The testing of spherical surfaces using the three-measurement technique outlined by Jensen requires very precise alignment of the sphere relative to the interferometer. An easier technique for the absolute measurement of spherical surfaces has been developed which does not require the precise alignment of the Jensen technique and uses only two measurements. As long as the test surface does not contain any aberrations with odd symmetry, these aberrations can be subtracted from the measurement and an absolute measurement of the test surface can be obtained. This paper describes and compares these two techniques and shows results of testing a A112 P-V (jeak-to-valley) sphere (N.A.=0.4) using both techniques with a phase-measuring Fizeau interferometer. These measurement techniques are repeatable to waves PV.

Surface deviations of spherical mirrors from a best fitting, mathematically ideal sphere were measured to an absolute precision of 0.25 nm rms. Because of the long radius of curvature, a Hindle-type arrangement was used as interferometric setup, resulting in a test arm length of about 1.4 m. A special calibration procedure was implemented to eliminate systematic, setup-dependent errors. A very fast data acquisition technique was combined with real-time wavefront averaging to eliminate the effects of random errors, such as wavefront variations due to the turbulent atmosphere in the beam path. For the evaluation of one mirror surface, all in all 400,000 individual wavefront measurements at 400 x 400 points were combined, requiring an overall measurement time of only one to two days.

We will report on a new interferometer developed at Carl Zeiss, which has real-time measuring capability with instant visualization of results, is nearly insensitive to vibrations, has a variable fringe spacing from one lambda to lambda/1O (lambda represents the wavelength of the light used in the interferometric test), and can give lambda/100 accuracy through a simple calibration procedure. It can be handled with the same ease and in just the same way as conventional interferometers.

Prisms are common in many optical instruments, however, widespread application of coherent sources has tightened the already stringent requirements set on these in principle simple optical components. One of the main problems in manufacturing prisms is the fast measurement of the angles. The objective of this paper is to suggest an instrument for the simple nd rapid testing of angles in prisms, thus saving valuable time for the technician.

The Fresnel rhomb is used for the production 0of 0cirularly polarized light. In this paper, we propose a set of two 45 -90 -45 prisms which is equivalent to the Fresnel rhomb. The analysis of polarization properties of such a set of prisms was done and it was found that fo orginary experiments on polarizatin, it may satisfactory to use two 45 -90- 45 prisms for producing the circularly polarized light. It is not necessary to fabricate a special prism i.e. the Fresnel rhomb with special angles. If the faces of the prisms are coated with antireflection coatings, a high degree of circularly polarized light is obtained. The polarization calculations were done and verified by experiment.

Diffraction at a low frequency grating is used for the measurement of focal length of a positive lens.

Industrial inspection of optical component surface flaws requires objective, high efficient and fast measurement methods and instruments. In this paper, a novel method, which is practical for on- line inspecting optical component surface flaws in manufacturing industry, is discribed. Laser beam goes through a lean- placed mirror with a slot in center onto the surface of the specimen. The imformation of surface flaws is obtained through analysing the frequency spectrum of reflective light which is detected by a photomultiplier, the specimen scanning control and signal processing are finished by a low - cost and handy single- board microcomputer. The theory that applies the scanning frequency spectrum method , the method for determining flaw size and measuring sensitivity as well as control model for various specimen are analysed in detail .A system has been built according to the idea discribed above. By using the system, several specimen are measured, the comparison and analysis between exprimental results and actual flaw conditions are given. The minimum detectable flaw is 3 micrometer, the measuring error is also given.

Several techniques for testing aspheric surfaces are presented, with the emphasis on interferometry. Techniques for setting up configurations to produce easily understandable null interferograms are described. Attention is also given to techniques that improve the readability of the more complicated interferograms produced when an asphere is tested at its center of curvature.

This paper describes an interferometer system for testing aspheric surfaces with computer-generated holograms (CGHs). A CGH written by electron-beams achieves the precise testing of a wide range of aspheric surfaces. A new algorithm removes the error caused by misalignment of a CGH and a testing surface. This system can test aspheric surfaces with accuracy 2/10 (P-V value) and repeatability /5OO (RMS value).

