In this paper we present a novel system concept for large scale flatness and dimensional measurement applications. This system has several advantages over traditional projection moire systems. The system has a robust optomechanical structure which avoids the tight mounting tolerances seen in current projection moire implementations. This means that the system concept will be more applicable to installations in a hostile production environment, at lower cost. The system can be installed close to the object under measurement and produces a good quality moire signal on variable surfaces including impurities, texture, markings, etc. The competitiveness of the flatness measurement system is increased by building a multipurpose system and adding such measurements as edges, width or surface quality to the same unit.

Most optical topology systems use a single wavelength laser for projection, using a swept spot, a moving line, or a projected grating. In a typical projected grating system, the gratings are shifted and a series of images are used to recover the 3D shape of the target. When the series of images is analyzed in the normal phase shift manner, the resulting 2D phase map typically has phase unwrapping problems due to noise and Nyquist limits. Surfaces with large vertical discontinuities present the biggest problem in 3D shape recovery. This paper looks at simultaneously projecting multiple wavelengths onto a surface to help avoid problems in unwrapping the 2D phase map. Using multiple wavelengths, the interferometer mirror does not have to be rotated to change the grating pitch and some operations can be done in parallel which reduces scanning time. Limitations and improvements in the current system will be discussed.

A Mach-Zehnder based variable resolution fringe projection system has been built for 3D video moire machine vision. This system uses the three main advantages of the Mach- Zehnder - 1) There is no optical feedback to the laser source; 2) The interferometer can accept two different laser wavelengths simultaneously; and 3) The interferometer produces two orthogonal output beams. The lack of optical feedback makes the Mach-Zehnder especially attractive for use with high power laser diode sources which are sensitive to optical feedback. When the two input ports are used with two different wavelength laser, the target can be illuminated by simultaneous projection of two different sets of colored fringes with two different spatial frequencies. This can allow more reliable reconstruction of the 3D surface over discontinuous jumps. Finally, the lack of feedback to the source coupled with the dual outputs means that he Mach-Zehnder fringe projector is very efficient in that 100 percent of the laser light is projected onto the prime and reference targets. Setup and alignment of this interferometer will be discussed for both parallel and diverging light. Plots of fringe visibility will be given for both outputs and both inputs.Application to a video moire based real time 3D error map machine vision system will be discussed.

Structured light methods, including moire contouring have been demonstrated both technically and commercially as a useful tool for a class of 3D contouring applications. Popular analysis methods such as phase shifting, Fourier analysis, and simple fringe counting all have their strengths and weaknesses. Phase shifting has proven to be a powerful tool for complete surface mapping with good performance, but requires multiple images for the analysis. Fringe counting and Fourier methods offer good speed, but are more sensitive to noise factors. The trends in manufacturing toward tools for in-process measurements has lead to a variety of methods for high sped methods including color moire, multi-camera/sensor systems, and combinations of methods such as stereo with moire methods. This paper provides a comparison of some of these methods with particular emphasis on sources of noise and system error functions. Specific systems discussed include single image fringe counting using gradient methods, multiple camera systems using phase analysis, color encoding methods, motion based systems and image multiplexing methods. The analysis will attempt to separate accountable systematic error sources from random noise and the degree of participation of each in the data analysis methods.

Adopting phase-shifting technique in moire topography provides many advantages in measuring complex surface profiles with varying reflectance. However, still the so- called 2(pi) -ambiguity problem remains, which limits the maximum measurable step height difference between two neighboring sample points to be less than half the equivalent wavelength of moire fringes. To cope with the problem in this investigation, a two-wavelength scheme of projection moire topography is proposed along with necessary hardware design considerations. Test results prove that the proposed scheme is capable of finding absolute fringe orders automatically, so that the 2(pi) -ambiguity problem can be effectively overcome so as to treat large step discontinuities in measured surfaces.

A 3D vision system named SVIS is developed for 3D shape measurement that integrates three methods: (i) multiple- baseline, multiple-resolution Stereo Image Analysis (SIA) that uses colore image data, (ii) Image Defocus Analysis (IDA), and (iii) Image Focus Analysis (IFA). IDA and IFA are less accurate than stereo but they do not suffer from the correspondence problem associated with stereo. A rough 3D shape is first obtained using IDA and then IFA is used to obtain an improved estimate. The result is then used in SIA to solve the correspondence problem and obtain an accurate measurement of 3D shape. SIA is implemented using color images recorded at multiple-baselines. Color images provide more information than monochrome images for stereo matching. Therefore matching errors are reduced and accuracy of 3D shape is improved. Further improvements are obtained through multiple-baseline stereo analysis. First short baseline images are analyzed to obtain an initial estimate of 3D shape. In this step, stereo matching errors are low and computation is fast since a shorter baseline result in lower disparities. The initial estimate of 3D shape is used to match longer baseline stereo images. This yields more accurate estimation of 3D shape. The stereo matching step is implemented using a multiple-resolution matching approach to reduce computation. First lower resolution images are matched and the result are used in matching higher resolution images. This paper presented the algorithms and the experimental result of 3D shape measurements on SVIS for several objects. These results suggest a practical vision system for 3D shape measurement.

