Three-dimensional particle tracking velocimetry using shallow neural network for real-time analysis

Gim, Yeonghyeon; Jang, Dong Kyu; Sohn, Dong Kee; Kim, Hyoungsoo; Ko, Han Seo

doi:10.1007/s00348-019-2861-8

Cited by 29 publications

(16 citation statements)

References 25 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The first category aims to enhance the reconstruction of particles in the classic multi-camera arrangement used in tomographic PIV and LPT. Gim et al [56] proposed the use of a shallow neural network to reconstruct 3D particle positions for particle tracking instead of performing traditional line-of-sight-based particle triangulation. The main advantage ise the reduced computational cost.…”

Section: Three-dimensional Piv With Nnsmentioning

confidence: 99%

Machine learning for flow field measurements: a perspective

Discetti¹,

Liu²

2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

Advancements in machine-learning (ML) techniques are driving a paradigm shift in image processing. Flow diagnostics with optical techniques is not an exception. Considering the existing and foreseeable disruptive developments in flow field measurement techniques, we elaborate this perspective, particularly focused to the field of particle image velocimetry. The driving forces for the advancements in ML methods for flow field measurements in recent years are reviewed in terms of image preprocessing, data treatment and conditioning. Finally, possible routes for further developments are highlighted.

show abstract

Section: Three-dimensional Piv With Nnsmentioning

confidence: 99%

Machine learning for flow field measurements: a perspective

Discetti¹,

Liu²

2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…In these processes, the flows are inherently 3D and unsteady, mostly highly turbulent on a large temporal-spatial dynamic scale. For an in-depth insight into these complex flow structures, it is necessary to develop an advanced flow diagnostic technique for instantaneous 3D flow velocity measurement [3][4][5][6][7]. In most industrial processes, the optical access to the measurement area of internal flows is limited due to the confined space, high temperature and pressure conditions.…”

Section: Introductionmentioning

confidence: 99%

Approach to select optimal cross-correlation parameters for light field particle image velocimetry

Zhu

Hossain

et al. 2022

Physics of Fluids

View full text Add to dashboard Cite

The light field particle image velocimetry (LF-PIV) has shown a great potential for three-dimensional (3D) flow measurement in space-constrained applications. Usually, the parameters of the cross-correlation calculation in the LF-PIV are chosen based on empirical analysis or introduced from conventional planar PIV, which lowers the accuracy of 3D velocity field measurement. This study presents an approach to selecting optimal parameters of the cross-correlation calculation and thereby offers systematic guidelines for experiments. The selection criterion of the interrogation volume size is studied based on the analysis of the valid detection probability of the correlation peak. The optimal seeding concentration and the size of tracer particles are then explored through synthetic Gaussian vortex field reconstruction. The optimized parameters are employed in a cylinder wake flow measurement in a confined channel. A comparative study is conducted between the LF-PIV and a planar PIV system. Results indicate that the LF-PIV along with the optimized parameters can measure the 3D flow velocity of the cylinder wakes accurately. It has been observed that the mean and max errors of velocity decrease by 32.6% and 18.8%, respectively compared to the related LF-PIV techniques without consideration of optimal parameters. Therefore, it is suggested that the optimized cross-correlation parameters in the LF-PIV can improve the accuracy of 3D flow measurement.

show abstract

“…В результате обзора литературы по методам машинного обучения для задачи измерения формы поверхности была выбрана нейросеть PIV-LiteFlowNet-en [10], которая основана на сети LiteFlowNet [11]. Обе эти сети являются потомками сети FlowNet [12], предназначенной для приема двух изображений на вход и оценки смещения оптического потока на выходе.…”

Section: Introductionunclassified

Exploring the Application of Convolutional Neural Networks for Photogrammetric Image Processing

Pinchukov

Poroykov

Shmatko

2022

Proceedings of the 32nd International Conference on Computer Graphics and Vision

View full text Add to dashboard Cite

Close-range photogrammetry is widely used to measure the surface shape of various objects and its deformations. Usually, a stereo pair of images of the object under study, obtained from different angles by means of several digital video cameras, is used for this purpose. The surface shape is measured by triangulating a set of corresponding two-dimensional points from these images using a predetermined location of cameras relative to each other. Various algorithms are used to find these points. Several photogrammetric methods use cross-correlation for this purpose. This paper discusses the possibility of replacing the correlation algorithm with neural networks to determine offsets in the images. They allow to increase the calculation speed and the spatial resolution of the measurement results. To verify the possibility of their application, a series of experimental images of surfaces with different deformations were obtained. Computational experiments were performed to process these images using selected neural networks and a classical cross-correlation algorithm. The limitations on the use of the compared algorithms were determined and their error in restoring the three-dimensional shape of the surface was estimated. The physical simulation to verify the selected neural networks for image processing for the task of photogrammetry showed their performance and efficiency.

show abstract

Three-dimensional particle tracking velocimetry using shallow neural network for real-time analysis

Cited by 29 publications

References 25 publications

Machine learning for flow field measurements: a perspective

Machine learning for flow field measurements: a perspective

Approach to select optimal cross-correlation parameters for light field particle image velocimetry

Exploring the Application of Convolutional Neural Networks for Photogrammetric Image Processing

Contact Info

Product

Resources

About