This paper addresses the problem of testing generalized aspheric surfaces. The technique involves the use of diamond machined null reflective compensators. These null mirrors can be designed, fabricated and qualified in-house quickly which is a big advantage in a production environment. Methods of null reflector design as well as test configuration alignment will be discussed. Several examples of this method will be used to illusirate the technique.

The optical aspheric surface profiler was developed. In the profiler a phase shift interferometric technique is used to measure an optical wavefront precisely and modified to be applied for an aspheric wavefront measurement. The profiler makes it easier to measure aspherical surfaces than the profiler using an optical probe or a stylus. The intensity distributions of fringe patterns from a Twyman-Green interferometer are obtained by a linear image sensor while a phase shifter driven by a piezoelectric transducer (PZT) moves. The charge coupled device (CCD) linear image sensor, which has smaller pixels than an area image sensor, detects fine fringe patterns from aspherical surfaces. The fringe data are transmitted into a computer and calculated into the wavefront phase data. In order to analyze the phase data, a non-linear least squares phase fitting method is used to eliminate the effects of the setting error of a tested surface. The profile error from the designed surface, whose parameters are previously saved into the computer, is calculated. The measurement results for the aspherical mirrors were compared to another profiler using a stylus. Finally, the new approach to analyze the phase data is proposed to solve the problem for this profiler and simulated for some data.

Some simple methods for calculating the radii of curvature of the null mirrors for a given Ritchey-Chretein system are described. The cases considered include the null for the concave primary, the null for the convex secondary, and examples of unobscured design, on-axis design, and off-axis design.

Nondestructive inspection by holographic interferometry (HI) is quickly gaining acceptance in the electronics industry as a sensitive and accurate method of locating manufacturing and assembly flaws in a wide range of electronics, from individual components to assembled modules. This paper describes the specific application of real-time HI in the nondestructive analysis of circuit board heat exchangers and multiple-layer printed wiring boards to locate areas of debonding and delamination. In the application of HI, the choice of a stressing method is often as important as the choice of a specific HI technique. Methods for component stressing include thermal, vibrational, and pressure-induced stressing methods, and these are described in detail. In addition, two techniques for sensitivity enhancement, phase shift interferometry and beam tilt correction, are discussed in detail.

An external corrective filter is used to achieve image reconstruction from transmission holograms in white light. In effect, the corrective filter re-distributes the light diffracted from a suitably oriented hologram by stretching it horizontally and perpendicular to the direction of the dispersion of the hologram. The corrective filter, that can be described as a one dimensional random diffuser, also serves as a projection screen upon which the real image from the hologram is displayed. The method has been applied to study distortion of surfaces and aberrations of optical systems.

Image processing has been used to analyze TV-holographic fringe patterns of deformation and vibrations. Recent results using such image processing algorithms for deformation and vibration measurements are described. It is shown how these techniques has been used on practical objects such as concrete, high-frequency transducers, and underwater transducers.

Measurement requirements and candidates for measuring crystal growth in space are described, emphasizing holographic instrumentation. Existing instrumentation planned for the IML-1 Spaceflight is described along with advanced concepts for future application which incorporate diode lasers, fiber optics, and holographic optical elements. Particle image displacement velocimetry in crystal growth chambers is described.

Van der Lugt optical correlation, involving Fourier transform holography, has been shown to be successful in detecting and accurately measuring leak rates in hermetically sealed packages in the range of 10' to iO6 atm cc/sec. The technique depends on the measurement of the relaxation time of sealed packages under vacuum conditions, from which leak rates can be calculated. This technique has the advantage over conventional 'leak testing, in that it requires no backfilling with tracer gas, and therefore is more cost effective and faster. Also, packages such as pacemakers can be examined at the end of the manufacturing process, after the polymer neck has been inserted, eliminating any absorption of the helium or radioactive tracer gas that normally reduces the accuracy of conventional methods. Furthermore, this new technique can be used to rapidly test packages with large internal free volumes, in that it requires only a small differential pressure across the walls of the package, and avoids measurements depending on slowly changing internal helium or tracer gas concentrations. A comparison will be made between this HNDT technique and other leak testing methods commonly used in the packaging industry, in terms of cost and time factors, and range of accuracy.