Traditionally, gear tooth profile is measured by coordinate measurement machines using mechanical probe scanning, which is very time consuming. Therefore, normally only a few lines are scanned across the gear tooth surface, which often do not faithfully represent the whole tooth surface profile. In this paper, a phase-shift optical triangulation technique is applied to the gear profile measurement, which has the advantages of redundant information, speed and non-contact nature. When examining a 2 cm² area, comparative measurement accuracy better than 1 micrometers and resolution about 0.1 micrometers has been successfully demonstrated with the system; this corresponds to a 5 by 10^-5 relative accuracy. The measurement result from the optical system displaced good correlation with those form a mechanical probe on a coordinate measurement machine. A few teeth/second measurement speed can also be expected. Different factors to effect the measurement accuracy are discussed, and the possible solutions are provided.

Although laser sources offer more advantages over white light sources in some phase-shifting projected fringe profilometry applications, these advantages are at the cost of speckle. This paper presents some basic statistics of the speckle-induced phase measurement errors that are investigated based on the multiplicative-noise model for image-plane speckles. The dependence of the phase error distribution and measurement uncertainty on the speckle size and grating pitch is found based on the Karhunen-Loeve expansion of speckle field. This analysis shows that phase errors caused by speckles can be modeled as additive white Gaussian noise. Based on these simulation results we can discuss the designs of the optical system and noise reduction algorithms.

A nondestructive laser profiler based on optical scanning method using a commercial compact laser triangulation displacement probe is described to measure the profile of the pipe inner wall surface. According to the optical feature of measured pipe inner wall surface, the adjusting posture method of laser triangulation displacement probe is put forward to reduce the disturbance and increase the measuring accuracy. the experiments are given. The method of the wavelet transform to analyze the measure results is used. The resolution of the nondestructive laser profiler is 30 micrometers .

We present a new concept for 3D shape recovery using Defocused Structure Light (DSL) images. DSL technique externally extracts the depth information for the scene by using projections of cylindrical wavefronts on the object. These projections show different degrees of defocus as a function of the depth.

This paper describes the principle of the spatial encoding- based rangefinder. Its parameters to be calibrate are found out. Mathematical expressions are deduced of its measurement errors caused by calibration errors. Analysis shows that it is difficult to calibrate the parameters separately, especially some parameters are unable to calibrate. So we present a new calibrating method based on table. In the normal operation of the rangefinder, it directly obtains the ranges the pixels of the image of the object from the table after giving its codes. Our calibrating method has the advantages of simple and high accuracy.

3D measurement systems based on active triangulation using projection of structured light and recording in CCD- cameras are gaining more and more importance in industrial applications. Their advantages compared to conventional coordinate measurement machines are the non-contact measurement and fast acquisition of dense point clouds.

This report documents system uncertainty, best practices, and process limitations for the real-time videogrammetry system developed by Metronor, ASA. The report describes the system and the tests used to establish its performance. It then documents the test result and draws conclusions. An uncertainty of 23.6 ppm within the 3.2 by 2.8 by 1.5-meter test volume was achieved with the baseline between the cameras parallel to the length of the target field. A slightly lower uncertainty of 18.5 ppm was achieved when the cameras were setup with their baseline normal to the length of the target field.Results suggest that the orientation of the camera baseline relative to the target field does not play a significant role when targets are more than 2 meters apart. Testing shows a significant dependence in system accuracy relative to the apex angle between the cameras and the targets. When the apex angle was between 60 degrees and 95 degrees the performance specification was met. Conclusions drawn from a series of two-sample F-tests of RSS error values establish that sampling a point more than three times does not significantly reduce its uncertainty. This result is in part explained by the redundant fitting of the five-diode points on the light pen: the fitting effects an averaging process even on samples of one.

