Evaluation of Spherical Fitting as a technique for sizing iron ore pellets is performed. Size measurement of pellet in industry is usually performed by manual sampling and sieving techniques. Automatic on-line analysis of pellet size would allow non-invasive, frequent and consistent measurement. Previous work has used an assumption that pellets are spherical to estimate pellet sizes. In this research we use a 3D laser camera system in a laboratory environment to capture 3D surface data of pellets and steel balls. Validation of the 3D data against a spherical model has been performed and demonstrates that pellets are not spherical and have physical structures that a spherical model cannot capture.

Size measurement of pellets in industry is usually performed by manual sampling and sieving techniques. Automatic on-line analysis of pellet size based on image analysis techniques would allow non-invasive, frequent and consistent measurement. We make a distinction between entirely visible and partially visible pellets. This is a significant distinction as the size of partially visible pellets cannot be correctly estimated with existing size measures and would bias any size estimate. Literature review indicates that other image analysis techniques fail to make this distinction. Statistical classification methods are used to discriminate pellets on the surface of a pile between entirely visible and partially visible pellets. Size estimates of the surface of a pellet pile show that the overlapped particle error is overcome by only estimating the surface size distribution with entirely visible pellets.

A novel approach for monitoring and control of the coal powder injection in a blast furnace is presented and discussed. Image analysis of video recordings is used as a means to estimate the instantaneous coal flow. Initial experiments at the blast furnace no 3 of SSAB Tunnplat AB Lulea, Sweden, have been performed and first hand results on modelling and control of a single injection line are given.

A novel approach to monitoring and control of the coal powder injection in a blast furnace is presented and discussed. Image analysis of video recordings is used as a means to estimate the instantaneous coal flow. Initial experiments at blast furnace number 3 of SSAB Tunnplat AB, Lulea, Sweden, are reported and firsthand results on modeling and control of a single injection line are given.

Determining lubricant film thickness between contacting bodies under elastohydrodynamic (EHD) conditions is often simulated by using a ball/cylinder and transparent disc apparatus together with an interferometry technique. The simulated contact will have a point or elliptic shape and the light used can be white or monochromatic. The interference pattern is normally photographed with a regular camera or a video camera and the pictures are then evaluated by the naked eye of the observer. In most cases, only central or minimum thicknesses are evaluated.

In this paper an image processing method for the analysis of film thickness is presented. This method makes it possible to extract considerably more information about film thickness fluctuations than is achievable by the naked eye. The method primarily matches hue (but also saturation and intensity values) from digitized colour interferometric images of the unknown film shapes with calibration values obtained with known geometric shapes.

The method is shown to work well in the range from 95 up to 700 nm with white light and makes the results unbiased by the observer. Furthermore, absolute film thickness can be evaluated without prior knowledge about the fringe order in the interferogram.

Image analysis results of video-recorded slow fine-grain granular flows, using the accumulating particle trajectories method, are presented. Analytic and simulation results, stating conditions on numbers of particle passages and on particle velocities related to particle and pixel size which are necessary for accurate measurements, are presented

Soap lumps entering a grease-lubricated elastohydrodynamic lubrication (EHL) contact were traced as they passed through the contact by using optical interferometry in a standard Ball & Disc apparatus and a high-speed video camera with light enhancer. One specific soap lump could be traced when it was passing through the contact and sharp images of the passing lump could be captured. This paper presents a combined experimental and numerical approach to determine what is happening when such lumps pass through the contact. From the fringe pattern, obtained by optical interferometry, a film thickness map is created by using image processing. This is done for every time step (1 ms) as the lump is passing through the contact. These maps serve as input to a numerical computation of the pressure by assuming elastic deformation theory. Consequently, no assumptions about the rheological behaviour of the grease have to be made. Two greases were studied, based on the same synthetic poly-α-olefin but thickened with Li-12-OH and Li-complex respectively. It is seen that the soap thickened lumps can cause deep elastic indentations accompanied by large pressure fluctuations. The pressure level will in some cases be more than doubled due to lumps entering the contact region. The effects on noise level and operational life are also discussed.

A hybrid evaluation scheme for EHL film thickness determination is proposed and discussed. The film thickness profile in the contact region is measured using interferograms produced with a novel multi channel interferometry method. Since the refractive index distribution in the contact is pressure-dependent, and the initial film thickness profile will be evaluated assuming atmospheric pressure, a refractive index correction scheme is employed. The correction scheme is based on the Lorenz-Lorentz equation and a pressure-density relation together with a numerical pressure solver taking the initial film thickness measurement as input. The film thickness determination scheme is applied to an interesting phenomenon that can be observed at sliding conditions when the discrepancy occurred in the form of a deep and large dimple in the conjunction. Such a dimple appeared instead of the conventional plateau. The phenomenon was studied under different degrees of sliding. The detailed film thickness maps and pressure distributions for highly loaded EHL conjunctions at high degrees of sliding are produced using a hybrid evaluation scheme. The results are analyzed and discussed.