A mathematical model relating deformation to electrochemical behaviors of metals in aqueous solution has been developed. The model was established based on principles of holographic interferometry for measuring microscopic, deformation and with those of electrochemistry for measuring the corrosion of metals, i.e., anodic dissolution current. The model can be utilized to predict the mechanochemical behavior i.e., stress corrosion, of metals under a constant load or displacement in a polarized or unpolarized condition. In other words. the current density of metals can be predicted as a function of imposed deformations. In the present work, the model is used to predict the stress corrosion behavior of a Nickel electrode under a constant load in iN H2S04 solution. The prediction is obtained based on a hypothetical optical arrangement for the holographic interferometry as well as on empirical parameters of Ni electrochemistry in the solution.

A fundamental study on the effect of deformation on the electrochemical behavior, i.e., corrosion, of metals in aqueous solution has been conducted. The study was successful in developing a novel experimental technique for investigating the influence of deformation on corrosion. The development of the new technique is established based on incorporating electrochemical methods with those of holographic inferometry. In other words, the technique is capable of measuring microscopic deformation and the corrosion, anodic dissolution current, of the surface of metal, simultaneously. As a result, data on stress corrosion tests of a Molybdenum electrode in 0.75 N KCI is given in this paper. The data are plotted against a mathematical model developed in relation to the present work.

The effect of imperfectness of detector's characteristics on measurement accuracy in phase-stepping interferometry is investigated by numerical stimulation method. Based on the results, the determination method of system parameters to achieve desired accuracy requrement is discussed.

The distortion of wavefronts by the process of light-in-flight recording by holography is graphically and mathematically investigated. It is shown how distortion caused by the holographic process can be used to compensate for distortions due to relativistic effects caused by the limited speed of light.

An optical method for the precise measurement of the angular errors of a table moving along a slideway is described. The system is arranged so that angular errors of the table give rise to sideways shifts of a random wavefront generated by a ground glass screen. The shifted wavefront is made to interfere with a reconstruction of the original unshifted wavefront produced by a hologram, and the angular errors determined from the interference fringes which appear. This technique has several advantages over usual autocollimator methods.

The structure and operation o a holographic intererometer that employs photorefractive crystals (o the sillenite Family) as reusable recording materials and computer image processing or rapid and precise interpretation o the fringe patterns are discussed. Different types of two-wave mixing (2WM) and four-wave mixing (4VM) were applied to optical testing by timeaverage and double..exposure hoaographic intererometry. A very convenient matching was found in the setup or the BTO crystals and usual Re—Ne lasers, when the exposures were o r.lOs or light energy densities o 4 x lO" W/cm2.

Recent work on phase conjugate resonators and their bistabilities and applications is reviewed. Emphasis is given to resonators involving degenerate four-wave mixing and those involving Brillouin enhanced four-wave mixing.

The properties of phase conjugate (PC) beams generated by an erythrosin/polyvinyl alcohol film are briefly reviewed along with demonstrations of PC interferometries, and an experimental study of the time dependence of the phase difference between saturable absorption and holographic components of the PC signal is reported. The results show that the phase difference between two components of PC signals changed temporally with a period of about nine minutes and reached 110 deg five minutes after the initiation of degenerate four-wave mixing at the pump beam intensity of 1 W/sq cm.

An application of a phase conjugate Twyman-Green interferometer for testing a parabolic mirror is demonstrated. The interferometer is free from aberrations due to the self-focusing property of the phase conjugate mirror in one arm of the interferometer. It does not require a precision spherical mirror in the reference arm.

In the paper, the principle of phase shifting interferometry has been briefly described. B using this technique to the non-destructive testing of solder joints on printed circuit board, the defective solder joints can be easily distinguished b the phase changes on the joints. A suitable aglorithm for the processing of the interferogram was also given. Keywords: Holographic interferometry, Non-destructive testing Phase-sh I ft

Industrial flaw detection is not always amenable to template matching approaches. Some flaws have fuzzy descriptions which serve better than prototypes for representation. Optical Fuzzy Syntactic Pattern Recognition is well adapted to this situation.