The laser range-finder presented here has been designed as a part of a non-destructive testing system. The whole conception of the system is detailed. The method is based on phase-shift measurement, but using electronic heterodynage to provide a good resolution. An RF card is the heart of the system, including two VCO and two active mixers, providing the modulation signal and both reference and phase shifted signal at intermediate frequency of 125 kHz. The emitter is a laser diode at 12 mW average output power, the receiver uses an avalanche photodiode. We also find an amplification board with an 80 dB dynamic range AGC, and the frequencies control board. The phase shift is estimated by direct counting with more than 10 bits resolution, that means about 0.4 mm for a 60 cm range without phase ambiguity. The amplification level, obtained from the AGC loop, gives an estimation on the measurement quality and on the target surface. A PC computer controls the range-finder through an interface specially developed. It is based on two PLD, one for the general sequencement and one for the averaging of the data. The obtained data are stored in a 2Kb FIFO memory waiting a data request for the PC or generating an interrupt if it is full. The interface is easy to drive from the PC with the choice for averaging value, and amplification criterion. The imaging should be provided by scanning mirrors. Measurement of the target distance and estimation of the target surface conducts to laser imaging. Altogether the system is rather compact due to the disposition of the cards and also rather low-cost.

This paper is a preliminary description of three technologies for use in scanning and printing. There aren't a lot of experimental data here, unfortunately, because the ideas are new. They came out of a current effort to build a pocket-sized, battery-operated, non-contact 3D input device. The concept of this pocket 3D scanner is to allow someone to take simultaneous range and intensity images of a 10-50 cm diameter area in half a second, store a hundred or so of them, then play them back into a PC IR pot for OCR, printing, archival storage, or further processing. Such areas include flat or crinkled paper, hands and faces, machined parts, textures, and many others. Besides their use in input devices, these technologies could greatly improve the performance of low-end printers, at very low cost. None of these techniques is yet at a high state of development. The first is a scanning technique that should allow increasing pixel rates by a factor of 10 or more without significant additional optical or mechanical complexity; the second is an extremely fast focus actuator that should reduce the field flatness and accuracy requirements of the scan lens and scanner assembly, by allowing fast focus correction even within a scan line; an the third is a 'mass customizing' wavefront aberration correction method for producing very high quality laser beams from low quality optics, without requiring any hand work.

The drive to short development times and closed-loop process control has created a demand for new tools to collect the needed dimensional data. Optical technologies in fields such as sensors, signal processing, metrology, and instrumentation offer unique solutions to many areas of monitoring, diagnostics and control. The Advanced Fresnel Tracer (AFT), an innovative instrumentation for in-process micromeasurement consisting of a smart optical sensors and an automatic follow-up system, based on a temperature controlled grated glass scale or interferometer will be presented. This device may readily be integrated into a turning or grinding machine, e.g. for the needs of quality assurance and to enable an on-line automatic compensation of diameter deviations/1/2. The device contains an optical Fresnel diffraction sensor allowing a fast measurement of the surface topography, achieving three goals: 1) improvement of the instantaneous diameter measurement, 2) surface quality inspection, and 3) determination of the edge gradient or the waviness of the workpiece. The new compact, smart, and precise optical multiparamter sensor, the AFT has been developed and tested.

The surface metrology market toady is moving towards non- contact modular computer-controlled system for measuring and analyzing roughness, contour and topography. In this paper we present a new optical instrument based on the concept of confocal microscopy. In this instrument, which is especially suitable for measurements on smooth surfaces, either a pinhole or a structured light pattern in imaged by a very high numerical aperture optical system on the surface of the sample to be measured. The reflected light is observed wit a CCD array and analyzed with different image data processing algorithms. Two different experimental prototypes were developed to allow the measurement not only of surfaces with good accessibility but also of those with intricate geometries, difficult access and small dimensions. Various samples such as high precision optical surfaces, master gratings, and diamond drawing dies were measured. All the results obtained show that the confocal optical profiler is robust enough to provide a surface topography with spatial resolution lower than 1 micrometers and uncertainty about 10 nm. In addition to the replacement of the existing stylus system, there are also important new potential applications for this kind of instrument.

The printability of newsprint and liner is addressed through the analysis of CLSM images. The surface topography and porous structure of the paper are physical properties heavily correlated with the paper's formation, and, hence, important factors affecting the printability of the paper. The surface 3D structure of the paper is measured by using topographic CLSM images from which the pore size distribution can be estimated using image analysis. The experimental results have both been compared to a theoretical model of the cross-sectional area of void volumes in paper and also with paper physics measurements, specially related to the offset print-through problem. Good correlations have been found, and results so far are promising, but still there is a need for improved analysis of the 3D structure of paper, both on the measurement and modeling side.

