Particle Tracking Velocimetry Algorithm Research Articles

Improving depth uncertainty in plenoptic camera-based velocimetry

This work describes the development of a particle tracking velocimetry (PTV) algorithm designed to improve three-dimensional (3D), three-component velocity field measurements using a single plenoptic camera. Particular focus is on mitigating the longstanding depth uncertainty issues that have traditionally plagued plenoptic particle image velocimetry (PIV) experiments by leveraging the camera’s ability to generate multiple perspective views of a scene in order to assist both particle triangulation and tracking. 3D positions are first estimated via light field ray bundling (LFRB) whereby particle rays are projected into the measurement volume using image-to-object space mapping. Tracking is subsequently performed independently within each perspective view, providing a statistical amalgamation of each particle’s predicted motion through time in order to help guide 3D trajectory estimation while simultaneously protecting the tracking algorithm from physically unreasonable fluctuations in particle depth positions. A synthetic performance assessment revealed a reduction in the average depth errors obtained by LFRB as compared to the conventional multiplicative algebraic reconstruction technique when estimating particle locations. Further analysis using a synthetic vortex ring at a magnification of − 0.6 demonstrated plenoptic-PIV capable of maintaining the equivalent of 0.1–0.15 voxel accuracy in the depth domain at a spacing to displacement ratio of 5.3–10.5, an improvement of 84–89% compared to plenoptic-PIV. Experiments were conducted at a spacing to displacement ratio of approximately 5.8 to capture the 3D flow field around a rotor within the rotating reference frame. The resulting plenoptic-PIV/PTV vector fields were evaluated with reference to a fixed frame stereoscopic-PIV (stereo-PIV) validation experiment. A systematic depth-wise (radial) component of velocity directed toward the wingtip, consistent with observations from prior literature and stereo-PIV experiments, was captured by plenoptic-PTV at magnitudes similar to the validation data. In contrast, the plenoptic-PIV did not discern any coherent indication of radial motion. Our algorithm constitutes a significant advancement in enhancing the functionality and versatility of single-plenoptic camera flow diagnostics by directly addressing the primary limitation associated with plenoptic imaging.Graphical abstract

Measuring three-dimensional bubble dynamics for hydrogen production via water electrolysis

The commercialization of mass hydrogen production via water electrolysis is presently limited by low operational current densities. The optimal performance of electrolysis cells is significantly influenced by the substantial formation and residence of bubbles at high current densities. Thus, it is crucial to design electrodes with the ability for rapid bubble discharge to ensure appropriate bubble management. However, the quantitative volumetric measurements required to determine the bubble discharge ability of an electrode are not yet sufficiently accurate. This paper describes a quantitative volumetric method that combines a stereoscopic shadowgraph imaging system with particle tracking velocimetry (PTV) to measure the three-dimensional position, size, and velocity of micrometer-sized bubbles. The proposed method successfully captures hydrogen bubbles larger than 30 μm bubbles in an alkaline water electrolyzer. Considering the different luminance patterns of small bubbles (r ≤ 4 pixels) and large bubbles (r > 4 pixels) in the current imaging system, a bubble-size adaptive detection algorithm is established based on the cascade correlation method to obtain the two-dimensional centroid coordinates and radius of the observed bubbles. The bubble size information is also introduced into a two-view PTV algorithm for retrieving the Lagrangian trajectory of each bubble. Both the bubble detection and PTV algorithms are validated using synthetic datasets. Once the bubble trajectories are resolved successfully, the three-dimensional bubble velocity is obtained, and the actual bubble sizes are further corrected using the depth information. Analysis of the trajectory and velocity components indicates the existence of lateral bubble motion, reflecting interactions among bubbles. The rise velocity is positively correlated with the bubble radius in two regions, respectively, and the deviation from the theoretical value reveals the influence of non-buoyancy factors. The proposed technique provides effective diagnostics of the three-dimensional dynamic characteristics of micrometer-sized bubbles and can be used to evaluate and design bubble management systems for various electrochemical energy conversion devices.