A range evaluation system is presented which uses a unique method of fringe projection to establish the range of points upon the surface of an object, viewed along the optical axis using a high resolution sensor. A range profile is obtained and by computing over a matrix of points, which represent a segment of the surface under view, the topographical form of the object can be determined.

A laser stereo triangulation device is described which permits easier observation of objects with specular surfaces. The device's two cameras are first calibrated with the light pattern. The calibration data are used to compute the unique and reversible function to mapping the image sensed by the second camera within the image plane of the first camera. Simple correlation techniques are used to extract the originally projected light pattern from the numerous artifacts. Because of its simplicity, the algorithm can be computed in real time at video rates to produced a clean image representative of the surface geometry. Experiments conducted using a laboratory prototype produce reliable results wih promise for practical applications.

This paper describes a visual inspection system for factory automation. The system is based on a multi-directional 3-D imager. Three dimensional object recognition has become increasingly important in factory automation. For example, automatic assembly of printed circuit (PC) boards can use a 3-D visual inspection system to detect incorrectly assembled devices. To be effective, measurement should be done from more than one angle. A visual inspection system has been developed based on a multi-directional 3-D imager and laser cross scanning. It can obtain range and intensity information of objects simultaneously. Range measurement is based on laser triangulation using a position sensitive detector. The system features: (1) Multi-directional 3-D measurement. The quad 3-D imager and X-Y laser scanner enable multi-directional 3-D measuremenL (2) High-speed. Measurement speed is 1 million pixels per second. Each pixel contains data for 256-height-level range and 256-gray-level intensity. One quad flat package with 160 leads can be measured in 4 seconds. (3) High-resolution. The inspection resolution is 25 jim in the X and Y directions and 30 pm in the Z direction. The visual inspection system uses the 32-bit MC68030 and 12 megabytes of image memory. The system was capable of detecting missing, shifting, and floating leads, and solderjoint defects.

A novel edge extraction method that employs an active defocusing technique is presented. The method is based on the principle that a Laplacian of Gaussian(LOG) operator can be well approximated by a Difference of Gaussian(DOG) operation.. While such operation can be done by digital processing, it is more effective to be conducted in a combination of optical and digital processing techniques. In this edge extraction process, a focused image of object in scene is first acquired. Image of the scene is then slightly defocused by changing the focal length of camera lens. Real time subtraction is applied to the defocused and the previously acquired images. It produces a residual image that emphasizes abrupt intensity variations, which are typical of edges in the image. An objective evaluation called edge index is performed on the resulting image. Amount of defocusing is carefully adjusted according to this measurement so that a desired edge image is generated. Boundaries of objects can then be obtained by further enhancement of the edge image. Since this edge detection method is an optical-based process aided by digital processing, it is rather fast and less expansive.

This paper describes a system calibration and part alignment approach for inspection of 2D electronic circuit pattems. Fast and robust techniques for measuring image features in these procedures are presented. The approach and techniques have been validated on a prototype system implementation consisting of a PC-AT, a stationary microscope with a CCD camera interfaced to a commercial frame grabber board, and a rotation platform mounted on an X-Y stage that transports the part during inspection. System calibration finds the parameters that map pixels into physical units and that map X-Y stage units into physical units. Each mapping is modeled with a 2D affine transformation obtained by least-squares fitting measured image features to physical units. System calibration also determines the axis of the rotation platform accurately and corrects for nonuniformity in pixel sensitivity. Features of a calibration part are detected with robust techniques that reduce the likelihood of spurious detection and hence relax constraints for a completely contamination-free calibration part. Subpixel measurements are conducted in edge profiles of row and column projections of image sections. Edges of opposite polarity and similar magnitude which are sufficiently close to each other are grouped into pairs and considered spurious. Feature detection is conducted at two different image regions to verify authenticity and obtain a more accurate measurement. The axis of the rotation platform is calculated by rotating CW and CCW and measuring corresponding features of the calibration part after each rotation. Measurement of rotation during this step and in the alignment procedure is obtained by mapping a quadrilateral in 2D measurement space onto a rectangle. This compensates for minor global variations in height that the part may exhibit due to placement. It also makes the alignment procedure independent of knowledge of the physical coordinates of the features that defme the vertices of the rectangle. Analytical solutions for the measurement of rotation and the axis of the rotation platform are presented.

