基于多步匹配的超声多目标跟踪方法与系统构建

朱星龙; 艾丹妮

doi:null

您当前的位置：

首页 >

文章列表页 >

基于多步匹配的超声多目标跟踪方法与系统构建

光学传感与测量 | 更新时间：2026-01-22

- 基于多步匹配的超声多目标跟踪方法与系统构建
- Ultrasound multi-object tracking method based on multi-step matching and system construction
- 光学技术 2026年52卷第1期页码：115-122
- 作者机构：
  
  北京理工大学光电学院，北京 100081
- 作者简介：
  
  朱星龙（1996— ），男，硕士研究生，从事医学图像处理研究。xlzhu1996@163.com
  艾丹妮（1988— ），女，特别研究员、博士生导师，博士，从事医学图像智能分析、手术导航、虚拟现实与增强现实、智能传感及人工智能等研究。dannibj@163.com
- 基金信息：
  
  国家自然科学基金面上项目(61971040)
- DOI：
  中图分类号： TP394.1
- 收稿：2022-06-23，
  
  修回：2022-09-03，
  
  纸质出版：2026-01-20
- 稿件说明：
移动端阅览
朱星龙,艾丹妮.基于多步匹配的超声多目标跟踪方法与系统构建[J].光学技术,2026,52(01):115-122.

ZHU Xinglong,AI Danni.Ultrasound multi-object tracking method based on multi-step matching and system construction[J].OPTICAL TECHNIQUE,2026,52(01):115-122.
朱星龙,艾丹妮.基于多步匹配的超声多目标跟踪方法与系统构建[J].光学技术,2026,52(01):115-122. DOI：

ZHU Xinglong,AI Danni.Ultrasound multi-object tracking method based on multi-step matching and system construction[J].OPTICAL TECHNIQUE,2026,52(01):115-122. DOI：

摘要

在肝脏肿瘤治疗中，采用超声引导放射治疗能够最大程度减少对病人正常组织的伤害。超声引导放射治疗通过超声探头实时获取病人的肿瘤位置信息，并控制放射源跟随肿瘤移动。超声引导放射治疗技术的难点在于如何通过超声图像预测目标位置，这个步骤也称为超声目标跟踪。文章研究了基于多步匹配的超声多目标跟踪方法，该方法能同时跟踪肝脏超声图像序列中的多个肝脏血管。首先检测-跟踪多任务网络根据序列当前帧初步预测目标位置的若干提案，同时卡尔曼滤波器会根据目标的历史位置信息预测目标的先验位置，然后多步匹配方法会对目标位置提案和先验位置进行匹配，最后利用轨迹活跃度对输出的目标进行筛选。文章构建了超声目标跟踪系统。该系统可以通过图形用户界面完成多目标跟踪演示、图像标注和评估功能。文章根据MOT竞赛的指标将提出的方法与其他方法进行了对比及消融实验，实验结果证明文章方法具有更高的精确度和鲁棒性。同时，通过功能测试和性能测试证明了超声目标跟踪系统能够稳定地在准实时条件下完成设计的各项功能。

Abstract

In the treatment of liver tumors

the use of ultrasound-guided radiation therapy can minimize the damage to the normal tissue of the patient. Ultrasound-guided radiation therapy obtains the patient's tumor location information by the ultrasound probe in real time

and controls the radiation source to follow the tumor movement. As the essential technology of ultrasound-guided radiation therapy. ultrasound object tracking aims to obtain the positions of a specific target in the ultrasound sequences. A multi-object tracking method for ultrasound images based on multi-step matching is proposed

allowing multiple liver blood vessels to be tracked in a sequence of liver ultrasound images. The proposed method first inputs the image into the detection-tracking multi-task network to obtain the target position proposal

and filters the prior position of the target through Kalman filter at the same time. Then the target position proposal and prior position are matched by the multi-step matching method. Finally the estimated target positions are picked by their trajectory activity. This paper also builds an ultrasound object tracking system based on the proposed method. Besides tracking multiple objects

the system can also demonstrate annotating images and evaluating tracking results in a graphical user interface The proposed method is evaluated with the indicators of MOT competition

and the ablation experiments reveal the availability of our improvement. The system are also evaluated with function tests and performance test

which prove the reliability of the proposed system.

关键词

Keywords

references

De Luca V , Banerjee J , Hallack A , et al . Evaluation of 2D and 3D ultrasound tracking algorithms and impact on ultrasound‐guided liver radiotherapy margins [J ] . Medical Physics , 2018 , 45 （ 11 ）: 4986 — 5003 .

De Luca V , Benz T , Kondo S , et al . The 2014 liver ultrasound tracking benchmark [J ] . Physics in Medicine & Biology , 2015 , 60 （ 14 ）: 5571 .

Liu F , Liu D , Tian J , et al . Cascaded one-shot deformable convolutional neural networks: Developing a deep learning model for respiratory motion estimation in ultrasound sequences [J ] . Medical Image Analysis , 2020 , 65 : 101793 .

Leal-Taixé L , Milan A , Reid I , et al . MOTChallenge 2015: Towards a benchmark for multi-target tracking [J/OL ] . arXiv preprint arXiv: 1504.01942 , 2015 .