ANDROMEDE — A software platform for optical surface velocity measurements

Methods and algorithms for measuring stream surface velocities have been continuously developed over the past five years to adapt to specific flow typologies. The free software ANDROMEDE allows easy use and comparison of these methods with image processing capabilities designed for measurements in natural environments and with unmanned aerial vehicles. Conventional particle tracking velocimetry, optical flow and particle image velocimetry algorithms, as well as filtering and statistical analysis methods, are available on a single platform. The code is written in Python, exploits available Python libraries, and allows accessible open-source code development. The validation of the integrated algorithms is presented on three case studies that represent the targeted applications: the study of currents for eco-hydraulics, the measurement of low water flows and the diagnosis of hydraulic structures. The field measurements are in very good agreement with the optical measurements and demonstrate the usefulness of the tool for rapid flow diagnosis for all the intended applications.

Development and assessment of a particle tracking velocimetry (PTV) measurement technique for the experimental investigation of oil drops behaviour in dispersed oil–water two-phase flow within a centrifugal pump impeller

The objective of the current work is to present the development of a Particle Tracking Velocimetry (PTV) algorithm for the analysis of oil drops behaviour in two-phase oil–water dispersions within a centrifugal pump impeller. The drop tracking was realized through high-speed camera images in a transparent pump prototype, which enabled the visualization of oil drops dispersed in water in all the impeller channels. The PTV algorithm is based on deep-learning techniques for image processing. The drops are detected by a combined U-Net and Convolutional Neural Network (CNN) method, with the former generating a binary image and the latter detecting valid oil drop contours. After detection, the Labelled Object Velocimetry (LOV) is adopted to calculate the instantaneous oil drop velocity. A synthetic image generator based on a Generative Adversarial Network (GAN) is then developed to assess the results from the U-Net, CNN, and LOV models. Additional validation studies are performed using the results from Perissinotto et al. (2019a). The results reveal that the presented deep-learning PTV algorithm is robust and provides consistent and reliable data for the dispersed oil phase in two-phase oil–water flows.

Experimental study of free drainage of mixed cubic and spherical particles in a monolayer granular system

The granular system of varied shapes is important in industrial production. Once the particles are deformed and crushed into parts, the characteristics of free discharge will become complicated and different from the spheres greatly. In this work, to further study the free-discharging characteristics of the mixed granular system for cubic and spherical particles, an unloading experiment of the mixed granular system is designed, with relevant experimental platforms. The algorithm for the particle recognition used in the particle tracking velocimetry (PTV) system mixed with cubic and spherical particles is applied. The effect of non-spherical particle fractions on the flow rate of discharge, residence time, and translational and rotational movement for non-spherical particles are analyzed based on PTV and the method of marked particles. It was found that a larger percentage of cubic particles has the effect of impeding particles' movement, especially in the near-wall region. The rotational movement of cubic particles becomes more intense and evident in the region below the middle of the silo. Number of contacted cubic particles during the discharge for cases when R cp0 = 10% and 15% (a), and some images during the discharge (b)(c)(d). • The discharge process of the mixed cubic and spherical particle system is analyzed. • A recognition and PTV algorithm for a mixed particle system is proposed. • Larger components of cubic particles impede the movement of the whole system. • Appropriate addition of cubic particles can locally promote the central discharge. • The rotational movement of a non-spherical particle is revealed by marked particles.

Physics-Driven PTV Interpolation Through Learned Regularization

This paper deals with spatial interpolation of instantaneous Particle Tracking Velocimetry (PTV) data. Such a task can be handled using generic signal processing tools like B-splines, radial basis functions (RBF) or Kriging. We deal with non time-resolved PTV as gathered by dual frame particle image velocimetry (PIV) systems followed by dual frame PTV algorithms. In such a dual frame context, Physics-based interpolation techniques typically rely on enforcement of divergence free constraints on the interpolated field. The aim of this communication is to explore the wider field of Physics-based PTV interpolators based on learning a deep neural network. These interpolators are learned on a database of flow examples that may be generated by high resolution PIV or DNS. We propose an instance of such a Physics-based PTV interpolator based on the machine learning concept of unrolled optimization. This concept is extended in order to deal with random measurement location, characteristic of PTV data. We also show how to use this concept to perform pressure estimation from PTV data, along with velocity interpolation. The concept is illustrated on a 2D laminar flow, DNS data is used to train and benchmark the proposed method against RBF interpolation. The proposed method is shown to be much more robust than RBF interpolation against low density seeding and noise.

