Table of contents

The spectral density of the current fluctuations (noise) of the channel electron multiplier (channeltron) working in the analogue mode has been measured. The results for the channeltron operating in the pulse counting mode have been analysed and compared with those for the analogue mode. In both cases the existence of `flicker noise' (i.e. a spectral density function of the 1/f^α (f is the frequency) type) has been observed unless the current has reached high enough values.

Very few investigations of bubbly two-phase flows have used X-hot-film probes for the measurement of the turbulence field in the continuous phase. Therefore the interactions between the sensors of the X probe and the bubbles are not well known and thus can lead to errors in the determination of turbulence parameters. To gain understanding of the interaction, investigations of static and dynamic bubble-sensor interactions were carried out. In the static investigations a probe that is moved through stagnant water interacts with a bubble that is fixed to a small nozzle. These experiments help us to understand signals for situations in which the liquid and the bubble move with the same speed. In this case, the so-called post-signal occurring at the exit of the sensor penetration through the bubble is caused by the deformation of the bubble surface by the sensor and its prongs rather than by the flow field. In the dynamic investigations the probe is fixed in a vertical flow and it interacts with single bubbles of a bubble stream, as is the case in any two-phase flow situation. A bubble-probe interaction map is proposed that classifies the behaviour into four different modes of interaction. In both series of measurements, various parameters characterizing bubble signals were determined. A synchronized video system is used for visual investigations of deformations of the bubble surface. The results will be used for the future development of a so-called bubble elimination algorithm that is used for filtering the bubble signals out of the two-phase signal.

A photon event counting imaging detector, originally developed for astronomical applications, has been adapted for use on a low light level fluorescence microscope. It is based on a 40 mm diameter, three-microchannel plate image intensifier with a photon gain of 10⁷ when operating in the photon event counting mode. The intensifier's output screen is lens coupled to a full frame progressive scan CCD camera which transfers the photon event data into a PC via a framegrabber. The image is built up from the photon events in the frames using software-based event processing. The high sensitivity of the photon counting approach means that low excitation light levels and/or a low probe concentration can be used. Moreover, the inherent linearity between incident intensity and the number of detected events provides intrinsic photometric accuracy independent of camera gain and offset.

To demonstrate the viability of photon event counting imaging applied to fluorescence studies, we have imaged 4-dicyanomethylene-2-methyl-6-(p-dimethylaminostyryl)-4H-pyran- (DCM-) and rhodamine 6G-doped thin films produced by the sol-gel technique using tetraethylorthosilicate (TEOS) as a precursor. We show that fluorescence imaging is a feasible tool for finding cracks, streaks or other defects and for assessing the uniform thickness of the finished films. In addition, we image the fluorescence in response to an excitation beam illuminating rhodamine 6G-doped bulk monolith sol-gel samples. The results are analysed using the Lambert-Beer law.

Lock-in thermography is a technique which is increasingly being used for the evaluation of subsurface defects in composite materials such as carbon-fibre-reinforced polymers (CFRPs) in aircraft structures. Most CFRP structures have a finite thickness and non-destructive inspection is performed in a natural ambient environment. In this paper, a photothermal model is developed in order to investigate the behaviour of thermal waves in homogeneous plates and layered plates with finite thicknesses under convective conditions. The model is then utilized to predict the phase differences produced by multi-layer subsurface defects and optimum inspection parameters. The theoretical results are compared with the experimental results. The detectivity of lock-in thermography for CFRP is also presented in this paper.

The integration of an acoustography ultrasonic imaging system with a mechanical testing machine is described. The apparatus is designed for real-time ultrasonic imaging of impact damage growth in composite material samples subjected to either static or fatigue loading. Experimental results are presented of the damage growth to failure of an aerospace standard carbon-fibre composite sample subjected to compressive cyclic loading. Experimental results for static loading show a reversible increase and decrease in damage area, indicating an opening and closing of delamination defects, which points to the possibility of an under-estimation of defect size by the conventional ultrasonic testing of unloaded components.

We present a high-resolution heterodyne interferometric system using a high-output-power, three-longitudinal-mode He-Ne laser. In this system, an interferometric wavelength of λ/2 is synthesized with the super-heterodyne detection of the three modes, resulting in twice the sensitivity for the target displacement in comparison with the conventional two-frequency interferometer. In the experiment, the achievement of subnanometre resolution has been confirmed.