As the micromachining techniques are developed, in-process measurement and evaluation technique for the 3D micro- profile in the microparts production system are required. In this paper, we propose a new optical measuring method which can be applied to in-process measurement of 3D micro-profile with an accuracy of nanometer order. The laser inverse scattering method has a possibility to measure 3D micro- profile in the whole area of a workpiece which is illuminated by the laser beam simultaneously without scanning. The 3D micro-profile is reconstructed by measuring only Fraunhofer diffraction intensities. The principle of measurement is based on the optical Fourier transform and phase-retrieval technique. The hybrid phase-retrieval algorithm as a fusion of the logarithmic Hilbert transform with the Fourier series expansion and the iterative algorithm is developed. We proposed the process error evaluation system by employing the laser inverse scattering method with introducing design model based scheme and the computer simulation system of measurement procedure is built up. In order to verify the validity of the proposed method, the computer simulations for the 3D profile such as microparts fabricated by photolithography process are performed. We show that the proposed method is effective for evaluating defective 3D micro-profile result from process error.

The aim of this project was to propose an original approach c to predetermine the topology of a surface from the point cloud obtained during a digitizing process. This method has been applied for the predefinition of the different main regions of a surface, in order to be able to divide the point cloud in different separate clouds corresponding to these regions. It should be also possible to optimize locally the surface when taking into account tangency and curvature conditions. The main advantage of this method is that, for a given accuracy, the complexity of the surface is lower than a traditional method, an the computations are quite the same than those obtained when using traditional surface modeling algorithms. This method allows a qualitative decomposition of the surface defined from a function z equals f(x,y). In the paper, the fuzzification and defuzzification process and a study of the influence of the main parameters will also be presented. Some fuzzy logic algorithms and functions have been compared and analyzed when applied on the example.

The tripod operator consists of a class of feature extraction operators for range images. Each tripod consist of three points in a 3D space fixed at the vertices of an equilateral triangle. A triangle is moved as a rigid body unit the three vertices lie on the range image surface. Several angle measurements can be made from the linkable tripod operators. The resulting angles between linkable triangles are local shape features which are invariant to scale, transformation and rotation. In this report a new approach to shape classification is proposed. The first step in the approach is to take angle measurement through random placement of the tripod operators in a range image. The second step is to generate two histograms based on two resulting angle features generated from the first step. The third step is to extract a number of invariant features, such as SIGN of coefficients, using the Fourier transforms of the two histograms. Based on the analysis of these invariant features, 3D surfaces can be classified. The experiments show that tripod operators can generate a large number of surface features for object classification and further for recognition.

Implicit surface methods for surface reconstruction have received much attention in recent years. But implicit algorithms reported in literature are usually limited to a class of closed surfaces or require knowledge of the surface topologies prior to surface reconstruction. Here, a general implicit approach is proposed for surface reconstruction of various types of surfaces, including manifolds with boundaries and non-manifolds, without a prior knowledge of surface topology. A framework and an associated methodology have also been defined to identify the main tasks an the issue in surface reconstruction. An assessment of the methods and algorithms reported in surface reconstruction has been also carried out.

This paper introduces the first results of a research work carried out on the automation of digitizing process of complex part using a precision 3D laser senor. Indeed, most of the operations are generally still manual to perform digitization. In fact, redundancies, lacks or forgettings in point acquisition are possible. Moreover, digitalization time of a part, i.e. immobilization of the machine, is thus not optimized overall. After introducing the context in which evolves the reverse engineering, we quickly present non-contact sensors and machines usable to digitalize a part. Considered environment of digitization is also modeled, but in a general way in order to preserve an upgrading capability to the system. Machine and sensor actually used are then presented and their integration exposed. Current process of digitization is then detailed, after what a critical analysis from the considered point of view is carried out and some solutions are suggested. The paper concludes on the laid down prospects and the next projected developments.

The main concern of this paper is to develop a technique for analyzing a set of binary image taken for a 3D solid object to determine its basic physical properties. The binary images are easier to acquire, sore, and process than images in which many brightness levels are represented. The physical properties investigated in this paper include mass, zeroth and higher orders moments, surface area and flux density and representation of object in the Euclidean space. The developed techniques may be useful in many engineering applications such as robots, medical diagnosis, non- destructive testing and process control. Computer simulation is used to test the efficiency of the developed techniques with encouraging results.

We report on a novel set-up using image shearing speckle pattern interferometry for the determination of 2D strain distributions of an object surface. This system is based on simultaneous illumination of the object with three diode lasers that emit at different wavelengths between 810 nm and 850 nm. Their speckle images are separated within the shearographic set-up, consisting of a variable shear element, a special color separation optics and 3 b/w CCD cameras, in such a way that each camera records the speckle image corresponding to one laser source only. The shearographic camera in combination with the appropriate illumination geometry allowed us to isolate all six displacement derivatives from phase stepped fringe patterns. The good suitability and accuracy of the system for the determinant of 2D strain distributions are demonstrated on the basis of shearographic measurements during tensile and comparison with strain gage measurements.

