TY - GEN
T1 - UAV Target Tracking Based on Adaptive Spatio-Temporal Regularization
AU - Meng, Lingbo
AU - Wu, Dongjian
AU - Zhang, Boheng
AU - Sun, Mingjian
N1 - Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
PY - 2022
Y1 - 2022
N2 - Aiming at the problem of fast motion and motion blur in UAV target tracking process, an adaptive spatiotemporal regularization correlation filtering target tracking algorithm is proposed. Firstly, in order to meet the real-time requirements of UAV target tracking, a fast correlation filtering target tracking algorithm framework is adopted. Secondly, based on the correlation filtering target tracking algorithm, a temporal regularization and spatial regularization method is proposed to deal with the problem of fast-moving target and solve the boundary effect problem of correlation filtering. Then, the optimal solution of the constructed ridge regression function is obtained. In order to solve the problem quickly, the alternating direction multiplier (ADMM) iterative method is come up and used as the solution for the optimal correlation filter template to ensure the target tracking in the subsequent frames. Finally, the fast motion challenge of target is tested on three different datasets: OTB100, UAV123 and DTB70. The experimental results show that the algorithm proposed in this paper can better deal with the boundary effect caused by the rapid movement of the target compared with other relevant filtering algorithms.
AB - Aiming at the problem of fast motion and motion blur in UAV target tracking process, an adaptive spatiotemporal regularization correlation filtering target tracking algorithm is proposed. Firstly, in order to meet the real-time requirements of UAV target tracking, a fast correlation filtering target tracking algorithm framework is adopted. Secondly, based on the correlation filtering target tracking algorithm, a temporal regularization and spatial regularization method is proposed to deal with the problem of fast-moving target and solve the boundary effect problem of correlation filtering. Then, the optimal solution of the constructed ridge regression function is obtained. In order to solve the problem quickly, the alternating direction multiplier (ADMM) iterative method is come up and used as the solution for the optimal correlation filter template to ensure the target tracking in the subsequent frames. Finally, the fast motion challenge of target is tested on three different datasets: OTB100, UAV123 and DTB70. The experimental results show that the algorithm proposed in this paper can better deal with the boundary effect caused by the rapid movement of the target compared with other relevant filtering algorithms.
KW - Adaptive spatiotemporal regularization
KW - Correlation filter
KW - UAV target tracking
UR - https://www.scopus.com/pages/publications/85130909838
U2 - 10.1007/978-981-16-9492-9_194
DO - 10.1007/978-981-16-9492-9_194
M3 - 会议稿件
AN - SCOPUS:85130909838
SN - 9789811694912
T3 - Lecture Notes in Electrical Engineering
SP - 1966
EP - 1975
BT - Proceedings of 2021 International Conference on Autonomous Unmanned Systems, ICAUS 2021
A2 - Wu, Meiping
A2 - Niu, Yifeng
A2 - Gu, Mancang
A2 - Cheng, Jin
PB - Springer Science and Business Media Deutschland GmbH
T2 - International Conference on Autonomous Unmanned Systems, ICAUS 2021
Y2 - 24 September 2021 through 26 September 2021
ER -