EOG-DS：Zoom Dual-Mode Tracking Algorithm for Dynamic Targets

doi:10.3778/j.issn.1002-8331.2203-0057

Abstract

Abstract: Dual-mode fusion tracking with high magnification optical system is widely used in ground air surveillance and forensics. The research of image-based target tracking algorithm is mainly single-mode, but there is a lack of in-depth research on dual-mode, image tracking algorithm, especially in the process of dynamic zoom. To solve the problem that the target is easily lost due to blurred image and the rapid change of scale in dual-mode tracking, a dynamic target zoom dual-mode tracking algorithm based on DS theory is proposed. Through the fusion of visible and infrared tracking target response map at DS decision level, the target position is comprehensively judged by the complementary advantages of dual-mode, and the fusion tracking of large and small field of view is realized. The EOG of visible and infrared image is detected in real time, and the blur information is introduced into the fusion to improve the adaptability of image zoom blur. The tracking problem caused by the rapid change of the scale is avoided by the normalization of the field of view of dual-mode size. The experimental results show that the proposed EOG-DS fusion tracking algorithm improves the stability of tracking, the tracking accuracy is 0.833, the success rate is 0.41, and the tracking rate reaches 26.87 FPS, which meets the real-time requirements.

Key words: dual-mode fusion tracking, zoom blur, Dempster-Shafer（DS） fusion

摘要： 配置大倍率光学系统的双光复合跟踪在地面空中监视取证等领域有广泛的应用。目前基于图像的目标跟踪算法研究主要是单光跟踪，在双光图像跟踪算法，特别是动态变焦过程中图像跟踪方面缺乏深入研究。针对双光跟踪视场变焦时的图像模糊、尺度快速变化容易导致目标丢失的问题，提出基于Dempster-Shafer（DS）理论的动态目标变焦双光跟踪算法。通过DS决策级融合可见光与红外跟踪目标响应图，利用双光互补优势综合判断目标位置，实现大小视场的复合跟踪；实时检测可见光和红外图像目标的能量梯度（energy of gradient，EOG），在融合中引入模糊度信息，提高图像变焦模糊的适应性。通过双光大小视场尺度归一化的处理避免尺度快速变化导致的跟丢问题。实验结果表明，提出的EOG-DS融合跟踪算法提高了跟踪的稳定性，跟踪精确率为0.833，成功率为0.41，均有明显提升，跟踪速率达到26.87?FPS，满足实时性要求。

关键词: 双光融合跟踪, 变焦模糊, DS融合

AO Xuecong, LIU Cheng, ZHANG Duxiang. EOG-DS：Zoom Dual-Mode Tracking Algorithm for Dynamic Targets[J]. Computer Engineering and Applications, 2023, 59(12): 113-121.

敖雪聪, 刘成, 张杜祥. EOG-DS：动态目标变焦双光跟踪算法[J]. 计算机工程与应用, 2023, 59(12): 113-121.