Birefringent optical materials can be used to convert mechanical strain into fringe patterns of optical intensity which have typically been used to measure surface stains or stresses. In this paper a system will be described that uses a photoelastic transducer, linear sensor array, and neural network image processing to estimate the load torque for stationary and rotating motor shafts up to 1500 rpm. A photoelastic polymer coupling is attached to the shaft, and illuminated by polarized light. As the shaft torque varies the photoelastic plastic coupling experiences torsional strain. This results in a corresponding 2D fringe pattern when viewed through an optical polarizer. The strain that causes this observed pattern in a complex function of the applied torque applied to the shaft. A neural network is trained with the fringe patterns corresponding to calibrated load torques as measured by a laboratory strain gauge torque sensor. Experimental results show that the neural network torque estimator can accurately estimate the applied torque for both static and rotating shafts.

In its classical application Electronic Speckle-Pattern Interferometry (ESPI) is used to measure deformations with high resolution. Additionally, this methods also able to measure the 3D topography of technical surfaces even with discontinuities. If adequate set-ups are used, combined measurements of shape and deformation can also be carried out. Especially in production of machines and other metal parts, the determination of the stress and strain e.g. of welding points is a very important issue for evaluating about the quality of the specimen under test. Here, the use of wire resistance strain gauges is state-of-the-art for measuring length variations of parts under mechanical load. It is very time-consuming to prepare the corresponding measurement environment for strain gauges. Moreover, only very limited information about the strain can be measured by this means because all information is integrated over the whole area covered by the strain gauges without lateral resolution. In order to extend this kind of metrology to a matrix of some thousands points even including sensitivity in the out-of-plane-direction, ESPI-methods can be used. As described in this paper, it is therefore necessary to perform both,the shape and the deformation measurement to obtain the necessary information. Of course, the directions of sensitivity depend on the contour of the test specimen ad can be determined due to the previously measured topography. In this paper current work on the field of stress and strain measurements with ESPI is described. The experimental result of several technical applications is shown and it is compared to measurements with conventional strain gauges. Further possible technical applications are discussed and the prototype of an ESPI stress sensor is presented.

This paper presents an optical measurement method of transient current. The current is induced by a Rogowski coil, and the induced voltage together with other constant source is used to drive a linear LED and make the light be modulated by the current. The modulated light is transmitted to electronic circuit by optical fiber and converted to voltage signa. A microcontroller processes the signal and types the current waveform.

We developed an on-line measurement system for the simultaneous measurement of the root-mean-square roughness and autocorrelation length which are the parameters of surface roughness. The measurement is based on the scattering theory of light on the rough surface. Computer simulation shows that the measurement range depends on the wavelength of the light source, and this is verified with the experiment. We installed the measurement system at the finishing line of a cold-rolling steel work, and measured the two parameters in situ. The rms roughness and autocorrelation length are measured and transformed in the average surface roughness and then umber of peaks per inch, respectively. The measured data for both of the parameters are compared with those of stylus method, an the optical method is well coincided with the conventional stylus method.

We describe the detailed design of a geometrically desensitized interferometer using two transmission diffraction gratings. A number of models of the instrument are used to eliminate object ghosts and stray light contributions. We then investigate analytically the influence of object slope variations on the instrument precision. We show that the part can be located at a measurement location where the metrology is optimized. Analytical and raytracing models demonstrate excellent agreement with experiment.

A geometrically desensitized interferometer (GDI) uses two beams incident on the same sample area at different angles of incident to generate an interference pattern with an equivalence wavelength larger than the illumination wavelength. The instrument is well adapted to the metrology of both smooth and rough samples that are beyond the range of conventional interferometers, while providing more accuracy than conventional moire techniques. In this paper, we extend the capabilities of a GDI with an equivalent wavelength of 12.5 micron using coherence scanning in a manner similar to that of scanning white light interferometry. We also present new analysis techniques to accommodate speckle phenomena that can be more prominent in GDI than in white light interferometry. Our scanning GDI can rapidly characterize the surface flatness and relative heights of discontinuous surface features over large measurement volumes.

The report presents several examples of holography applications in metrology: automatic control of spatial water solution concentration during crystal growth providing large homogeneous crystals such as KDP, DKDP, and alike; detection and measurement of defects in crystal and precious stones; measurements of the distribution of droplets in diesel engine sprayers; flow velocity distribution in spillway of river dams; optimization of blood freezing process to prevent cell damage; detection and measurement of the static and dynamic objects behind non-transparent media.