Measurements on re-emitted positrons from electro-deposited NiW show the alloy to be an excellent candidate for the positron moderator in a new generation of positron beam systems in which hitherto unattainable geometries can be used either to accommodate novel beam designs and/or to generate more intense beams of positrons. After annealing at 1350 °C the moderation efficiency of the NiW is found to be ~8×10^-4, comparable to that commonly reported for polycrystalline W. It emits positrons of energies up to ~3 eV, a value characteristic of the work-function emission of positrons from W. The re-emitted positron energy spectrum contains features characteristic of both Ni and W, suggesting that a degree of surface segregation takes place on annealing. The remoderation efficiency of the alloy is stable after a decline of approximately 10% in the first two days after annealing in situ.

A cost-effective methodology for operating the 349-pixel Cerenkov light imaging camera of the imaging element of the TACTIC γ-ray telescope array at stable single-counts rate and safe anode current values has successfully been tried out despite variations in the light of the night sky experienced by these pixels from time to time. Important details of this unconventional scheme are discussed and reference is made to the methods evolved for in situ absolute and relative gain calibrations of the camera pixels, which are particularly required for this mode of operation.

Local velocity measurements of interstitial flow through porous media and observation of the local pore geometry, which strongly influences the flow field, are necessary in order to understand the transport mechanism in a porous system. In order to investigate the effect of pore geometry on the flow field in porous media, three-dimensional velocity measurements of interstitial flow through two porous samples with different pore geometries have been experimentally performed using magnetic resonance imaging (MRI). The porous samples were made from cylindrical tubes in one case packed with crushed glass particles and in the other spherical beads. Based on the observed images of the porous samples, the porosity and pore size distributions as pore structure parameters were determined by image analysis. The two- and three-dimensional velocity vectors of the steady-state flow in the void spaces of the crushed glass sample have been measured using a spin-echo sequence with tagging pulses. The obtained velocity maps show a strong non-uniform flow. In places, reversed flow is induced by the pore geometry. The effect of the mean flow velocity on the change in the flow pattern has been investigated in the mean flow velocity range from 3.84 to 13.2 mm s^-1. The frequency distribution of the axial velocity component of the interstitial flows through the packed beads was broader than that of the crushed glass pack.

An intensity-based technique for the interrogation of fibre Bragg grating sensors is described. The technique utilizes two longitudinal modes of a single, pulsed laser diode as a dual-wavelength source for grating illumination. Time-division-multiplexed addressing of sensor arrays in serial and tree topologies is also demonstrated.

This paper discusses the experimental setup of a 3D LIF imaging facility. The advected scalar concentration field of a LIF dye is digitized quasi-instantaneously. The x- and y-resolution are determined by the camera optic, the z-resolution by the scan process. In the absence of scan noise, spatial resolution can be designed to be equal in all three directions. The thus obtained 3D field can be used as input signal for an optical flow measurement technique. The scope of the imaging is mainly constrained by the local Reynolds number of the flow and the state of current high-speed digital cameras and laser scanners. The reciprocal relationship of the important parameters is discussed along with some implementational examples for the turbulent free jet.

We describe a compensated common path lateral shearing interferometer that allows fast two-dimensional measurements of the optical thickness of transparent plates. The bench is able to provide accurate results even when heterogeneous residual birefringence is present in the sample. This bench brings several improvements to an optical system originally built by Mr G Nomarski.

An effective novel method for improving the detecting of interturn short circuits on rotor windings is described. The approach consists of measuring the magnetic flux density when the rotor is rotating at a nominal rotation speed and a current excitation is applied to the field windings. The measurements are obtained from a stationary search coil tangential to the rotor and near to its surface. A mathematical model that fits the tangential magnetic flux density is used to simulate failure conditions. Finally, a statistical analysis is employed to estimate the probability of making wrong decisions about the occurrence of failures. Finally, a practical application of this methodology is described.

A unique air flow rate measurement device has been developed and tested for use in the Axial Fan Research and Development facility at Michigan State University. The subject mass air flow meter is located downstream of a wide variety of test fan and component geometries. These can significantly change the velocity profile entering the metering nozzles. For this reason a point measurement of a characteristic velocity in the nozzle outlet would not provide a viable experimental technique for the flow rate evaluation. This also negates the option of using a simple nozzle discharge coefficient.

