Aperture Radar, 2014.
5	MILLER J, BISHOP E, DOERRY A. An application of backprojection for video SAR image formation exploiting a subaperature circular shift register[C]//Proc. of the Algorithms for Synthetic Aperture Radar Imagery XX, 2013: 66-79.
6	PU W , WANG X D , WU J J , et al. Video SAR imaging based on low-rank tensor recovery[J]. IEEE Trans.on Neural Networks and Learning Systems, 2021, 32 (1): 188- 202. doi: 10.1109/TNNLS.2020.2978017
7	梁健, 张润宁, 包敏凤. 天基视频SAR系统设计及成像算法研究[J]. 中国空间科学技术, 2016, 36 (6): 22.
	LIANG J , ZHANG R N , BAO M F , et al. Research on spaceborne video SAR system design and image formation algorithm[J]. Chinese Space Science and Technology, 2016, 36 (6): 22.
8	胡睿智. 视频合成孔径雷达成像理论与关键技术研究[D]. 成都电子科技大学, 2018.
	HU R Z. Research on imaging theory and key technology of video synthetic aperture radar[D]. Chengdu: University of Electronic Science and Technology of China, 2018.
9	苟立婷, 李勇, 朱岱寅, 等. 基于GPU的圆迹视频SAR实时成像算法[J]. 雷达科学与技术, 2019, 17 (5): 550- 556.
	GOU L T , LI Y , ZHU D Y , et al. A real-time algorithm for circular video SAR imaging based on GPU[J]. Radar Science and Technology, 2019, 17 (5): 550- 556.
10	WANG H , CHEN Z S , ZHENG S C . Preliminary research of low-RCS moving target detection based on Ka-band video SAR[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14 (6): 811- 815. doi: 10.1109/LGRS.2017.2679755
11	ZHANG Y, MAO X H, YAN H, et al. A novel approach to moving targets shadow detection in VideoSAR imagery sequence[C]//Proc. of the IEEE International Geoscience and Remote Sensing Symposium, 2017: 606-609.
12	ZHANG Y, YANG S, LI H, et al. Shadow tracking of moving target based on CNN for video SAR system[C]//Proc. of the IGARSS IEEE International Geoscience and Remote Sensing Symposium, 2018: 4399-4402.
13	LIANG Z H, LIANG C J, ZHANG Y, et al. Tracking of moving target based on SiamMask for video SAR system[C]//Proc. of the IEEE International Conference on Signal, Information and Data Processing, 2019.
14	吴晓桐. 视频SAR的运动目标检测与跟踪技术研究[D]. 西安: 西安电子科技大学, 2020.
	WU X T. A study of moving target detection and tracking techniques with video SAR[D]. Xi'an: Xidian University, 2020.
15	丁金闪. 视频SAR成像与动目标阴影检测技术[J]. 雷达学报, 2020, 9 (2): 321- 334.
	DING J S . Focusing algorithms and moving target detection based on video SAR[J]. Journal of Radars, 2020, 9 (2): 321- 334.
16	闫贺, 黄佳, 李睿安, 等. 基于改进快速区域卷积神经网络的视频SAR运动目标检测算法研究[J]. 电子与信息学报, 2021, 43 (3): 615- 622.
	YAN H , HUANG J , LI R A , et al. Research on video SAR moving target detection algorithm based on improved faster region-based CNN[J]. Journal of Electronics & Information Technology, 2021, 43 (3): 615- 622.
17	ZHAO B J , HAN Y Q , WANG H S , et al. Robust shadow tracking for video SAR[J]. IEEE Geoscience and Remote Sensing Letters, 2020, 18 (5): 821- 825.
18	VOJIR T , NOSKOVA J , MATAS J . Robust scale-adaptive mean-shift for tracking[J]. Pattern Recognition Letters, 2014, 49, 250- 258. doi: 10.1016/j.patrec.2014.03.025
19	张惊雷, 时鹏, 温显斌. 基于尺度自适应均值偏移优化的TLD跟踪算法[J]. 控制与决策, 2019, 34 (1): 144- 150.
	ZHANG J L , SHI P , WEN X B . A TLD tracking algorithm based on scale-adaptive mean-shift method[J]. Control and Decision, 2019, 34 (1): 144- 150.
20	吕晨, 程德强, 寇旗旗, 等. 基于YOLOv3和ASMS的目标跟踪算法[J]. 光电工程, 2021, 48 (2): 200175.
	LYU C , CHENG D Q , KOU Q Q , et al. Target tracking algorithm based on YOLOv3 and ASMS[J]. Opto-Electronics Engineering, 2021, 48 (2): 200175.
21	OLIVER C , QUEGAN S . Understanding synthetic aperture radar images[M]. Raleigh: SciTech Publishing, 2004.
22	WU Y, LIM J, YANG M H. Online object tracking: A benchmark[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2013: 2411-2418.
23	DANELLJAN M, KHAN F S, FELSBERG M, et al. Adaptive color attributes for real-time visual tracking[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2014: 1090-1097.
24	DANELLJAN M, HAGER G, KHAN F, et al. Accurate scale estimation for robust visual tracking[C]//Proc. of the British Machine Vision Conference, 2014.
25	ZHU G B, WANG J Q, WU Y, et al. Collaborative correlation tracking[C]//Proc. of the British Machine Vision Conference, 2015: 184.
26	HENRIQUES J F , CASEIRO R , MARTINS P , et al. High-speed tracking with kernelized correlation filters[J]. IEEE Trans.on Pattern Analysis and Machine Intelligence, 2015, 37 (3): 583- 596.
27	LI Y, ZHU J K. A scale adaptive kernel correlation filter tracker with feature integration[C]//Proc. of the European Conference on Computer Vision, 2014: 254-265.
28	GALOOGAHI H K, FAGG A, LUCEY S. Learning background-aware correlation filters for visual tracking[C]//Proc. of the IEEE International Conference on Computer Vision, 2017: 1135-1143.
29	WANG C, ZHANG L, XIE L H, et al. Kernel cross-correlator[C]// Proc. of the AAAI Conference on Artificial Intelligence, 2018.
30	邓玉辉, 孙光才, 邢孟道, 等. 基于面向对象的视频SAR成像方法[C]//第十五届全国雷达学术年会, 2020.
	DENG Y H, SUN G C, XING M D, et al. Object oriented ViSAR imaging method[C]//Proc. of the the 15th National Radar Academic Annual Conference, 2020.