The ability to measure lubricant film thickness in elastohydrodynamic point contacts was greatly improved when the method of interferometry was introduced in the late 1960s. Constant refinements of the technique have made it possible to measure the thickness of thin films with an accuracy of a few nanometers. In order to further develop the technique and provide a tool for advanced lubricant experiments, a Ball and Disc Apparatus was developed. The aim was to accomplish an apparatus with an open architecture for easy expansion and the ability to use several sub-methods for film thickness determination.

The possibilities for studying thin film EHL using optical interferometry has improved considerably with the introduction of spacer layer techniques. Film thicknesses down to zero can now be measured using different interferometric evaluation techniques. One technique is based on image analysis and a two-dimensional film thickness map can be obtained for the whole contact between a ball and a transparent disc. The image analysis techniques that have been employed normally requires a calibration table which couples separation with the measured colour parameters. There are a number of disadvantages with such a calibration procedure and the proposed Multi channel method makes it possible to avoid calibration in the same time as the evaluation time is much reduced. This paper presents a first attempt to apply the Multi channel method on the measurement of very thin film and a comparison with a calibration method (the HSI colour calibration method). A spacer layer is used in order to obtain high resolution even for very thin films. It is seen from the results that both methods correspond well at thicker films (>30 nm) while there are large differences for very thin films. The Multi channel results are promising and the method has a potential to be even further improved.

In this paper, a new method to reconstruct piecewise continuous phase estimates using inphase and quadrature components acquired from interferometry measurements is derived and discussed. The method, based on the concept of anisotropic evolution formulations, is shown to be far less noise sensitive than similar methods operating on modulo-mapped data (i.e., traditional phase unwrapping methods). The method is able to produce reliable phase estimates from data containing complex sheared structures in combination with high noise content without relying on user-defined weights.

Section A in this thesis is dedicated to measurements of lubricant films separating rotating surfaces in elastohydrodynamic experiments. Two specially designed interferometry-based methods are presented. A method to compensate for measurement errors due to the pressure dependence of the refractive index of the lubricant is proposed. Automatic fringe analysis is treated in section B. A noise-insensitive two-dimensional phase unwrapping method is presented, together with a pre-processing method that estimates local fringe density and direction. A method to determine a continuous phase from interferometric measurements, where no intermediate modulo-mapped phase data is produced and thus no phase unwrapping is necessary, is proposed.

A method for interferometric measurement comprises emitting waves onto a reference surface (1) and onto a measured object (2), which each reflects a part of the emitted waves, and receiving both the reflected parts of the waves by the same receiver (4). In the receiver the parts of the waves generate a representation of the measured object (2) in the form of an interferogram, from which the form of the measured object (2) is determined. The emitted waves comprise waves of three well-defined wavelengths nu1, nu2 and nu3, where nu1<> nu2<> nu3, which wavelengths are chosen to substantially satisfy the mutual relation nu1=(nu2.nu3)/(2.nu3-nu2) and generate an interferogram each in the receiver (4). An apparatus suited to implement the method comprises a light source (3) for emitting light, a reference surface for reflecting a first part of said light, and a receiver (4) which is arranged to receive the first part of the light and a second part of the light, reflected by a measured object (2). The light source is arranged to simultaneously emit light of the three above-mentioned well-defined wavelengths.

A new two-dimensional phase unwrapping method, based on an iterative computational procedure, is proposed. The method, which is derived with the use of a global cost function to minimize the phase discontinuities in the unwrapped phase map, has shown to produce robust and reliable results on very noisy phase data. Preprocessing operations, such as noise cleaning or segmentation, will in many cases be superfluous but may be included

A method for automated fringe analysis is presented. It robustly estimates local fringe density and direction in noisy wrapped phase maps. Such information can be used to improve the performance of two-dimensional phase unwrapping methods, to construct phase-jump-preserving filtering strategies, and also to perform robust segmentation of phase data. The method, which is highly insensitive to noise, is model based and performs the estimation in the Fourier domain.

The lubricant film thickness between contacting bodies may be determined from interferograms by image analysis. Look-up tables employed in the analysis are obtained from calibration interferograms of contacts of bodies of known shape. Calibration and measurement interferograms are, however, obtained under different pressure conditions and, since the refractive index of lubricants is pressure dependent, the use of look-up tables can introduce significant errors (in the range of 10 per cent). In this paper it is shown that a straightforward correction for this error is available. In an iterative procedure the uncorrected film thickness is first determined through looking up tables. This allows the pressure distribution over the contact to be calculated and a corrected refractive index may then be obtained. The corrected refractive index in turn allows a corrected film thickness to be calculated. This procedure can be followed iteratively. After two iterations the corrections have been found to be of the same order as the resolution yielded by the method, as has been implemented

