Improved GM-PHD filtering algorithm for multi-target tracking in sonar images

被引：0
|
作者：
Zhou T. [1 ,2 ,3 ]
Zhang L. [1 ,2 ,3 ]
Du W. [1 ,2 ,3 ]
Han T. [1 ,2 ,3 ]
机构：
[1] Acoustic Science and Technology Laboratory, Harbin Engineering University, Harbin
[2] Key Laboratory of Marine Information Acquisition and Security of Ministry of Industry and Information Technology, Harbin Engineering University, Harbin
[3] College of Underwater Acoustic Engineering, Harbin Engineering University, Harbin
来源：
Harbin Gongcheng Daxue Xuebao/Journal of Harbin Engineering University | 2020年 / 41卷 / 05期
关键词：
Gaussian mixture probability hypothesis density(GM-PHD) filter; Multi-target tracking; Random finite set; Sonar image; Track-before-detect(TBD);
D O I：
10.11990/jheu.201808089
中图分类号：
学科分类号：
摘要：
The standard Gaussian mixture probability hypothesis density (GM-PHD) filtering algorithm assumes that the newborn target intensity is known a priori; this assumption does not satisfy the multi-target tracking of sonar images. An improved GM-PHD filtering algorithm is proposed to solve this problem. The initialization is performed using measurement instead of uniform distribution, and the convergence rate of the algorithm is enhanced. In the algorithm loop at each moment, the current measurement is used to drive the newborn target, but no decision is made until the end of the next time step based on the tracking result, which can break through the limitation of the assumption "newborn target intensity is known a priori" and reduce the tracking error. Simulation results show that the improved GM-PHD filter reduces the tracking error and computation time. In addition, the effects of the detection probability and clutter density on the tracking performance are analyzed. Image sonar multi-target tracking pool experiment verifies that the improved GM-PHD filtering algorithm can perform multi-target tracking effectively in engineering applications. Copyright ©2020 Journal of Harbin Engineering University.
引用
下载
收藏
页码：691 / 697
页数：6
相关论文
共 15 条
  • [1] BAR-SHALOM Y, FORTMANN T E., Tracking and data association, (1988)
  • [2] BLACKMAN S S., Multiple hypothesis tracking for multiple target tracking[J], IEEE aerospace and electronic systems magazine, 19, 1, pp. 5-18, (2004)
  • [3] FORTMANN T, BAR-SHALOM Y, SCHEFFE M., Sonar tracking of multiple targets using joint probabilistic data association[J], IEEE journal of oceanic engineering, 8, 3, pp. 173-184, (1983)
  • [4] MAHLER R P S, EBRARY I., Statistical multisource-multitarget information fusion, Statistical Multisource-Multitarget Information Fusion, (2007)
  • [5] MAHLER R P S., Statistics 101" for multisensor, multitarget data fusion[J], IEEE aerospace and electronic systems magazine, 19, 1, pp. 53-64, (2004)
  • [6] MAHLER R., A theoretical foundation for the stein-winter Probability Hypothesis Density (PHD) multi-target tracking approach, Proceedings of the MSS National Symposium on Sensor and Data Fusion, pp. 99-117, (2000)
  • [7] MAHLER R P S., Multitarget Bayes filtering via first-order multitarget moments[J], IEEE transactions on aerospace and electronic systems, 39, 4, pp. 1152-1178, (2003)
  • [8] VO B N, MA W K., The gaussian mixture probability hypothesis density filter[J], IEEE transactions on signal processing, 54, 11, pp. 4091-4104, (2006)
  • [9] CLARK D E, PANTA K, VO B N., The GM-PHD filter multiple target tracker, Proceedings of the 2006 9th International Conference on Information Fusion, pp. 1-8, (2006)
  • [10] PANTA K, CLARK D E, VO B N., Data association and track management for the Gaussian mixture probability hypothesis density filter[J], IEEE transactions on aerospace and electronic systems, 45, 3, pp. 1003-1016, (2009)
12