Unsupervised learning about 4D features of microparticle motion

Wolfe, Bradley T.; Iaroshenko, Oleksandr; Chu, Pinghan; Kenyon, Garrett T.; Schaub, Terrance J.; Thresher, Austin Morgan; Watkins, Yijing Z.; Zhao, Lei; Wang, Zhehui

doi:10.1063/1.5037462

Title: Unsupervised learning about 4D features of microparticle motion

Abstract

Material clusters of different sizes are known to exist in high-temperature plasmas due to plasma-wall interactions. The facts that these clusters, ranging from sub-microns to above mm in size, can move from one location to another quickly and that there are a lot of them make high-speed imaging and tracking one of the best, effective, and sometimes only diagnostic. An unsupervised machine learning technique based on deconvolutional neural networks is developed to analyze two-camera videos of high-temperature microparticles generated from exploding wires. The neural network utilizes a locally competitive algorithm to infer representations and optimize a dictionary composed of kernels, or basis vectors, for image analysis. Our primary goal is to use this method for feature recognition and prediction of the time-dependent three-dimensional (or “4D”) microparticle motion. Features equivalent to local velocity vectors have been identified as the dictionary kernels or “building blocks” of the scene. The dictionary elements from the left and right camera views are found to be strongly correlated and satisfy the projection geometrical constraints. The results show that unsupervised machine learning techniques are promising approaches to process large sets of images for high-temperature plasmas and other scientific experiments. In conclusion, machine learning techniques can be usefulmore »« less

Authors:

^[1];

^[1]; Schaub, Terrance J. ^[1];

^[1];

^[1]; Zhao, Lei ^[1];

^[1]

Los Alamos National Lab. (LANL), Los Alamos, NM (United States)

Publication Date:: Mon Oct 01 00:00:00 EDT 2018

Research Org.:: Los Alamos National Lab. (LANL), Los Alamos, NM (United States)

Sponsoring Org.:: USDOE Office of Science (SC), Fusion Energy Sciences (FES)

OSTI Identifier:: 1514979

Report Number(s):: LA-UR-18-23510
Journal ID: ISSN 0034-6748

Grant/Contract Number:: 89233218CNA000001

Resource Type:: Accepted Manuscript

Journal Name:: Review of Scientific Instruments

Additional Journal Information:: Journal Volume: 89; Journal Issue: 10; Conference: High temperature plasma diagnostics, San Diego, CA (United States), 16 Apr 2018; Journal ID: ISSN 0034-6748

Publisher:: American Institute of Physics (AIP)

Country of Publication:: United States

Language:: English

Subject:: 47 OTHER INSTRUMENTATION; Magnetic Fusion Energy; Machine Learning; Camera

Citation Formats


                    Wolfe, Bradley T., Iaroshenko, Oleksandr, Chu, Pinghan, Kenyon, Garrett T., Schaub, Terrance J., Thresher, Austin Morgan, Watkins, Yijing Z., Zhao, Lei, and Wang, Zhehui. Unsupervised learning about 4D features of microparticle motion.  United States: N. p., 2018. 
Web.  doi:10.1063/1.5037462.

Copy to clipboard


                    Wolfe, Bradley T., Iaroshenko, Oleksandr, Chu, Pinghan, Kenyon, Garrett T., Schaub, Terrance J., Thresher, Austin Morgan, Watkins, Yijing Z., Zhao, Lei, & Wang, Zhehui. Unsupervised learning about 4D features of microparticle motion.  United States.  https://doi.org/10.1063/1.5037462

Copy to clipboard


                    Wolfe, Bradley T., Iaroshenko, Oleksandr, Chu, Pinghan, Kenyon, Garrett T., Schaub, Terrance J., Thresher, Austin Morgan, Watkins, Yijing Z., Zhao, Lei, and Wang, Zhehui. Mon .  
"Unsupervised learning about 4D features of microparticle motion".  United States.  https://doi.org/10.1063/1.5037462.  https://www.osti.gov/servlets/purl/1514979.

