Aiming at the problems of track interruption and ID jump in the battlefield perception from the perspective of military UAVs, this paper proposes a ground multi-target tracking algorithm VA-ByteTrack based on the perspective of UAVs. First, EfficientFormerV2 network was used as the backbone network of detector YOLOX, and CBAM attention mechanism was added to the output to improve the feature extraction capability of detector for small-size targets, and to solve the problem of track interruption caused by fuzzy features of small targets. Secondly, by introducing motion matching cost IOU and detection confidence S, an adaptive Kalman filter AKF based on sigmoid function is constructed to balance the different covariance requirements of high-score detection boxes and low-score detection boxes, and to solve the problem of frequent ID jump caused by dense targets and mutual occlusion. Experimental results show that compared with mainstream methods such as SORT, DeepSORT and ByteTrack, the tracking accuracy of the proposed method is increased by 1.7, 1.4 and 1.0 percent point, the tracking accuracy is increased by 0.7, 0.5 and 1.9 percent point, and the track interruption problem is reduced by 49.4%, 7.7% and 7.2%, respectively. The ID jump problem is reduced by 46.3%, 15.9% and 12.1% respectively, which meets the actual demand of battlefield dynamic perception.
JIANGChao, CUIYuwei, WANGHui. Review on battlefield situational awareness of UAV based on image technology[J]. Measurement & Control Technology, 2021, 40(12): 14-19. (in Chinese)
RUANGuoqing, YIKan, SUNJiadong, et al. Research status and development trend of intelligent battlefield perception technology[J]. Command Information System and Technology, 2022, 13(3): 17-22. (in Chinese)
SUNZhiqiang, SONGLili. Application prospect and realization of security monitoring system based on UAV in smart city[J]. Theoretical Research in Urban Construction, 2024(3): 211-213. (in Chinese)
ZHANGYi, CHENPing. Change detection of unmanned aerial vehicle images based on deep belief networks[J]. Journal of Test and Measurement Technology, 2020, 34(3): 190-196. (in Chinese)
[10]
ZHANGY, MAR G, LIANGC. Road target detection algorithm based on improved YOLOv5 in UAV images[J]. Journal of Measurement Science and Instrumentation, 2024, 15(1): 128-139.
OUYANGQuan, ZHANGYi, MAYan, et al. A review of UAV aerial photography target detection and tracking methods based on deep learning[J]. Electronics Optics & Control, 2024, 31(3): 1-7. (in Chinese)
LISonglin, JIANGJian. Improved UAV small object detection algorithm based on YOLOv5[J]. Journal of Test and Measurement Technology, 2024, 38(4): 354-362. (in Chinese)
[15]
梁天一. 基于深度学习的视觉多目标跟踪方法研究[D]. 长沙: 国防科技大学, 2022.
[16]
羊洁明. 基于深度学习的视频多目标跟踪算法研究[D]. 无锡: 江南大学, 2022.
[17]
BEWLEYA, GEZ, OTT L, et al. Simple online and realtime tracking[C]//2016 IEEE International Conference on Image Processing (ICIP), 2016: 3464-3468.
[18]
WOJKEN, BEWLEYA, PAULUSD. Simple online and realtime tracking with a deep association metric[C]//2017 IEEE International Conference on Image Processing (ICIP), 2017: 3645-3649.
LUOXi, ZHAORui, ZHUANGHuishan, et al. UAV multi-target tracking algorithm jointly optimized by YOLOv5 and deep-SORT[J]. Journal of Signal Processing, 2022, 38(12): 2628-2638. (in Chinese)
SHANGJian, LIUBo, et al. Research on vehicle multi-target tracking algorithm based on UAV aerial video[J/OL]. Acta Metrologica Sinica, 2024, 45(12): 1772-1779. (in Chinese)
[25]
LIUY, LIY, XUD, et al. Adaptive Kalman Filter with power transformation for online multi-object tracking[J]. Multimedia Systems, 2023, 29(3): 1231-1244.
[26]
ZHANGY, TAOQ, YINY. A lightweight man-overboard detection and tracking model using aerial images for maritime search and rescue[J]. Remote Sensing, 2023, 16(1): 165.
[27]
YUANK, WANGQ, MIY, et al. Improved feature fusion in YOLOv5 for accurate detection and counting of Chinese flowering cabbage (brassica campestris L. ssp. chinensis var. utilis tsen et lee) buds[J]. Agronomy, 2023, 14(1): 42.
[28]
GUOX, ZHAOX. Oriented-YOLOX: an improved YOLOX for rotated object detection[C]//Proceedings of the 2024 5th International Conference on Computing, Networks and Internet of Things, 2024: 261-265.
[29]
ASIRIA A, SHAFA, ALI T, et al. Advancing brain tumor classification through fine-tuned vision transformers: a comparative study of pre-trained models[J]. Sensors, 2023, 23(18): 7913.
[30]
LIY, HUJ, WENY, et al. Rethinking vision transformers for MobileNet size and speed[C]//2023 IEEE/CVF International Conference on Computer Vision (ICCV), 2023: 16843-16854.
[31]
WOO S, PARKJ, LEEJ Y, et al. CBAM: convolutional block attention module[C]//2018 IEEE/CVF International Conference on Computer Vision (ICCV), 2018: 3-19.
LIShunfang. Design of compensation-adaptive Kalman filtering for online multi object tracking[J]. Journal of Chuzhou University, 2023, 25(5): 30-34. (in Chinese)
[34]
CHENG, WANGW, HEZ, et al. VisDrone-MOT2021: the vision meets drone multiple object tracking challenge results[C]//2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), 2021: 2839-2846.
[35]
XUX, FENGZ, CAOC, et al. STN-track: multiobject tracking of unmanned aerial vehicles by swin transformer neck and new data association method[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022, 15: 8734-8743.