Improved joint probabilistic data association algorithm based on Meanshift clustering and Bhattacharya likelihood modification

doi:10.11772/j.issn.1001-9081.2014.05.1279

Journal of Computer Applications ›› 2014, Vol. 34 ›› Issue (5): 1279-1282.DOI: 10.11772/j.issn.1001-9081.2014.05.1279

Previous Articles Next Articles

Improved joint probabilistic data association algorithm based on Meanshift clustering and Bhattacharya likelihood modification

TIAN Jun,LI Dan,XIAO Liqing

Jiangsu Key Laboratory of Large Engineering Equipment Detection and Control, Xuzhou Institute of Technology, Xuzhou Jiangsu 221000, China

Received:2013-10-12 Revised:2013-12-25 Online:2014-05-30 Published:2014-05-01
Contact: TIAN Jun

基于Meanshift聚类Bhattacharya观测似然度修正的联合概率数据关联改进算法

田隽,厉丹,肖理庆

徐州工程学院江苏省大型工程装备检测与控制重点建设实验室，江苏徐州 221000

通讯作者: 田隽
作者简介:田隽(1981-)，女，重庆渝中人，副教授，博士，主要研究方向：非线性滤波技术、基于粒子滤波的视频跟踪技术；厉丹(1981-)，女，江苏徐州人，讲师，博士，主要研究方向：视频目标检测与跟踪；肖理庆（1981-），男，山东青岛人，讲师，主要研究方向：智能算法、智能控制。
基金资助:
江苏省高校自然科学研究项目;徐州市科技项目

Abstract

Abstract:

To reduce the calculation complexity of the Joint Probabilistic Data Association (JPDA) joint-association events, due to multiple targets' tracks aggregation, an improved JPDA algorithm, clustering by Meanshift algorithm and optimizing confirmation matrix by Bhattacharya coefficients,was proposed.The clustering center was created by Meanshift algorithm. Then the tracking gate was obtained by calculating Mahalanobis distance between the clustering center and targets' prediction observation. The Bhattacharya likelihood matrix which was as a basis for low probability events was created, consequently the computing complexity of JPDA joint-association events which was related to low probability events was reduced. The experimental results show that the new method is superior to the conventional JPDA both in computational complexity and precision of estimation for multiple targets' tracks aggregation.

摘要：

为降低多目标航迹聚集时联合概率数据关联（JPDA）联合关联事件的计算复杂度，提出一种基于Meanshift聚类〖CD*2〗Bhattacharya（Bhy）观测似然度修正的JPDA改进算法。利用Meanshift得到聚类中心，据聚类中心与目标预测量测马氏距离形成跟踪门；提出Bhy似然度矩阵，将Meanshift聚类中心与各量测Bhy距离所表征的观测似然度作为确认矩阵小概率事件划分依据，消除确认矩阵中小概率事件对联合关联事件计算复杂度的影响。实验结果表明：多目标航迹聚集时，该算法在减少计算复杂度同时保持了较高关联精度，跟踪性能明显优于经典JPDA。

CLC Number:

TP391

TIAN Jun LI Dan XIAO Liqing. Improved joint probabilistic data association algorithm based on Meanshift clustering and Bhattacharya likelihood modification[J]. Journal of Computer Applications, 2014, 34(5): 1279-1282.

田隽厉丹肖理庆. 基于Meanshift聚类Bhattacharya观测似然度修正的联合概率数据关联改进算法[J]. 计算机应用, 2014, 34(5): 1279-1282.