Copy to clipboard


                    
@article{osti_1514979,

  title        = {Unsupervised learning about 4D features of microparticle motion},

  author       = {Wolfe, Bradley T. and Iaroshenko, Oleksandr and Chu, Pinghan and Kenyon, Garrett T. and Schaub, Terrance J. and Thresher, Austin Morgan and Watkins, Yijing Z. and Zhao, Lei and Wang, Zhehui},

  abstractNote = {Material clusters of different sizes are known to exist in high-temperature plasmas due to plasma-wall interactions. The facts that these clusters, ranging from sub-microns to above mm in size, can move from one location to another quickly and that there are a lot of them make high-speed imaging and tracking one of the best, effective, and sometimes only diagnostic. An unsupervised machine learning technique based on deconvolutional neural networks is developed to analyze two-camera videos of high-temperature microparticles generated from exploding wires. The neural network utilizes a locally competitive algorithm to infer representations and optimize a dictionary composed of kernels, or basis vectors, for image analysis. Our primary goal is to use this method for feature recognition and prediction of the time-dependent three-dimensional (or “4D”) microparticle motion. Features equivalent to local velocity vectors have been identified as the dictionary kernels or “building blocks” of the scene. The dictionary elements from the left and right camera views are found to be strongly correlated and satisfy the projection geometrical constraints. The results show that unsupervised machine learning techniques are promising approaches to process large sets of images for high-temperature plasmas and other scientific experiments. In conclusion, machine learning techniques can be useful to handle the large amount of data and therefore aid the understanding of plasma-wall interaction.},

  doi          = {10.1063/1.5037462},

  journal      = {Review of Scientific Instruments},

  number       = 10,

  volume       = 89,

  place        = {United States},

  year         = {Mon Oct 01 00:00:00 EDT 2018},

  month        = {Mon Oct 01 00:00:00 EDT 2018}

}

Copy to clipboard

Journal Article:

Free Publicly Available Full Text

Accepted Manuscript (DOE)

Publisher's Version of Record

https://doi.org/10.1063/1.5037462

Other availability

Search WorldCat to find libraries that may hold this journal

Citation Metrics:

Cited by: 1 work

Citation information provided by
Web of Science

Save / Share:

Export Metadata

Save to My Library

Works referenced in this record:

Sparse coding of sensory inputs
journal, August 2004

Olshausen, B.; Field, D.
Current Opinion in Neurobiology, Vol. 14, Issue 4
DOI: 10.1016/j.conb.2004.07.007

A compressed sensing X-ray camera with a multilayer architecture
journal, January 2018

Wang, Zhehui; Iaroshenko, O.; Li, S.
Journal of Instrumentation, Vol. 13, Issue 01
DOI: 10.1088/1748-0221/13/01/c01035

Highly resolved measurements of dust motion in the sheath boundary of magnetized plasmas
journal, September 2015

Shalpegin, A.; Brochard, F.; Ratynskaia, S.
Nuclear Fusion, Vol. 55, Issue 11
DOI: 10.1088/0029-5515/55/11/112001

Sparse Coding via Thresholding and Local Competition in Neural Circuits
journal, October 2008

Rozell, Christopher J.; Johnson, Don H.; Baraniuk, Richard G.
Neural Computation, Vol. 20, Issue 10
DOI: 10.1162/neco.2008.03-07-486

Dust creation and transport in MAST
journal, September 2010

De Temmerman, G.; Bacharis, M.; Dowling, J.
Nuclear Fusion, Vol. 50, Issue 10
DOI: 10.1088/0029-5515/50/10/105012

Four-dimensional (4D) tracking of high-temperature microparticles
journal, July 2016

Wang, Zhehui; Liu, Q.; Waganaar, W.
Review of Scientific Instruments, Vol. 87, Issue 11
DOI: 10.1063/1.4955280

Highly overcomplete sparse coding
conference, March 2013

Olshausen, Bruno A.
IS&T/SPIE Electronic Imaging, SPIE Proceedings
DOI: 10.1117/12.2013504

Emergence of simple-cell receptive field properties by learning a sparse code for natural images
journal, June 1996

Olshausen, Bruno A.; Field, David J.
Nature, Vol. 381, Issue 6583
DOI: 10.1038/381607a0

Gradient-based learning applied to document recognition
journal, January 1998

Lecun, Y.; Bottou, L.; Bengio, Y.
Proceedings of the IEEE, Vol. 86, Issue 11
DOI: 10.1109/5.726791

Similar Records in DOE PAGES and OSTI.GOV collections:

Four-dimensional (4D) tracking of high-temperature microparticles

Journal Article Wang, Zhehui ; Liu, Qiuguang ; Waganaar, Bill ; ... - Review of Scientific Instruments

High-speed tracking of hot and molten microparticles in motion provides rich information about burning plasmas in magnetic fusion. An exploding-wire apparatus is used to produce moving high-temperature metallic microparticles and to develop four-dimensional (4D) or time-resolved 3D particle tracking techniques. The pinhole camera model and algorithms developed for computer vision are used for scene calibration and 4D reconstructions. 3D positions and velocities are then derived for different microparticles. As a result, velocity resolution approaches 0.1 m/s by using the local constant velocity approximation.
Cited by 3
https://doi.org/10.1063/1.4955280

