YE Chu,MA Shangchang,PAN Yun.Research on Trajectory Tracking of Precipitation Particles based on Machine Vision[J].Journal of Chengdu University of Information Technology,2026,41(02):174-179.[doi:10.16836/j.cnki.jcuit.2026.02.006]
基于机器视觉的降水粒子运动轨迹追踪研究
- Title:
- Research on Trajectory Tracking of Precipitation Particles based on Machine Vision
- 文章编号:
- 2096-1618(2026)02-0174-06
- Keywords:
- machine vision; precipitation particles; trajectory tracking; Kalman filter; Hungarian algorithm
- 分类号:
- P426.62+2
- 文献标志码:
- A
- 摘要:
- 在基于机器视觉的降水观测研究中,准确获取每个降水粒子的物理信息是提高降水现象和降水强度观测精度的关键。在视频采样空间内,单个降水粒子在下落过程中会被连续多帧图像捕获。为避免降水粒子数量被重复计数,使用卡尔曼滤波器和匈牙利算法联合应用策略实现降水粒子的运动轨迹追踪。实验表明,该策略能有效区分不同的降水粒子,实现对多降水粒子的跟踪。
- Abstract:
- In the research of precipitation reset based on vision machines, the accurate acquisition of the physical information of each precipitation particle is the key to improving the reset accuracy of precipitation phenomena and precipitation intensity. In the video collection space, a single precipitation particle is captured in text by multiple consecutive frames of images during its fall. In order to avoid counting the number of precipitation particles repeatedly, a joint application strategy of the Kalman filter and the Hungarian algorithm is used to track the movement trajectories of precipitation particles. Experiments show that this strategy can effectively distinguish different precipitation particles and achieve tracking of multiple precipitation particles.
参考文献/References:
[1] 刘振东.翻斗雨量计误差试验研究及改正措施[D].南京:河海大学,2007.
[2] 中国气象局.地面气象观测规范[M].北京:气象出版社,2003:54-59.
[3] 蔡彦,李应绪,单长吉,等.称重式光学雨量计的设计[J].物理通报,2017,46(6):77-79.
[4] BAGREER.Characterization and design of a readout circuit for a piezoelectric-based acoustic disdrometer[J].Microelectronics & Computer Engineering,2012,2(1):4-9.
[5] 朱力强,罗婷,蔺潇.基于压电效应的雨量与雨滴谱检测[J].振动.测试与诊断,2020,40(1):155-162,209.
[6] Parsivel激光雨滴谱仪与雨量计观测降水的一致性分析[J].气象科技,2020,48(2):147-153.
[7] 许华春.激光雨滴谱仪的研究与改进[D].淮南:安徽理工大学,2017.
[8] 张建云,张垚.LPA10激光雨滴谱仪的定标及观测对比[J].气象科技进展,2018,8(6):62-70.
[9] Thurai,M,Bringi,& V N.Drop axis ratios from a 2D video disdrometer[J].JOURNAL OF ATMOSPHERIC AND OCEANIC TECHNOLOGY,2025,22(7):966-978.
[10] Schuur T J,Ryzhkov A V,Zrnic D S,et al.Drop size distributions measured by a 2D video disdrometer:Comparison with dual-polarization radar data[J].J.Appl.Meteor.,2001,40:1019-1034.
[11] Grazioli J,Tuia D,Monhart S,et al.Hydrometeor classification from two-dimensional video disdrometer data[J].Atmospheric Measurement Techniques,2014,7(9):2869-2882.
[12] Kathiravelu Gopinath,Lucke Terry,Nichols Peter.Rain Drop Measurement Techniques:A Review[J].WATER,2016,8(1),29.
[13] 安栋.复杂场景下目标追踪算法研究[D],长沙:中南大学,2023.
[14] 吕红强,黄涛,聂兴家,等.光流法在地铁车辆测速中的应用研究[J].机车电传动,2022(2):129-134.
[15] 王广玉,窦磊,窦杰.基于自适应卡尔曼滤波的多目标跟踪算法[J].计算机应用,2022,42(S1):271-275.
[16] 任珈民,宫宁生,韩镇阳.基于YOLOv3与卡尔曼滤波的多目标跟踪算法[J].计算机应用与软件,2020,37(5):169-176.
[17] 谷稳.基于进化匈牙利算法的目标分配问题研究及应用[D].西安:西安电子科技大学,2013.
[18] 刘磊,高太长,李浩.三种基于光学原理的降水类型识别技术[C].中国气象学会2006年年会“提高大气监测自动化水平、为业务技术体制改革作贡献”分会场论文集,2006:202-205.
[19] Atlas D,Srivastava R C,Sekhon R S.Doppler radar characteristics of precipitation at vertical incidence,Reviews of Geophysics,1973,11(1):1-35.
备注/Memo
收稿日期:2024-08-21