An integral measurement technique was therefore implemented. A 90° turning passage supported on a knife edge pivot was used to create a measurable net moment-of-momentum flux. The resistive moment provided by a strain gauge force transducer was used to infer this net flux. The acceleration of the flow from inlet to outlet of the 90° turn and the larger contribution of the outlet moment-of-momentum flux provide the desired insensitivity to the non-uniform inlet flow conditions. A calibrated nozzle was used to determine the `force-mass flow rate' transfer function.

This note presents an analytical model that relates the displacement found by peak finding in cross-correlation particle image velocimetry using light-sheet illumination to the weighted volume average of the fluid velocity field. General equations are derived describing the cross-correlation function in light-sheet PIV. An approximate model is derived for light-sheet illumination by approximating the particle images as Gaussian. This results in a weighting function that can be used to define the measurement volume and to find biasing of the velocity measured for this volume.

To use Nature as a model for a new sensor technique - that is a concept that has become very attractive in recent years. This biomimetic approach deals with sensors and sensor systems based on mechanisms found in biological systems. One branch consists of the concept of biosensors, a biosensor being a combination of a biologically active material with a transducer element. Another branch consists of the concept of artificial senses, such as so-called electronic noses and electronic tongues. The principles of these are analogous to the sense of smell: a signal pattern from a sensor array with different selectivities is processed with multivariate data analysis for recognition and learning. Electronic noses have already become an established technique, and the establishment of electronic tongues is rapidly growing.

Professor K Toko at Kyushu University, who is a pioneer in these fields, has now written a book to introduce the biomimetic approach to sensor technology. He has also developed an electronic tongue based on lipid/polymer membranes. The aim of this system is to mimic the human taste sensation; therefore it is called a `taste sensor'. This taste sensing system has now also been commercialized. The book also covers this taste sensing system in detail.

In the introductory part, the principles of measurement systems and different techniques of multivariate data analysis are described and discussed. This also includes some discussions concerning the five human senses and their integration to form an overall decision of quality. In the next part, reception mechanisms of biological systems are briefly described with emphasis on the taste system of the tongue. A large section on properties of some lipid membranes then follows, including their potential as membranes for the taste sensing system. The next section covers principles and properties of biosensors, followed by a chapter on odour sensors - so-called electronic noses. The principles of olfaction and its basis for the electronic nose are discussed. The taste sensing system is thoroughly described in a large section. The abilities of different membranes to detect the five tastes of the tongue (sweet, sour, bitter, salt and umami) are described. Details of many applications, mainly in the food industry, are given. In the last part, other methods to measure taste based on impedance, surface plasmon resonance and surface photovoltage are described. Finally, some future aspects towards a common sensing system based on artificial analogues to all human senses are touched upon.

To summarize the book, it covers all essential areas in biomimetic sensor technologies, biosensors, electronic noses and electronic tongues. It also covers multivariate data processing techniques, which are a very important aspect of these systems. It examines the term `deliciousness', which actually is the combination of the information from all five human senses to give an overall impression of the experience of the food. The initial part on sensors and measurement methods is very valuable in putting the subsequent chapters in perspective. The section dealing with artificial membrane properties, however, is rather large, and there is little description of how biological membranes work. Thus, a section dealing more deeply with olfaction and biological receptors would have made the book more complete. The hypothesis of how membrane binding can start excitation in a receptor cell and how this can be a used as a prototype for a taste sensor, however, is interesting.

Although the term `electronic tongue' suggests measurements on taste, most of the research groups working with this concept are trying to find not the taste but the quality. The taste sensing system describes the result in terms of the five basic senses of the tongue. It is a little confusing, then, when in some cases this system also refers to more general properties of quality. It is also surprising that no other concepts of electronic tongues are mentioned in the book.

This book is recommended to researchers in the field of electronic noses and tongues. Part of it could also be useful for introducing postgraduate students to the biomimetic approach in sensor technology.

Fredrik Winquist

Within metrology the evaluation of measurement uncertainty plays a special and very important role. In the past the generally accepted procedure for establishing the uncertainty was the so-called error analysis. Within this procedure probability is interpreted strictly in the statistical (frequentist) sense. Since the available information needed to infer the measurand almost never consists of only observed statistical data, the inference on the basis of error analysis showed difficulties that in 1993 were overcome by an international recommendation to apply, at least within the various branches of metrology, Bayesian Statistics to measurement data. But the price to pay for the resulting consistent and satisfying evaluation procedure is to change one's view of probability and to accept the consequences. This needs guidance, particularly for those educated in frequency-based statistics.