Camera calibration algorithm development for a two-view collimated shadowgraph system

Shadowgraph imaging is a promising technique for volumetric velocity measurements, which features a high framing rate, long depth focus, and a cheap light source. The main objective of the current study is to develop a camera calibration algorithm for collimated shadowgraph systems, which is an essential procedure for 3D particle tracking velocimetry (PTV) strategies. First, the optical model of a two-view collimated shadowgraph system is established, which can be described by the orthographic projection model. The image distortion effect is also taken into consideration. Then, the calibration algorithm is developed using a flexible planar-target-based method. Aiming towards 3D PTV applications, the extrinsic parameters, including rotation and translation relationships between the two camera imaging coordinates, have been derived. The ambiguity around the sign confirmation of the extrinsic parameters has been solved by introducing extra information from the relative positions of the two views. Moreover, extrinsic parameters self-calibration (EPSC) has been implemented to deal with unavoidable camera drifts during the experiments. The results indicate that the EPSC is effective in removing the global system error in the current two-view system. The proposed calibration algorithm has been verified using synthetic images, which has shown a mean reprojection error of less than 0.1 pixels. In a water jet experiment, the mean reprojection error is around 0.3 pixels (about 0.019 mm in reality) after the board calibration. The relative error evaluated from the reconstruction points is less than 1%. The results indicate that the proposed calibration procedure is effective and feasible for collimated shadowgraph imaging systems. The 3D-particle positions of a sample frame have been reconstructed successfully. It is believed that the high quality shadowgraph images can offer high precision measurements for further implementations of 3D PTV algorithms.

A fast region homogenization method based on experimental data for pebble flow

Particle movement in pebble bed has important applications in engineering. A fast region homogenization method based on experimental silo discharging process is proposed in this paper. It can obtain the particle information from the particle motion image, and quickly calculate the homogenization information of the region. Hough transform is applied to the image of silo unloading process in order to obtain the information of particle centroid and radius. The particle tracking velocimetry (PTV) algorithm processes two continuous images to realize particle matching. Based on the trajectories of the particle, a fast region homogenization method is developed to obtain the homogenization information of the region. By analyzing the possible relative geometric relationships between mesh elements and particles, the hash set is constructed in advance to quickly calculate the regional homogenization parameters. Compared with the traditional integral method, the calculation speed is greatly accelerated on the premise of ensuring the calculation accuracy. Finally, the particle area characteristics (average velocity and filling rate) of the near wall area and the central area of the pebble bed are compared.

A conveyor belt experimental setup to study the internal dynamics of granular avalanches

This paper shows how a conveyor belt setup can be used to study the dynamics of stationary granular flows. To visualise the flow within the granular bulk and, in particular, determine its composition and the velocity field, we used the refractive index matching (RIM) technique combined with particle tracking velocimetry and coarse-graining algorithms. Implementing RIM posed varied technical, design and construction difficulties. To test the experimental setup and go beyond a mere proof of concept, we carried out granular flow experiments involving monodisperse and bidisperse borosilicate glass beads. These flows resulted in stationary avalanches with distinct regions whose structures were classified as: (i) a convective-bulged front, (ii) a compact-layered tail and, between them, (iii) a breaking size-segregation wave structure. We found that the bulk strain rate, represented by its tensor invariants, varied significantly between the identified flow structures, and their values supported the observed avalanche characteristics. The flow velocity fields’ interpolated profiles adjusted well to a Bagnold-like profile, although a considerable basal velocity slip was measured. We calculated a segregation flux using recent developments in particle-size segregation theory. Along with vertical velocity changes and high expansion rates, segregation fluxes were markedly higher at the avalanche’s leading edge, suggesting a connection between flow rheology and grain segregation. The experimental conveyor belt’s results showed the potential for further theoretical developments in rheology and segregation-coupled models.Graphic

Photogrammetry for Free Surface Flow Velocity Measurement: From Laboratory to Field Measurements

This study describes a multi-camera photogrammetric approach to measure the 3D velocity of free surface flow. The properties of the camera system and particle tracking velocimetry (PTV) algorithm were first investigated in a measurement of a laboratory open channel flow to prepare for field measurements. The in situ camera calibration methods corresponding to the two measurement situations were applied to mitigate the instability of the camera mechanism and camera geometry. There are two photogrammetry-based PTV algorithms presented in this study regarding different types of surface particles employed on the water flow. While the first algorithm uses the particle tracking method applied for individual particles, the second algorithm is based on correlation-based particle clustering tracking applied for clusters of small size particles. In the laboratory, reference data are provided by particle image velocimetry (PIV) and laser Doppler velocimetry (LDV). The differences in velocities measured by photogrammetry and PIV, photogrammetry and LDV are 0.1% and 3.6%, respectively. At a natural river, the change of discharges between two measurement times is found to be 15%, and the corresponding value reported regarding mass flow through a nearby hydropower plant is 20%. The outcomes reveal that the method can provide a reliable estimation of 3D surface velocity with sufficient accuracy.

An improved high accuracy PTV algorithm for pebble flow

The Particle Tracking Velocimetry (PTV) has a wide range of applications in the field of fluid and particle flow measurement. Due to the uncertainty caused by the optimal probability in the matching algorithm, the long-term tracking of multiple particles will eventually lead to lower accuracy with increasing time. In this study, a full-field high-accuracy PTV algorithm for pebble flow measurement is developed. Based on a two-dimensional pebble flow experiment, improved matching algorithms for pebble flow is proposed. Analyzing two kinds of errors in the experimental data—unidentified centroids and misidentified centroids, two amending methods are developed. The original PTV algorithm for pebble flow reached an average accuracy of 90% under 5 different flowrates. After adopting the three-point co-circular amending method, the accuracy is increased to 98%. Finally, this value exceeds 99% when the LP polygon amending method is applied. The adaptability and stability of the algorithm have also been tested.

A spatial-temporal algorithm for three-dimensional particle tracking velocimetry using two-view systems

In the present study, a novel temporal three-dimensional particle tracking velocimetry (3D PTV) algorithm for flow measurements with only two views is developed and validated with synthetic particles. The spatial information in image and object spaces, as well as the temporal predictions, are strongly coupled to improve the particle tracking accuracy. A well-designed cost function, simultaneously penalizing the reconstruction and tracking processes, is minimized to determine the most reliable traces. The algorithm shows a correctness over 98% up to 0.0273 ppp (particles per pixel) when using ideal synthetic particle positions, which is superior to several artificial intelligence methods. Moreover, an improved particle identification algorithm is proposed to handle overlapped particles and reduce the error introduced into the 3D PTV scheme. The algorithm adopts a particle position shifting process to tackle the correct particle numbers iteratively, which shows better performance than some other methods. A comparative study indicates that particle identification accuracy has a significant effect on the subsequent 3D reconstruction and tracking processes. The 3D PTV and particle identification algorithms show good consistencies under two types of flow conditions: a homogeneous isotropic turbulent flow and a vortex ring flow. Comparing with multiple-view setups, two-view systems are more compact and cost-effective, especially in conditions requiring high-speed cameras. With the newly established algorithms, a two-view system is now able to handle higher particle-seeding densities and thus can resolve higher spatial resolutions, which is significant for applications in turbulent flow and particle motion measurements.

A hybrid 3D particle matching algorithm based on ant colony optimization

Particle Tracking Velocimetry (PTV) is a popular optical method to measure the velocity field of complex flow at high spatial resolution. One of the practical limitations of this technique is that in general, the accuracy of matching individual particles between a pair of images is limited by the particle image displacement. To deal with this limitation, the present work proposed a hybrid ant colony optimization (ACO) algorithm for particle matching in three-dimensional (3D) scenarios . It can be regarded as an update of the conventional ACO PTV algorithm (Takagi J Vis 27:89–90. https://doi.org/10.3154/jvs.27.Supplement2_89 , 2007, Ohmi et al. Exp Fluids 48(4):589–605. https://doi.org/10.1007/s00348-009-0815-2 , 2010). The key concept is to seek a global solution of the minimization of a displacement-pattern function (DPF) via improved ant colony optimization (ACO). The object function, i.e., DPF, hybrids the measure of particle image displacement and the measure of pattern similarity as the particle matching criterion, the latter of which is constructed as the similarity level of Voronoi polygons (VPs) of paired particles. Performance evaluation was based on both the standard particle image database of Visualization Society of Japan (Okamoto et al. Meas Sci Technol 11(6):685–691. https://doi.org/10.1088/0957-0233/11/6/311 , 2000) and the laboratory-made synthetic flow. It was shown that this hybrid ACO algorithm has higher matching accuracy than those of exiting ACO methods based on either minimum displacement function or relaxation function. Its credibility in dealing with the scenarios of large relative particle displacement, i.e., the cases where particle image displacement is comparable to or even larger than the mean spacing of neighboring particles was also empirically demonstrated. Other features including fast convergence speed and regular pattern of outliers were also seen. All these make this algorithm a suitable candidate for 3D particle matching in PTV.

Performance comparison of particle tracking velocimetry (PTV) and particle image velocimetry (PIV) with long-exposure particle streaks

Particle tracking velocimetry (PTV) and particle image velocimetry (PIV) are popular experimental methods to quantitatively measure flow fields. In many practical applications, hardware limitations result in longer exposure times, causing particle images to elongate into particle streak images. In this study, the performances of PTV and PIV in relation to particle streak images are evaluated systematically by means of both synthetic and experimental images. For the synthetic images, particle streak images are created via the integration over time of the standard Gaussian approximation of particle images, plus the effects of exposure time, and parameters such as particle image diameter, intensity, and density were investigated in terms of the velocity fields of a 1D uniform flow and the rotational 2D Hill’s vortex. The results show that PTV performs well for short exposure times, and its peak-finding criteria can be verified for nondimensional exposure times up to 10. As the increases from 10 to 70, the reliability of the PTV algorithm drops significantly, while the yield drops only slightly. Longer exposure times cause an increase in the number of RMS displacement errors; the PTV algorithm is more likely to fail when particle diameter, image intensity and particle density are larger. The displacement error of PTV RMS ranges from 0.064 pixels to 0.157 pixels. In comparison to PTV, PIV is more robust in relation to streak images with long-exposure times, and not sensitive to the effects of particle image diameter and intensity. The RMS dispersion errors for PIV range from 0.015 pixels to 0.316 pixels. A comparison of the RMS displacement errors exhibited by PIV and PTV shows that where the seeding of particles is not an issue, PIV is more robust, and can handle long-exposure streak images. The results of experimental streak images show that PIV produces lower RMS dispersion error values at low Reynolds numbers, whereas it produces more significant errors for high Reynolds numbers. At regions of low seeding density, PTV can resolve local fluid motions with superior accuracy, but it is vulnerable to high exposure time. Based on these results, where a longer exposure time is a requirement for an application, the use of PIV is recommended, by virtue of its robustness to image streaking, provided that a good seeding and high particle image density are available. For applications such as some microscale flows or flows with separation or recirculation regions, PTV is more capable of fulfilling their requirements.

Particle velocity measurement of binary mixtures in the riser of a circulating fluidized bed by the combined use of electrostatic sensing and high-speed imaging

The flow dynamics of binary particle mixtures in the fluidized bed needs to be monitored in order to optimize the related industrial processes. In this paper, electrostatic sensing and high-speed imaging are applied to measure the velocities of polyethylene and sand particles in the binary particle mixtures in fluidization. Experimental studies were conducted on a lab-scale cold circulating fluidized bed. Correlation function between electrostatic signals from upstream and downstream electrodes placed along the riser shows two peaks that represent transit times for the two types of particles. To verify the above results, high-speed imaging was adopted to capture the flow images of particle mixtures. Particle Image Velocimetry and Particle Tracking Velocimetry algorithms were utilized to process the resulted images in order to measure the velocities of polyethylene and sand particles. The reasons for two-peak correlation functions are illustrated based on the frequency spectrums of the mono-solid-phase electrostatic signals and the velocity difference between polyethylene and sand particles. Finally, comparisons on the velocities obtained from electrostatic sensing and high-speed imaging demonstrate the electrostatic sensor can roughly estimate the particle velocity of binary particle mixtures in the near wall region of the circulating fluidized bed.

Deep Learning-Based Super-resolution Ultrasound Speckle Tracking Velocimetry

Deep ultrasound localization microscopy (deep-ULM) allows sub-wavelength resolution imaging with deep learning. However, the injection of contrast agents (CAs) in deep-ULM is debatable because of their potential risk. In this study, we propose a deep learning-based super-resolution ultrasound (DL-SRU), which employs the concept of deep-ULM and a convolutional neural network. The network is trained with synthetic tracer images to localize positions of red blood cells (RBCs) and reconstruct vessel geometry at high resolution, even for CA-free ultrasound (US) images. The proposed algorithm is validated by comparing the full width at half-maximum values of the vascular profiles reconstructed by other techniques, such as the standard ULM and the US average intensity under in silico and in vitro conditions. RBC localization by DL-SRU is also compared with that by other localization approaches to validate its performance under in vivo condition, especially for veins in the human lower extremity. Furthermore, a two-frame particle tracking velocimetry (PTV) algorithm is applied to DL-SRU localization for accurate flow velocity measurement. The velocity profile obtained by applying the PTV is compared with a theoretical value under in vitro condition to verify its compatibility with the flow measurement modality. The velocity vectors of individual RBCs are obtained to determine the applicability to in vivo conditions. DL-SRU can achieve high-resolution vessel morphology and flow dynamics in vasculature, mapping 110 super-resolved images per second on a standard PC, regardless of various imaging conditions. As a result, the DL-SRU technique is much more robust in localization compared with previous deep-ULM. In addition, the performance of DL-SRU is nearly the same as that of deep-ULM in rapid computational processing and high measurement accuracy. Thus, DL-SRU might become an effective and useful instrument in clinical practice.

Applications of a robust particle tracking velocimetry algorithm in the multiple-parametric measuring system for coal exploration industry

With the gradual increase in the average exploration depth of domestic coal industry, the conflict between deep coal exploration technology and exploration requirements is becoming increasingly prominent. A series of exploration disaster accidents caused by geological disasters require a high level of survey technology before the deep coal exploration. In this study, a high-resolution single hole measuring system for aquifers in coal seam is proposed. The entire system can be divided into two parts: hardware and software. The hardware probe integrates various sensors for different physical parameters using optical fibre to communicate with the controlling part above the ground. The home-made software is then used to receive and process the signals of temperature, pressure and orientation, as well as the instantaneous image sequence of aquifer with flowing contaminants. A home-made novel particle tracking velocimetry algorithm based on Voronoi diagram is used as the core algorithm to analyse the image sequence and obtain the magnitude and direction of flow at the measuring spot of aquifer. After practical tests, the system is confirmed to be able to effectively monitor the hydrological conditions of the aquifer in coal seam. Therefore, the proposed algorithm can be applied to the evaluation of hydrogeological conditions before well establishment for coal-exploration and to the prevention of potential dangers during the exploration work.

A microscopic experimental study of nanoparticle motion for the enhancement of oxygen absorption in nanofluids

Abstract The behavior of nanoparticle motion has a great influence on gas-liquid mass transfer. However, it has been very difficult to characterize the motion of nanoparticles from a micro view in mass transfer experiments. In this study, a novel method was proposed to investigate nanoparticle Brownian motion through the application of the total internal reflection fluorescence microscope in a self-designed sample (a quasi-static liquid micro-groove) and the mass transfer enhancement of nanoparticles. Nanoparticle movement behavior was photographed using an electron-multiplying charge coupled device, and 100 consecutive images were recorded using Micro-Manager software at a rate of 20 fps. The images were processed through the particle tracking velocimetry algorithm to calculate two-dimensional motion rates of nanoparticles caused by Brownian movement. It showed that nanoparticle loadings influenced the motion rates significantly, and the motion rates were larger with smaller particle sizes under the same operating condition. The mass transfer coefficients in the quasi-static gas-liquid mass transfer system were calculated and analyzed through microscopic measurement. Based on the above thought, three important non-dimensional numbers [Sherwood (Shp ), Reynolds (Rep ), and Schmidt (Scp ) numbers] for mass transfer theory were studied.

Exploring the optimal experimental setup for surface flow velocity measurements using PTV.

Advances in flow monitoring are crucial to increase our knowledge on basin hydrology and to understand the interactions between flow dynamics and infrastructures. In this context, image processing offers great potential for hydraulic monitoring, allowing acquisition of a wide range of measurements with high spatial resolution at relatively low costs. In particular, the particle tracking velocimetry (PTV) algorithm can be used to describe the dynamics of surface flow velocity in both space and time using fixed cameras or unmanned aerial systems (UASs). In this study, analyses allowed exploration of the optimal particle seeding density and frame rate in different configurations. Numerical results provided useful indications for two field experiments that have been carried out with a low-cost quadrocopter equipped with an optical camera to record RGB videos of floating tracers manually distributed over the water surface. Field measurements have been carried out using different natural tracers under diverse hydraulic and morphological conditions; PTV's processed velocities have been subsequently benchmarked with current meter measurements. The numerical results allowed rapid identification of the experimental configuration (e.g., required particle seeding density, image resolution, particle size, and frame frequency) producing flow velocity fields with high resolution in time and space with good agreement with the benchmark velocity values measured with conventional instruments.

Image processing techniques for the measurement of two-phase bubbly pipe flows using particle image and tracking velocimetry (PIV/PTV)

Particle Image Velocimetry (PIV) measurement in gas-liquid bubbly flows is a challenging task, mainly due to the dispersion of the laser light caused by the gas-liquid interfaces. A common solution adopted is the use of fluorescent seeding particles associated with a bandpass filter for the laser light. Therefore, the camera captures only the light fluoresced by the seeding particles, filtering the laser light dispersed by the gas-liquid interfaces and measuring a velocity field which corresponds to the liquid phase, seeded with the particles. However, even for relatively low gas fractions, the fluoresced light reflected by the interfaces distorts the measurements, as it illuminates particles out from the laser plane. In addition, the fluoresced light also reflects at the interfaces, distorting the measurements and trending to overshoot the measured liquid velocities In this work, PIV measurements of air-water bubbly flows in a small diameter pipe (D = 26.2 mm) are performed with fluorescent tracer particles (PIV/LIF). A new method for the phase discrimination in PIV was developed to overcome the problems caused by the presence of bubbles in the flow, which uses the pixels intensity information of each interrogation window, to identify if that window corresponds to liquid or bubble region. To validate the PIV procedure, a Particle Tracking Velocimetry (PTV) algorithm was also developed to measure the bubbles velocities, based on similar implementations found in literature, but some corrections were also proposed to overcome the bubble overlap phenomena which arises in bubbly pipe flows, when backward illumination is used. The PIV and the PTV methods were tested and validated by a series of distinct gas void fraction and bubble diameters experimental cases, confirming the accuracy and reliability of both methods. The proposed method was used to analyze a set of upward laminar and turbulent bubbly flows, showing that the averaged axial velocity and the dispersed gas bubbles modify the turbulence intensity profiles.