In this paper, a simple and effective algorithm for the generation of look-up range table in triangulation is presented. Based on this algorithm, the look up range table can be generated by imaged several points of different height of the sample plate on CCD instead of system parameters.

The reflectional display method for information extracting by the speckle pattern is presented in this paper. Combining objective speckle and reflection holography, this method reduces the restriction to optical density of the film. Two applications are presented: the in-. displacement of a specimen of the three point curved bean of coiplex material and the homogeneous degree of deformation of the bearing cap of top motor under axial load are measured respectively by the objective speckle and reflection holography..

A new optical fiber sensor with 2 light beams is developped for detecting variation of scattered light distribution caused by surface defects, like flaws, surface inferiority, etc. Its feature consists in compact structure using only light transmission characteristic proper to the optical fiber, using no optical components like a lens. High response speed of maximum 2m/s is suited to purpose of non con- tact in-process or in-line defect detection and with automation of inspection process performed in the past visual or sampling method, the stabilization of inspection level is achieved and inspection eff i- ciency is improved by introduction of this sensor. In this paper, the principle of surface defect detection with the optical fiber sensor and the quantification of surface properties by conversion of detection output into arithmetical mean deviation are described, and the system structure on representative examples of practical application in actual inspection process is explained.

The relative merits of a number of fiber optic sensors are compared in terms of their suitability for use in smart structures. It is shown that the Michelson fiber optic sensor is suitable for detecting damage-induced acoustic emission within composites in the laboratory and for testing the concept of an optical strain rosette. However, its limited application in terms of smart structures has led to the development of a Fabry-Perot fiber optic sensor for strain sensing within composite structures. A damage assessment system for composite material structures based on embedded optical fibers has been developed and fabricated within the composite leading edge of an aircraft. The results clearly support the concept of structurally integrated fiber optic damage assessment systems for composites.

In order to achieve centimeter scale spatial resolution along an optic fiber, it is necessary to employ optical sources and detectors with responses on a 100 ps time scale. In this paper, the results are described of an investigation into the limiting capability in terms of spatial resolution, return signal strength, and data acquisition time which can be achieved using a photon counting method of detecting the backscatter signals from a distributed sensor based on standard telecommunications fiber. Particular attention is given to the possibility of obtaining 'real time' update rate on the sensor output while maintaining the centimeter-scale resolution capability.

A novel image scanning technique, the pulse counting method, has been developed for geometric parameter measurement of optical fibers. The theory and implement of this technique are described. Experimental results show that the resolution and the repeatability of the system with this technique are O.OO5pm and O.lpm (3a) , respectively. Indexing Words: Image Scanning; Geometric Measurement; Optical Fibers;

The sensitivities of interferometer-based fiber optic sensors to both strain and temperature are compared. The first compact fiber optic strain rosette has been developed and its performance is compared with a conventional foil strain rosette when embedded within composite material. It is shown that an embedded Michelson fiber optic sensor can detect acoustic emission resulting from threshold delaminations within composite materials.

First test results of a multilayered fiber optic impact damage detection system fabricated within an aircraft wing composite leading edge are reported. The graphically presented results indicate that embedded optical fiber will track the growth of a delamination region. These results strongly support the concept of structurally integrated fiber optic damage assessment system for composites.

It is shown how spatial division multiplexing (SDM) offers a low-cost solution to polling large numbers of passive optical sensors in large sensor arrays and also provides the possibility of low-cost fiber optic sensor systems. The geometric versatility of SDM allows optical multiplexing to be at the source, in the fiber distribution system, or at the detector. In most cases, electronic scanning can be driven from low-cost microcomputers. The simplicity of SDM promises reliable and maintainance-free operation.

A novel image scanning technique, the pulse counting method, has been developed for geometric parameter measurement of optical fibers. The theory and implement of this technique are described. Experimental results show that the resolution and the repeatability of the system with this technique are O.OO5pm and O.lpm (3a) , respectively. Indexing Words: Image Scanning; Geometric Measurement; Optical Fibers;

Use of GRIN Fiber Lenses as SMF-to-SMF connectors is analyzed. Coupling conditions are derived under Gaussian mode approximation in order to obtain maximum coupling efficiency.