Full Text Available
Four-dimensional (4D) tracking of high-temperature microparticles

Journal Article Wang, Zhehui ; Liu, Q. ; Waganaar, W. ; ... - Review of Scientific Instruments

High-speed tracking of hot and molten microparticles in motion provides rich information about burning plasmas in magnetic fusion. An exploding-wire apparatus is used to produce moving high-temperature metallic microparticles and to develop four-dimensional (4D) or time-resolved 3D particle tracking techniques. The pinhole camera model and algorithms developed for computer vision are used for scene calibration and 4D reconstructions. 3D positions and velocities are then derived for different microparticles. Velocity resolution approaches 0.1 m/s by using the local constant velocity approximation.
https://doi.org/10.1063/1.4955280
Application of unsupervised deep learning to image segmentation and in-situ contact angle measurements in a CO₂-water-rock system

Journal Article Wang, Hongsheng ; Dalton, Laura ; Guo, Ruichang ; ... - Advances in Water Resources

Rock surface wettability is a critical property that regulates multiphase flows in porous media, which can be quantified using the surface contact angle (CA). X-ray micro-computed tomography (μCT) provides an effective approach to in-situ measurements of surface CAs. However, the CA measurement accuracy depends significantly on the quality of CT image segmentation, which is the clustering of CT pixels into separate phases. Inspired by this, we developed a deep learning (DL)-based CA measurement workflow. Motivated by the recent tremendous progress in unsupervised learning techniques and aiming to avoid expensive manual data annotations, an unsupervised DL pipeline for CT image segmentationmore »« less
https://doi.org/10.1016/j.advwatres.2023.104385

Full Text Available
Multistage neural network model for dynamic scene analysis

Miscellaneous Ajjimarangsee, P

This research is concerned with dynamic scene analysis. The goal of scene analysis is to recognize objects and have a meaningful interpretation of the scene from which images are obtained. The task of the dynamic scene analysis process generally consists of region identification, motion analysis and object recognition. The objective of this research is to develop clustering algorithms using neural network approach and to investigate a multi-stage neural network model for region identification and motion analysis. The research is separated into three parts. First, a clustering algorithm using Kohonens' self-organizing feature map network is developed to be capable of generatingmore »« less
Reification of latent microstructures: On supervised unsupervised and semi-supervised deep learning applications for microstructures in materials informatics

Technical Report Tran, Anh ; Rodgers, Theron ; Wildey, Timothy Michael

Machine learning (ML), including deep learning (DL), has become increasingly popular in the last few years due to its continually outstanding performance. In this context, we apply machine learning techniques to "learn" the microstructure using both supervised and unsupervised DL techniques. In particular, we focus (1) on the localization problem bridging (micro)structure (localized) property using supervised DL and (2) on the microstructure reconstruction problem in latent space using unsupervised DL. The goal of supervised and semi-supervised DL is to replace crystal plasticity finite element model (CPFEM) that maps from (micro)structure (localized) property, and implicitly the (micro)structure (homogenized) property relationships, whilemore »« less
https://doi.org/10.2172/1673174

Full Text Available

Similar Records

Title: Unsupervised learning about 4D features of microparticle motion

Abstract

Citation Formats

Sparse coding of sensory inputs journal, August 2004

A compressed sensing X-ray camera with a multilayer architecture journal, January 2018

Highly resolved measurements of dust motion in the sheath boundary of magnetized plasmas journal, September 2015

Sparse Coding via Thresholding and Local Competition in Neural Circuits journal, October 2008

Dust creation and transport in MAST journal, September 2010

Four-dimensional (4D) tracking of high-temperature microparticles journal, July 2016

Highly overcomplete sparse coding conference, March 2013

Emergence of simple-cell receptive field properties by learning a sparse code for natural images journal, June 1996

Gradient-based learning applied to document recognition journal, January 1998

Sparse coding of sensory inputs
journal, August 2004

A compressed sensing X-ray camera with a multilayer architecture
journal, January 2018

Highly resolved measurements of dust motion in the sheath boundary of magnetized plasmas
journal, September 2015

Sparse Coding via Thresholding and Local Competition in Neural Circuits
journal, October 2008

Dust creation and transport in MAST
journal, September 2010

Four-dimensional (4D) tracking of high-temperature microparticles
journal, July 2016

Highly overcomplete sparse coding
conference, March 2013

Emergence of simple-cell receptive field properties by learning a sparse code for natural images
journal, June 1996

Gradient-based learning applied to document recognition
journal, January 1998