基于机器学习方法的三维粒子重构技术
Particle reconstruction of volumetric particle image velocimetry with strategy of machine learning
查看参考文献28篇
|
文摘
|
通过三维粒子重构获取粒子场的分布情况是层析粒子图像测速的关键步骤,有限二维投影下的三维粒子重构是一个欠定的反问题,其精确解往往很难得到。一般情况下,可以通过优化方法得到近似解。为了获取质量更高的粒子场并用于层析粒子图像测速,提出了一种基于卷积神经网络(Convolutional Neural Networks,CNN)的粒子重构方法。所提出的技术可以从基于传统的代数重构技术(Algebraic Reconstruction Technique,ART)的方法所得到的粗略粒子分布中进一步提高粒子重构质量。与现有的基于ART的算法相比,新技术在重构质量方面有了显著的改进,可以有效剔除虚假粒子并更准确地还原粒子形状,并且在粒子浓度较稠密的情况下计算速度至少快了一个数量级。 |
|
其他语种文摘
|
Three-dimensional particle reconstruction with limited two-dimensional projections is an underdetermined inverse problem that the exact solution is often difficult to be obtained.In general,approximate solutions can be obtained by optimization methods.In order to obtain a better quality particle field for Tomographic PIV,in the current work,apractical particle reconstruction method based on convolutional neural network(CNN)is proposed.The proposed technique can refine the particle reconstruction from a very coarse initial guess of particle distribution from any traditional algebraic reconstruction technique (ART)based methods. Compared with available ART-based algorithms,the novel technique makes significant improvements in terms of reconstruction quality.It can effectively eliminate ghost particles and restore the shape of particles more accurately,and is at least an order of magnitude faster with dense particle concentration. |
|
来源
|
实验流体力学
,2021,35(3):88-93 【核心库】
|
|
DOI
|
10.11729/syltlx20200141
|
|
关键词
|
机器学习
;
粒子重构
;
层析粒子图像测速
;
卷积神经网络
;
重构质量
|
|
地址
|
1.
浙江大学航空航天学院, 杭州, 310027
2.
中国科学院力学研究所, 北京, 100190
3.
鞍钢集团钢铁研究院冶金工艺研究所, 辽宁, 鞍山, 114009
4.
海洋装备用金属材料及其应用国家重点实验室, 海洋装备用金属材料及其应用国家重点实验室, 辽宁, 鞍山, 114009
5.
北京立方天地科技发展有限责任公司, 北京, 100083
6.
北京航空航天大学, 流体力学教育部重点实验室, 北京, 100191
|
|
语种
|
中文 |
|
文献类型
|
研究性论文 |
|
ISSN
|
1672-9897 |
|
学科
|
力学;自动化技术、计算机技术 |
|
基金
|
国家自然科学基金面上项目
;
海洋装备用金属材料及其应用国家重点实验室开放基金课题
;
国家重点研发计划资助
|
|
文献收藏号
|
CSCD:7008077
|
参考文献 共
28
共2页
|
|
1.
Elsinga G E. Tomographic particle image velocimetry.
Experiments in Fluids,2006,41(6):933-947
|
CSCD被引
57
次
|
|
|
|
|
2.
Scarano F. Tomographic PIV:principles and practice.
Measurement Science and Technology,2013,24(1):012001
|
CSCD被引
26
次
|
|
|
|
|
3.
Gao Q. Review on development of volumetric particle image velocimetry.
Chinese Science Bulletin,2013,58(36):4541-4556
|
CSCD被引
13
次
|
|
|
|
|
4.
Discetti S. Spatial filtering improved tomographic PIV.
Experiments in Fluids,2013,54(4):1-13
|
CSCD被引
6
次
|
|
|
|
|
5.
Worth N A. Acceleration ofTomo-PIV by estimating the initial volume intensity distribution.
Experiments in Fluids,2008,45(5):847-856
|
CSCD被引
9
次
|
|
|
|
|
6.
Atkinson C. An efficient simultaneous reconstruction technique for tomographic particle image velocimetry.
Experiments in Fluids,2009,47(4/5):553-568
|
CSCD被引
13
次
|
|
|
|
|
7.
Elsinga G E. Ghost hunting-an assessment of ghost particle detection and removal methods for tomographic-PIV.
Measurement Science and Technology,2014,25(8):084004
|
CSCD被引
1
次
|
|
|
|
|
8.
De Silva C M. EnhancingTomo-PIV reconstruction quality by reducing ghost particles.
Measurement Science and Technology,2013,24(2):024010
|
CSCD被引
3
次
|
|
|
|
|
9.
Schanz D. 'Shake The Box'-a 4D PTV algorithm: Accurate and ghostless reconstruction of Lagrangian tracks in densely seeded flows.
Proc of the 17th International Symposium on Applications of Laser Techniques to Fluid Mechanics,2014
|
CSCD被引
1
次
|
|
|
|
|
10.
Schanz D. Shake-The-Box:Lagrangian particle tracking at high particle image densities.
Experiments in Fluids,2016,57(5):1-27
|
CSCD被引
6
次
|
|
|
|
|
11.
Wieneke B. Iterative reconstruction of volumetric particle distribution.
Measurement Science and Technology,2013,24(2):024008
|
CSCD被引
1
次
|
|
|
|
|
12.
Lynch K P. An efficient and accurate approach to MTE-MART for time-resolved tomographic PIV.
Experiments in Fluids,2015,56(3):1-16
|
CSCD被引
2
次
|
|
|
|
|
13.
Novara M. Motion tracking-enhanced MART for tomographic PIV.
Measurement Science and Technology,2010,21(3):035401
|
CSCD被引
4
次
|
|
|
|
|
14.
Wang H P. Intensity-enhanced MART for tomographic PIV.
Experiments in Fluids,2016,57(5):1-19
|
CSCD被引
3
次
|
|
|
|
|
15.
Gesemann S. Recasting Tomo-PIV reconstruction as constrained and L1-regularized nonlinear least squares problem.
Proc of the 15th International Symposium on Applications of Laser Techniques to Fluid Mechanics,2010
|
CSCD被引
1
次
|
|
|
|
|
16.
Ye Z J. Dual-basis reconstruction techniques for tomographic PIV.
Science China Technological Sciences,2015,58(11):1963-1970
|
CSCD被引
1
次
|
|
|
|
|
17.
Bajpayee A. Fast volume reconstruction for 3DPIV.
Experiments in Fluids,2017,58(8):1-4
|
CSCD被引
1
次
|
|
|
|
|
18.
Ben Salah R. Tomographic reconstruction of 3Dobjects using marked point process framework.
Journal of Mathematical Imaging and Vision,2018,60(7):1132-1149
|
CSCD被引
1
次
|
|
|
|
|
19.
Cai S Z. Dense motion estimation of particle images via a convolutional neural network.
Experiments in Fluids,2019,60(4):1-16
|
CSCD被引
5
次
|
|
|
|
|
20.
Cai S Z. Particle image velocimetry based on a deep learning motion estimator.
IEEE Transactions on Instrumentation and Measurement,2020,69(6):3538-3554
|
CSCD被引
5
次
|
|
|
|
|
了解气象卫星更多信息，请访问