References

［1］BAR-SHALOM Y, LI X R. Multitarget-multisensor tracking: principles and techniques ［M］. Storrs, CT: YBS Publishing, 1995.
［2］SUN J, CHEN W, TANG L, et al. Face detection in the low-luminous-density and low-resolution images within limited room ［J］. Journal of China University of Mining and Technology, 2008,37(3):373-377.（孙继平，陈伟，唐亮，等.有限空间低照度低分辨率图像中人脸检测［J］.中国矿业大学学报，2008,37(3):373-377.）
［3］TIAN J. Research on visual targets tracking algorithm based on particle filter ［D］. Xuzhou: China University of Mining and Technology, 2010.（田隽.基于粒子滤波的视频目标跟踪算法的研究［D］.徐州：中国矿业大学，2010.）
［4］OUSSALAH M, de SCHUTTER J. Hybrid fuzzy probabilistic data association filter and joint probabilistic data association filter ［J］. Information Sciences, 2002,142(1/2/3/4):195-226.
［5］AOKI E H, KIENITZ K H. Suboptimal JPDA for tracking in the presence of clutter and missed detections ［C］// Proceedings of the 2009 12th International Conference on Information Fusion. Piscataway: IEEE Press, 2009:818-825.
［6］VERMAAK J, GODSILL S J, PEREZ P. Monte Carlo filtering for multi target tracking and data association ［J］. IEEE Transactions on Aerospace and Electronic Systems, 2005,41(1):309-332.
［7］HAN H, ZHI J, JIAO L, et al. Data association for multiple targets based on meanshift in clutter ［J］. Journal of System Simulation, 2009,21(11):3351-3355.（韩红，智建纬，焦李成，等.杂波环境下基于meanshift的多目标数据关联［J］.系统仿真学报,2009,21(11):3351-3355.）
［8］CLARK D,VO B T,VO B N. Gaussian particle implementations of probability hypothesis density filters ［C］// Proceedings of the 2007 IEEE Aerospace Conference. Piscataway: IEEE Press, 2007:1-11.
［9］VO B N, MA W K. The Gaussian mixture probability hypothesis density filter ［J］. IEEE Transactions on Signal Processing, 2006,54(11):4091-4104.
［10］ZHANG J, JI H. Gaussian mixture particle probability hypothesis density based passive bearing-only multi-target tracking ［J］. Control and Decision, 2011,26(3):413-417.（张俊根，姬红兵.高斯混合粒子PHD滤波被动测角多目标跟踪［J］.控制与决策，2011,26(3):413-417.）
［11］COMANICIU D, MEER P. Mean shift analysis and applications ［C］// ICCV '99: Proceedings of the Seventh IEEE International Conference on Computer Vision. Washington, DC: IEEE Computer Society, 1999,2:1197-1203.
［12］COMANICIU D, MEER P. Mean shift: a robust approach toward feature space analysis ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002,24(5):603-619.

[1]	. Pavement defect detection algorithm with enhanced morphological perception [J]. Journal of Computer Applications, 0, (): 0-0.
[2]	YANG Jianfeng, CHEN Bin, LI Yuxuan. Self-supervised point cloud anomaly detection method based on point cloud reconstruction [J]. Journal of Computer Applications, 0, (): 0-0.
[3]	XUE Zhenhua, LI Qiang, HUANG Chao. Vision foundation model-driven pixel-level image anomaly detection model [J]. Journal of Computer Applications, 0, (): 0-0.
[4]	. Binocular vision object location algorithm for robot arm grasping [J]. Journal of Computer Applications, 0, (): 0-0.
[5]	. Multi-attention contrastive learning for infrared small target detection [J]. Journal of Computer Applications, 0, (): 0-0.
[6]	LI Zhonghua, ZHONG Gengxin, FAN Ping, ZHU Hengliang. Camouflaged object detection by boundary mining and background guidance [J]. Journal of Computer Applications, 0, (): 0-0.
[7]	. Multi-view difficult airway recognition based on discriminant region guidance [J]. Journal of Computer Applications, 0, (): 0-0.
[8]	. Privacy preserving localization of surveillance images based on large vision models [J]. Journal of Computer Applications, 0, (): 0-0.
[9]	. Gravity data denoising method based on multilevel wavelet residual network#br# #br# [J]. Journal of Computer Applications, 0, (): 0-0.
[10]	. Data augmentation method for abnormal passenger behavior in elevators based on dynamic graph convolutional network [J]. Journal of Computer Applications, 0, (): 0-0.
[11]	. Text-based person retrieval method based on multi-granularity shared semantic center association [J]. Journal of Computer Applications, 0, (): 0-0.
[12]	ZHANG Qing, YANG Fan, FANG Yuhan. Chinese Spelling Correction Algorithm Based on Multi-Modal Information Fusion [J]. Journal of Computer Applications, 0, (): 0-0.
[13]	WANG Hao, WANG Jinwei, CHENG Xin, ZHANG Jiawei, WU Hao, LUO Xiangyang, MA Bin. A Review of JPEG Recompression Forensics for Color Images [J]. Journal of Computer Applications, 0, (): 0-0.
[14]	. Distribution adaptation and dynamic curriculum pseudo-labeling framework for semi-supervised fire detection [J]. Journal of Computer Applications, 0, (): 0-0.
[15]	. Multimodal harmful content detection method based on weakly supervised modality semantic enhancement [J]. Journal of Computer Applications, 0, (): 0-0.

Improved joint probabilistic data association algorithm based on Meanshift clustering and Bhattacharya likelihood modification

基于Meanshift聚类Bhattacharya观测似然度修正的联合概率数据关联改进算法

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics