PIV for granular flows

 Particle image velocimetry (PIV) has been adapted for use in measuring particle displacement and velocity fields in granular flows. “Seeding” is achieved by using light and dark particles. The granular flow adjacent to a clear bounding wall is illuminated with a strobe, and the recorded images are analyzed using standard PIV techniques. The application is demonstrated by measuring convection rolls in a granular bed undergoing vertical oscillations. The PIV measured displacement is consistent with displacement of a marked layer of particles. Received: 29 January 1998/Accepted: 8 April 1999     

Intensity Capping: a simple method to improve cross-correlation PIV results

A common source of error in particle image velocimetry (PIV) is the presence of bright spots within the images. These bright spots are characterized by grayscale intensities much greater than the mean intensity of the image and are typically generated by intense scattering from seed particles. The displacement of bright spots can dominate the cross-correlation calculation within an interrogation window, and may thereby bias the resulting velocity vector. An efficient and easy-to-implement image-enhancement procedure is described to improve PIV results when bright spots are present. The procedure, called Intensity Capping, imposes a user-specified upper limit to the grayscale intensity of the images. The displacement calculation then better represents the displacement of all particles in an interrogation window and the bias due to bright spots is reduced. Four PIV codes and a large set of experimental and simulated images were used to evaluate the performance of Intensity Capping. The results indicate that Intensity Capping can significantly increase the number of valid vectors from experimental image pairs and reduce displacement error in the analysis of simulated images. A comparison with other PIV image-enhancement techniques shows that Intensity Capping offers competitive performance, low computational cost, ease of implementation, and minimal modification to the images.     

The influence of peak-locking errors on turbulence statistics computed from PIV ensembles

The influence of peak-locking errors on turbulence statistics computed from ensembles of PIV data is considered. PIV measurements are made in the streamwise–wall-normal plane of turbulent channel flow. The PIV images are interrogated in three distinct ways, generating ensembles of velocity fields with absolute, moderate, and minimal peak locking. Turbulence statistics computed for all three ensembles of data indicate a general sensitivity to peak locking in the single-point statistics, except for the mean velocity profile. Peak-locking errors propagate into the fluctuations of velocity, rendering single-point statistics inaccurate when severe peak locking is present. Multi-point correlations of both streamwise and wall-normal velocity are also found to be influenced by severe levels of peak locking. The displacement range of the measurement, defined by the PIV time delay, appears to affect the influence of peak-locking errors on turbulence statistics. Smaller displacement ranges, particularly those that produce displacement fluctuations that are less than one pixel in magnitude, yield inaccurate turbulence statistics in the presence of peak locking.     

A kilohertz frame rate cinemagraphic PIV system for laboratory-scale turbulent and unsteady flows

A kilohertz frame rate cinemagraphic particle image velocimetry (PIV) system has been developed for acquiring time-resolved image sequences of laboratory-scale gas and liquid-phase turbulent flows. Up to 8000 instantaneous PIV images per second are obtained, with sequence lengths exceeding 4000 images. The two-frame cross-correlation method employed precludes directional ambiguity and has a higher signal-to-noise ratio than single-frame autocorrelation or cross-correlation methods, facilitating acquisition of long uninterrupted sequences of valid PIV images. Low and high velocities can be measured simultaneously with similar accuracy by adaptively cross-correlating images with the appropriate time delay. Seed particle illumination is provided by two frequency-doubled Nd:YAG lasers producing Q-switched pulses at the camera frame rate. PIV images are acquired using a 16 mm high-speed rotating prism camera. Frame-to-frame registration is accomplished by imaging two pairs of crossed lines onto each frame and aligning the digitized image sequence to these markers using image processing algorithms. No flow disturbance is created by the markers because only their image is projected to the PIV imaging plane, with the physical projection device residing outside the flow field. The frame-to-frame alignment uncertainty contributes 2% to the overall velocity measurement uncertainty, which is otherwise comparable to similar film-based PIV methods. Received: 11 July 2000 / Accepted: 21 June 2001 Published online: 29 November 2001     

An extension of digital PIV-processing to double-exposed images

 In this paper digital processing techniques for PIV (Partical Image Velocimetry) using double-exposed particle images have been studied. It has been found that a pattern matching technique is significantly superior to the traditional autocorrelation method in the case that a large particle displacement between the double exposures is present on the image. In PIV using double-exposed images, the image shifting technique is usually used to solve the directional ambiguity problem. The performance of PIV using autocorrelation technique is dependent on the flow speed and the amount of image shift applied. This dependence, for example, causes a difficulty of autocorrelation in flows close to a solid boundary. The present study shows that a pattern matching technique eliminates such a difficulty. At the same signal-to-noise ratio, the pattern matching techndique has a better spatial resolution than that of autocorrelation. In concert with the pattern matching technique, PID (Particle Image Distortion) can be applied to double-exposed images, further improving the reliability and accuracy of velocity estimates of PIV in the presence of large velocity gradients. Generally speaking, PIP-matching and PID extend the validity of PIV using double-exposed images. The total processing time required by the PIV using the pattern matching technique and one PID iteration is of the same order as that required by the PIV using autocorrelation. Received: 7 July 1995 / Accepted: 11 September 1997     

Two-color digital PIV employing a single CCD camera

 An extension of two color particle image velocimetry (PIV) is described where the color images are recorded onto a single high-resolution (3060×2036 pixel) color CCD sensor. Unlike mono-color CCD sensors, this system not only eliminates the processing time and the subsequent digitization time of film-based PIV but also resolves the directional ambiguity of the velocity vector without using conventional image-shifting techniques. For comparing the spatial resolutions of film and CCD data, a calibration experiment is conducted by recording the speckle pattern onto 35 mm color film and using a CCD sensor under identical conditions. This technique has been successfully implemented for simulated turbine film-cooling flows in order to obtain a more detailed characterization of the coolant-injection phenomenon and its interaction with freestream disturbances. Received: 20 November 1996/Accepted: 29 January 1998     

On the accuracy of a MJPEG-based digital image compression PIV-system

 In this communication, the Digital Image Compression (DIC) – PIV system is introduced. The present system allows the measurement of mean and RMS velocities in turbulent flow fields, using JPEG digital image compression technique for on-line recording of thousands of images. The decompression and subsequent analysis of the images, performed by means of digital cross-correlation technique, is carried out off-line. Errors incurred by the application of the compression method are assessed and discussed. The effect of the compression is firstly analysed by linearly traversing (synthetic) computer-generated PIV-images at constant velocity. Secondly, accurate LDA measurements and data from direct numerical simulation (DNS) are used as a basis for the analysis in a low Reynolds number open water channel flow. The results show that excellent agreement between LDA and DIC–PIV measurements for mean and RMS velocities can be achieved using a compression factor up to 12. Received: 27 August 1996 / Accepted: 15 December 1998     

PIV measurement of flow around an arbitrarily moving body

This paper presents a PIV (particle image velocimetry) image processing method for measuring flow velocities around an arbitrarily moving body. This image processing technique uses a contour-texture analysis based on user-defined textons to determine the arbitrarily moving interface in the particle images. After the interface tracking procedure is performed, the particle images near the interface are transformed into Cartesian coordinates that are related to the distance from the interface. This transformed image always has a straight interface, so the interrogation windows can easily be arranged at certain distances from the interface. Accurate measurements near the interface can then be achieved by applying the window deformation algorithm in concert with PIV/IG (interface gradiometry). The displacement of each window is evaluated by using the window deformation algorithm and was found to result in acceptable errors except for the border windows. Quantitative evaluations of this method were performed by applying it to computer-generated images and actual PIV measurements.     

Spatial resolution of PIV for the measurement of turbulence

Recent technological advancements have made the use of particle image velocimetry (PIV) more widespread for studying turbulent flows over a wide range of scales. Although PIV does not threaten to make obsolete more mature techniques, such as hot-wire anemometry (HWA), it is justifiably becoming an increasingly important tool for turbulence research. This paper assesses the ability of PIV to resolve all relevant scales in a classical turbulent flow, namely grid turbulence, via a comparison with theoretical predictions as well as HWA measurements. Particular attention is given to the statistical convergence of mean turbulent quantities and the spatial resolution of PIV. An analytical method is developed to quantify and correct for the effect of the finite spatial resolution of PIV measurements. While the present uncorrected PIV results largely underestimate the mean turbulent kinetic energy and energy dissipation rate, the corrected measurements agree to a close approximation with the HWA data. The transport equation for the second-order structure function in grid turbulence is used to establish the range of scales affected by the limited resolution. The results show that PIV, due to the geometry of its sensing domain, must meet slightly more stringent requirements in terms of resolution, compared with HWA, in order to provide reliable measurements in turbulence.

---

Background extraction from double-frame PIV images

This study presents a simple image pre-processing scheme to extract background information from double-frame particle image velocimetry (PIV) images. Everything that stays stagnant in the image (e.g., image background and light reflections from stationary objects) is assumed to be a source of disturbance and is removed by subtracting the second frame of the image pair from the first frame. This yields a single frame difference image, which is transferred back to a background extracted double-frame image. After the procedure the background in the image does not correlate with itself anymore and therefore a bias error in PIV analysis towards zero displacement is avoided. The simulations show that the procedure conserves the profile of tracer particle images when a displacement between the frames is larger than a particle image size. The performance of this procedure is emphasized with various examples, and extensions of the procedure are introduced. The extended procedure extracts background objects that move between the image frames, e.g. dispersed phase particles in a two-phase flow or laser light sheet reflections from moving objects.