跟踪器	本文方法	BACF	CCT	CN	DSST	KCC	KCF	SAMF	ASMS
FPS	108.00	46.00	57.00	318.00	108.00	13.00	58.00	3.00	148.00
平均CLE	2.50	86.27	79.82	79.08	80.99	72.56	93.46	89.36	3.63

跟踪器	本文方法	BACF	CCT	CN	DSST	KCC	KCF	SAMF	ASMS
FPS	38.00	16.00	23.00	31.00	8.00	5.00	21.00	1.00	20.00
平均CLE	8.34	23.82	68.45	309.78	344.27	33.82	53.60	29.44	17.34

跟踪器	本文方法	BACF	CCT	CN	DSST	KCC	KCF	SAMF	ASMS
FPS	85.00	20.00	19.00	102.00	36.00	5.00	20.00	1.00	87.00
平均CLE	3.10	411.74	368.03	453.71	380.60	335.30	292.06	180.76	4.17

跟踪器	本文方法	BACF	CCT	CN	DSST	KCC	KCF	SAMF	ASMS
FPS	101.00	16.00	13.00	234.00	50.00	3.00	17.00	1.00	159.00
平均CLE	3.53	262.03	224.45	260.02	255.34	211.12	243.76	136.07	80.25

[1]	毕文豪, 周杰, 张安, 刘力. 杂波环境下基于最大熵模糊聚类的JPDA算法[J]. 系统工程与电子技术, 2023, 45(7): 1920-1927.
[2]	王森, 鲍庆龙, 潘嘉蒙, 祝茜. 基于改进概率假设密度滤波器的非合作双基地雷达目标跟踪[J]. 系统工程与电子技术, 2023, 45(7): 2002-2009.
[3]	舒刊, 万显荣, 易建新, 谢德强, 胡越洋, 童云. 基于状态解耦的两坐标外源雷达3D目标跟踪[J]. 系统工程与电子技术, 2023, 45(6): 1658-1666.
[4]	刘政玮, 陈映, 鲁耀兵. 适用于多目标轨迹小角度交叉的PHD滤波器[J]. 系统工程与电子技术, 2023, 45(4): 982-990.
[5]	张毅, 于浩, 杨秀霞, 姜子劼. 无人机集群分组编队控制跟踪一体化设计[J]. 系统工程与电子技术, 2023, 45(3): 848-858.
[6]	张昀普, 单甘霖, 黄燕, 付强. 考虑盲区的多移动传感器地面目标检测跟踪调度方法[J]. 系统工程与电子技术, 2023, 45(2): 453-464.
[7]	安雷, 李召瑞, 吉兵. 杂波环境下可移动主被动传感器长时调度方法[J]. 系统工程与电子技术, 2023, 45(1): 165-174.
[8]	仇祝令, 查宇飞, 李振宇, 李禹铭, 张鹏, 朱川. 基于多模型蒸馏的时间正则化相关滤波跟踪算法[J]. 系统工程与电子技术, 2022, 44(8): 2448-2456.
[9]	侯子林, 程婷, 彭瀚. 基于量测转换序贯滤波的GMPHD机动目标跟踪[J]. 系统工程与电子技术, 2022, 44(8): 2474-2482.
[10]	史浩然, 卢发兴, 祁江鑫, 杨光. 基于辅助信标的无人机协同目标跟踪[J]. 系统工程与电子技术, 2022, 44(7): 2302-2310.
[11]	金国栋, 薛远亮, 谭力宁, 许剑锟. 基于孪生神经网络的目标跟踪算法进展研究[J]. 系统工程与电子技术, 2022, 44(6): 1805-1822.
[12]	翟光, 王妍欣, 孙一勇. 基于低轨星网的多目标协同跟踪滤波技术[J]. 系统工程与电子技术, 2022, 44(6): 1957-1967.
[13]	王帅, 向建军, 彭芳, 唐书娟. 基于新最速下降法的目标跟踪算法[J]. 系统工程与电子技术, 2022, 44(5): 1512-1519.
[14]	辛怀声, 曹晨. 基于交互多模型的分组δ-广义标签多伯努利算法[J]. 系统工程与电子技术, 2022, 44(4): 1128-1138.
[15]	谢家豪, 黄树彩, 韦道知, 张曌宇, 王文豪. 基于P_EV准则的不确定混合多传感器联盟求解[J]. 系统工程与电子技术, 2022, 44(3): 819-826.