References

[1] 丁正彤，徐磊，张研，等.RGB-T目标跟踪综述[J].南京信息工程大学学报（自然科学版），2019，11（6）：690-697.
DING Z T，XU L，ZHANG Y，et al.A survey of RGB-T object tracking[J].Journal of Nanjing University of Information Science & Technology（Natural Science Edition），2019，11（6）：690-697.
[2] DANELLJAN M，H?GER G，KHAN F，et al.Accurate scale estimation for robust visual tracking[C]//British Machine Vision Conference，Michel，Canada，2014：1-65.
[3] LI Y，ZHU J.A scale adaptive kernel correlation filter tracker with feature integration[C]//European Conference on Computer Vision.Cham：Springer，2014：254-265.
[4] FENG M，SONG K，WANG Y，et al.Learning discriminative update adaptive spatial-temporal regularized correlation filter for RGB-T tracking[J].Journal of Visual Communication and Image Representation，2020，72（12）：102881.
[5] 章泉源，王昱程，陈曦，等.基于全卷积孪生网络的模糊目标跟踪[J].武汉大学学报（理学版），2021，67（5）：411-421.
ZHANG Q Y，WANG Y C，CHEN X，et al.Blurred target tracking by fully-convolutional siamese network[J].Journal of Wuhan University（Natural Science Edition），2021，67（5）：411-421.
[6] 王广龙，田杰，朱文杰，等.特征融合和自适应权重更新相结合的运动模糊目标跟踪[J].光学精密工程，2019，27（5）：1158-1166.
WANG G L，TIAN J，ZHU W J，et al.Feature fusion and weight adaptive updating based motion blur object tracking[J].Optical Precision Engineering，2019，27（5）：1158-1166.
[7] ZHANG X，YE P，LEUNG H，et al.Object fusion tracking based on visible and infrared images：a comprehensive review[J].Information Fusion，2020，63：166-187.
[8] ZHANG X，YE P，PENG S，et al.SiamFT：an RGB-infrared fusion tracking method via fully convolutional siamese networks[J].IEEE Access，2019，7：122122-122133.
[9] ZHANG X，YE P，PENG S，et al.DSiamMFT：an RGB-T fusion tracking method via dynamic siamese networks using multi-layer feature fusion[J].Signal Processing：Image Communication，2020，84：115756.
[10] ZHANG T，LIU X，ZHANG Q，et al.SiamCDA：complementarity-and distractor-aware RGB-T tracking based on siamese network[J].IEEE Transactions on Circuits and Systems for Video Technology，2022，32（3）：1403-1417.
[11] YUN X，SUN Y，YANG X，et al.Discriminative fusion correlation learning for visible and infrared tracking[J].Mathematical Problems in Engineering，2019：1-11.
[12] ZHAI S，SHAO P，LIANG X，et al.Fast RGB-T tracking via cross-modal correlation filters[J].Neurocomputing，2019，334：172-181.
[13] WANG Y，LI C，TANG J.Learning soft-consistent correlation filters for RGB-T object tracking[C]//Chinese Conference on Pattern Recognition and Computer Vision（PRCV）.Cham：Springer，2018：295-306.
[14] WANG Y，WEI X，TANG X，et al.Adaptive fusion CNN features for RGBT object tracking[J].IEEE Transactions on Intelligent Transportation Systems，2021，99：1-10.
[15] 李玲.基于相关性滤波器的实时跟踪算法研究[D].西安：西安电子科技大学，2018.
LI L.The real-time tracker based on correlation filter[D].Xi’an：Xi’an University of Electronic Science and Technology，2018.
[16] HENRIQUES J F，CASEIRO R，MARTINS P，et al.High-speed tracking with kernelized correlation filters[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2014，37（3）：583-596.
[17] 刘桐.基于视频音频联合识别的校园霸凌检测算法研究[D].哈尔滨：哈尔滨工业大学，2020.
LIU T.Research on campus bullying detection algorithm based on video and audio joint recognition[D].Harbin：Harbin Institute of technology，2020.
[18] 赵少锋.基于信息融合的复合制导及目标跟踪算法研究[D].兰州：兰州理工大学，2013.
ZHAO S F.Research on target tracking of muti-guidance using information fusion[D].Lanzhou：Lanzhou University of Technology，2013.
[19] 包丞啸，姜威，王玉潇.基于大津法分割和局部最大梯度的自动聚焦算法[J].光学技术，2019，45（6）：756-761.
BAO C X，JIANG W，WANG Y X.Automatic focusing algorithm based on Otsu segmentation and local maximum gradient[J].Optical Technology，2019，45（6）：756-761.
[20] WANG Y，LI C，TANG J，et al.Learning collaborative sparse correlation filter for real-time multispectral object tracking[C]//International Conference on Brain Inspired Cognitive Systems.Cham：Springer，2018：462-472.
[21] LUO C，SUN B，YANG K，et al.Thermal infrared and visible sequences fusion tracking based on a hybrid tracking framework with adaptive weighting scheme[J].Infrared Physics & Technology，2019，99：265-276.
[22] LI Y，FU C，DING F，et al.AutoTrack：towards high-performance visual tracking for UAV with automatic spatio-temporal regularization[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，2020：11923-11932.
[23] KIANI GALOOGAHI H，FAGG A，LUCEY S.Learning background-aware correlation filters for visual tracking[C]//Proceedings of the IEEE International Conference on Computer Vision，2017：1135-1143.
[24] DANELLJAN M，H?GER G，KHAN F S，et al.Discriminative scale space tracking[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2016，39（8）：1561-1575.
[25] LI C，CHENG H，HU S，et al.Learning collaborative sparse representation for grayscale-thermal tracking[J].IEEE Transactions on Image Processing，2016，25（12）：5743-5756.
[26] 吴秋峰.面向散焦图像的去模糊与深度估计研究[D].哈尔滨：哈尔滨工业大学，2013.
WU Q F.Research on deblurring and depth estimation of defocused images[D].Harbin：Harbin Institute of Technology，2013.
[27] LIU S，ZHOU F，LIAO Q.Defocus map estimation from a single image based on two-parameter defocus model[J].IEEE Transactions on Image Processing，2016，25（12）：5943-5956.