White light interferometry yields good resolution in thin film thickness measurements. Hue value variations vs. film thickness variations are complex and furthermore dependent on the power spectrum of the light source. Absolute film thickness measurements is possible using white light interferometry

Measurement of the thickness of thin lubricant films separating rotating surfaces in elastohydrodynamic experiments presents some challenging problems. The nature of the experimental apparatus inhibits the use of most commonly applied interferometric phase measurement methods. Also the absolute thickness of the separating film must be determined, as opposed to relative distances that would be sufficient in most other measurement scenarios where interferometry methods are used. In this paper, computer-based analysis of interferograms recorded using an elastohydrodynamic lubrication Fitzeu interferometer (a so-called ball-and-disc apparatus) is discussed, the main objective being to extract the absolute oil-film thickness. Intensity based methods (most importantly, calibration look-up procedures where colour parameters from recorded dynamic interferograms are compared with table values corresponding to known film thicknesses, but also a phase measurement approach based on multi-channel interferometry using trichromatic light) are described. A discussion regarding compensation for measurement errors due to the pressure dependence of the refractive index of the lubricant is also included

The extension of path-independent 2-D phase unwrapping algorithms, based on placement of branch cut lines between phase singularities of opposite sign, was recently proposed for phase volumes in a recent paper by Huntley. In 3-D the singularities were shown to form closed loops and path-independence was achieved by placing branch cut surfaces across the loops. In the current paper we describe in detail an optimised and extended version of Huntley's algorithm. This deals in particular with two aspects which are essential for practical phase volumes: (1) how to close partial loops which pass through arbitrary boundaries separating valid and invalid phase data, and (2) how to select the set of loops having shortest length. The second algorithm is necessary to deal with the ambiguous cases that can arise when the singularities form knots, i.e. two loops pass through a single phase volume element. The perfomance of the algorithm is demonstrated on 3-D phase maps from two types of medical imaging data: MRI and x-ray interferometry.

The extension of path-independent 2-D phase unwrapping algorithms, based on placement of branch cut lines between phase singularities of opposite sign, was recently proposed for phase volumes in a paper by Huntley. In 3-D, the singularities were shown to form closed loops, and path independence was achieved by placing branch cut surfaces across the loops. In the current work, we describe in detail an optimized and extended version of Huntley's algorithm. It deals in particular with two aspects that are essential for practical phase volumes: 1. how to close partial loops that pass through arbitrary boundaries separating valid and invalid phase data, and 2. how to select the set of loops having the shortest length. The second algorithm is necessary to deal with ambiguous cases that can arise when the singularities form knots, i. e., two loops pass through a single phase volume element. The performance of the algorithm is demonstrated on 3-D phase maps from two types of medical imaging data: medical resonance imaging (MRI) and x-ray interferometry.

Phase contrast magnetic resonance velocity imaging is a powerful technique for quantitative in vivo blood flow measurement. Current practice normally involves restricting the sensitivity of the technique so as to avoid the problem of the measured phase being 'wrapped' onto the range - π to + π. However, as a result, dynamic range and signal-to-noise ratio are sacrificed. Alternatively, the true phase values can be estimated by a phase unwrapping process which consists of adding integral multiples of 2π to the measured wrapped phase values. In the presence of noise and data undersampling, the phase unwrapping problem becomes non-trivial. In this paper, we investigate the performance of three different phase unwrapping algorithms when applied to three-dimensional (two spatial axes and one time axis) phase contrast datasets. A simple one-dimensional temporal unwrapping algorithm, a more complex and robust three-dimensional unwrapping algorithm and a novel velocity encoding unwrapping algorithm which involves unwrapping along a fourth dimension (the 'velocity encoding' direction) are discussed, and results from the three are presented and compared. It is shown that compared to the traditional approach, both dynamic range and signal-to-noise ratio can be increased by a factor of up to five times, which demonstrates considerable promise for a possible eventual clinical implementation. The results are also of direct relevance to users of any other technique delivering time-varying two-dimensional phase images, such as dynamic speckle interferometry and synthetic aperture radar

This project developed an industrial prototype 3D imaging and analysis system for measuring the size distribution of iron ore green pellets. The system is installed and operational at a pellet production plant capturing and analysising 3D surface data of piled pellets on the conveyor belt. It provides fast, frequent, non-contact, consistent measurement of the pellet sieve size distribution and opens the door to autonomous closed loop control of the pellet balling disk or drum in the future. Segmentation methods based on morphological image processing are applied to the 3D surface data to identify individual pellets. Determination of entirely visible pellets (non-overlapped) is made using a two feature classification strategy, the advantage being that this system eliminates the bias the results from sizing overlapped particles based on their limited visible profile. The system achieves what a number of commercial 2D fragmentation measurement systems could not satisfactorily achieve for the pellet producer, that is, accurate sizing of the green pellets.