Milan A , Leal-Taixé L , Reid I , et al . MOT16: A benchmark for multi-object tracking [J/OL ] . arXiv preprint arXiv: 1603.00831 , 2016 .

Dendorfer P , Rezatofighi H , Milan A , et al . MOT20: A benchmark for multi object tracking in crowded scenes [J/OL ] . arXiv preprint arXiv: 2003.09003 , 2020 .

Bewley A , Ge Z , Ott L , et al . Simple online and realtime tracking [C ] // 2016 IEEE International Conference on Image Processing （ICIP） . Phoenix, USA : IEEE , 2016 : 3464 — 3468 .

Wojke N , Bewley A , Paulus D . Simple online and realtime tracking with a deep association metric [C ] // 2017 IEEE International Conference on Image Processing （ICIP） . Beijing,China : IEEE , 2017 : 3645 — 3649 .

Zhang Y , Sun P , Jiang Y , et al . ByteTrack: Multi-object tracking by associating every detection box [C ] // European Conference on Computer Vision （ECCV） . Tel Aviv, Israel : Springer , 2022 : 1 — 17 .

Bergmann P , Meinhardt T , Leal-Taixé L . Tracking without bells and whistles [C ] // Proceedings of the IEEE/CVF International Conference on Computer Vision （ICCV） . Seoul, South Korea : IEEE , 2019 : 941 — 951 .

Tang S , Andriluka M , Andres B , et al . Multiple people tracking by lifted multicut and person re-identification [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition （CVPR） . Honolulu, USA : IEEE , 2017 : 3539 — 3548 .

Shuai B , Berneshawi A , Li X , et al . SiamMOT: Siamese multi-object tracking [C ] // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR） . Nashville, USA : IEEE , 2021 : 12372 — 12382 .

Kieritz H , Hubner W , Arens M . Joint detection and online multi-object tracking [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops （CVPRW） . Salt Lake City, USA : IEEE , 2018 : 1459 — 1467 .

Pearson K . LIII. On lines and planes of closest fit to systems of points in space [J ] . The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science , 1901 , 2 （ 11 ）: 559 — 572 .

Wang M , Liu Y , Huang Z . Large margin object tracking with circulant feature maps [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition （CVPR） . Honolulu, USA : IEEE , 2017 : 4021 — 4029 .

Kim S J , Nam J Y , Ko B C . Online tracker optimization for multi-pedestrian tracking using a moving vehicle camera [J ] . IEEE Access , 2018 , 6 : 48675 — 48687 .

Zhou X , Koltun V , Krähenbühl P . Tracking objects as points [C ] // European Conference on Computer Vision （ECCV） . Glasgow, UK : Springer , 2020 : 474 — 490 .

Feng W , Hu Z , Wu W , et al . Multi-object tracking with multiple cues and switcher-aware classification [J/OL ] . arXiv preprint arXiv: 1901.06129 , 2019 .

Weng X , Wang J , Held D , et al . AB3DMOT: A baseline for 3D multi-object tracking and new evaluation metrics [C ] // IEEE International Conference on Robotics and Automation（ICRA） . Xi’an, China : IEEE , 2021 : 10942 — 10948 .

Kim A , Ošep A , Leal-Taixé L . EagerMOT: 3D multi-object tracking via sensor fusion [C ] // IEEE/RSJ International Conference on Intelligent Robots and Systems （IROS） . Prague,Czech Republic : IEEE , 2021 : 8148 — 8155 .

Wang Z , Zheng L , Liu Y , et al . Towards real-time multi-object tracking [C ] // European Conference on Computer Vision （ECCV） . Glasgow,UK : Springer , 2020 : 107 — 122 .

Yu F , Wang D , Shelhamer E , et al . Deep layer aggregation [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition （CVPR） . Salt Lake City,USA : IEEE , 2018 : 2403 — 2412 .

He K , Gkioxari G , Dollár P , et al . Mask R-CNN [C ] // Proceedings of the IEEE International Conference on Computer Vision （ICCV） . Venice, Italy : IEEE , 2017 : 2961 — 2969 .

Li B , Yan J , Wu W , et al . High performance visual tracking with siamese region proposal network [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition （CVPR） . Salt Lake City, USA : IEEE , 2018 : 8971 — 8980 .

Bernardin K , Stiefelhagen R . Evaluating multiple object tracking performance: the CLEAR MOT metrics [J ] . EURASIP Journal on Image and Video Processing , 2008 , 2008 : 1 — 10 .

Ristani E , Solera F , Zou R , et al . Performance measures and a data set for multi-target, multi-camera tracking [C ] // European Conference on Computer Vision （ECCV） . Amsterdam, Netherlands : Springer , 2016 : 17 — 35 .

Li B , Wu W , Wang Q , et al . SiamRPN++: Evolution of siamese visual tracking with very deep networks [C ] // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR） . Long Beach,USA : IEEE , 2019 : 4282 — 4291 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于波前预测模型的波前校正方法

基于残差注意力机制的离轴数字全息重建

长距离光纤通信系统的非线性损伤补偿方法

基于RDCAN深度学习的三维共焦拉曼光谱高分辨成像方法