The book Bayesian Theory, first published in hardback in 1994, provides in about 600 pages a very clear, careful and well structured description of the foundations and key theoretical concepts of Bayesian Statistics. Although not written especially for metrologists and their needs to evaluate measurement data, the text to a large extent can be recommended without restriction to this community as an extensive and excellent introduction and guide to the world of Bayesian coherent reasoning. The detailed comparison with non-Bayesian theories, such as the frequentist procedures, is particularly enlightening in understanding the basic Bayesian concepts.

In the volume the authors concentrate on the answer to the question of why Bayesian Statistics should be used at all. Only in future volumes will they deal with analytical and numerical techniques to implement Bayesian procedures and the study of methods of analysis for various types of models and problems. The contents of the book go well beyond the problem of statistical inference, which is viewed as a special case of decision theory. The interpretation of probability in Bayesian Statistics as well as the foundations of probability theory and decision theory are presented. Apart from modelling and model comparison the central Bayesian problem of the prior distribution, particularly in case of ignorance, is addressed at length. It is discussed in conjunction with the introduction of information-theoretic concepts. Critical issues are explicitly mentioned and discussed. Numerous examples illustrate the clearly expounded theoretical considerations. There are about 1500 references (only a few later than 1994) including a list of other Bayesian textbooks. A list of abbreviations and symbols used is unfortunately missing. Whereas the authors throughout focus on statistical concepts rather than rigorous mathematics the reader should be prepared for a mathematical presentation on the level of advanced calculus.

The book will not be the primary source for an actual evaluation of the uncertainty of measurement given the model of evaluation and incomplete information about random and systematic effects occurring in measurement. But it certainly is an excellent primary source for those who wish to learn about the learning and decision process in a situation of uncertainty. It is this situation the metrologist faces after measurement when having to state what he has learned about the measurand.

Wolfgang Wöger

This book claims to fill a niche in the provision of textbooks devoted to image processing by being devoted to colour aspects. It is aimed at researchers and practitioners working in the area of colour image processing, particularly graduates in Computer Science and Electrical and Computer Engineering. The book is structured in such a way that, after reading an introductory chapter on colour, readers can work their way through the following four chapters on image processing, or omit them and move directly to the final three chapters that deal with colour image analysis and coding. It is not immediately apparent from reading the preface that companion image processing software is available online from the publisher's website, and this is not made use of as an integral part of the book.

Regrettably, the book shows much evidence of a lack of rigorous proof reading and editing. A number of errors can be found, particularly in the first chapter; this provides the fundamentals in colour science on which the book is based. For example, on the first page the visible spectrum is incorrectly reproduced with the wrong wavelength scale and photoreceptors are referred to as `roads'. A few pages on there is a complete mismatch between the explanatory text and diagrams concerning the CIE, XYZ and RGB colour matching functions. One diagram appears to have been reproduced twice, is incorrectly titled and not annotated. Also, the CIE chromaticity diagram lacks a wavelength scale and the colours of the diagram are incorrectly reproduced. Unfortunately, these fundamental errors appear in the first ten pages and have the unfortunate effect of detracting from the authoritative nature of the book as a whole. A further example of poor proof reading/editing can be found towards the end of the chapter in which photographic film is defined as follows: `The film which is used by conventional cameras contains three emulsion layers which are sensitive to red and blue light, which enters through the camera lens.' The chapters do improve, however, as one goes through the book, although in chapter 2 the description of the origins of photographic noise is incorrect and incomplete (`the noise is mainly due to the silver grains that precipitate during film exposure'). Also, the origins of noise in photoelectronic sensors are incompletely described.

Each chapter is accompanied by a substantial number of references to the primary sources of information, many of which are to recent literature in the field, in very useful summary or conclusion sections. It is puzzling that in view of the significance of the Fourier transform in image processing, it is not included, other than a brief mention of the Discrete Fourier Transform in the chapter on image compression. Adaptive image filters are described at length in chapter 3, which is followed by chapters dedicated to colour edge detection, enhancement and restoration, image segmentation, image compression and emerging applications. The latter makes interesting reading but is based almost exclusively on the detection and automatic location of the human face.

The index is not very exhaustive and as a consequence it is difficult to find many items that are discussed in the text but are not indexed. A few examples include: Wiener filter, Sobel, Prewitt and Robert's edge detection, all of which appear in the text and in the indices of most books on image processing but not in the index to this book. Also, most of the existing texts on image processing include many aspects of colour, which detracts somewhat from this book's claimed uniqueness, although it does contain more state-of-the-art aspects.

Ralph Jacobson