An optical triangulation technique using a solid state laser diode sourceand detector, providing sub-pixel resolution on a CCD detector was developed and is discussed. A novel approach to the problem of non-linear accuracy is outlined. Results of extensive research into improving the accuracy of the device with particular reference to configuration, optical components, the laser source, linearity, resolution, calibration and interpolation are presented.

Design and development of a fiber optic polarization sensor for pavement maintenance application is reported. Sensor is designed to measure stress distribution inside the sample caused by applied loads on the sample in the range of 0 to 1000 pounds. Sensor reaction to the temperature variations in the range of 70 °F to 90 0 F is investigated.

It is shown that improvement in the detection sensitivity of laser-ultrasonic systems may be obtained by generating narrowband acoustic signals using both temporal and spatial modulation of the generating laser. A laser-generated acoustic tone burst waveform will have lower peak amplitudes than a single acoustic pulse providing the same system SNR. Consequently, lower power density laser pulses may be used to avoid surface damage.

The high-spatial--fregiency behavior of rrethanical-profiling instrunents is determin principal ly by the non-linear geaitrica1 interaction beten the styl us tip and the surface irreularities. 'there is cons iderable interest in the nature of this interaction and its effects in practical applications. This paper explores the these issues analytically and pesents a riety of results with enphasis on the freuency-danain description. Smooth surfaces are treat1 by perturbation theory and are found to be rot4iened by tip-s ize effects , tiile rough surfaces are snoothed. In lieu of closel fotm results for rot4i surfaces, offer a simple conjecture rerding the nature of tip-s ize distortions and the rnaxinurn useable spatial freuenoj.

The qualification of semiconductor surfaces is an important aspect of IC lechnology, from checking the quality of the substrate material right through to the dimensioning of the finished device. Etching the surface of the substrate reveals details concerning chemical and crystal defects. Surface roughness measurements are required before preparing epitaxial layers and device processing. The height, width and shape of component features need to be checked throughout the device processing stage. In this paper we describe a three dimensional NDT optical profiler we have developed for the nanometrology of semiconductor surfaces. The system is based on interference microscopy, using phase stepping for automatic fringe analysis. Illumination is by high intensity LED or white light and detection is by a CCD camera. A vertical resolution of mm is achieved, with the lateral resolution being better than O.5i.m. Comparison of a two dimensional profile of a chemically etched surface using PSM with that obtained by a stylus profiler are within 10% agreement. Results are shown of three dimensional profiles of chemically etched InP and an MOCVD grown epitaxial layer, surface roughness of polished InP, and profiles of a recess in a combined laser/photodiode device.

The use of the phase shifting interferometric technique is discussed to make quantitative surface profiling using the Nomarski differential interference microscope. Lateral shift of the Nomarski prism introduces mutual phase shift between interfering two wavefronts with small amount of shear. Since the analyzed phase distribution corresponds to the differential of the surface profile under test, integration of the phase distribution gives the correct surface topography. The procedure for an analysis method and experimental results are presented.

A CCD camera system has been added to a surface profiler to display Nomarski enhanced images of a surface under test. The viewing system provides an expanded view of the test surface for identifying the location and extent of surface features. Practical engineering considerations of the optical and illumination systems will be discussed. Also, examples of surface images and their respective profile scans will be shown.

D SIGHTTM is an effective method for inspecting surfaces for localized shape defects. A complete theory is presented which describes how this phenomenon converts surface slope variations to grey scale information. The theory, based on graphical ray-tracing, accounts for all significant physical parameters and illustrates their interactions, as well as their effect on the resulting D SIGHT image. The theory's validity is demonstrated using a computer-based, ray-tracing simulation. Using the physical parameters as inputs, the computer model is shown to produce images which closely resemble those obtained from corresponding experimental hardware.

The authors present ne applications for the recently developed nori-contact optical inicrotopographer emphasizing the results of topographic inspections of thin silver films edges. These films were produced by sputtering of silver through different masks, using a planar magnetron source. The results show the influence ot the thickness and position of the masks on the topography of the film near its edge. Topographic information is obtained from the horizontal shift incurred by the bright spot on an horizontal surface, which is displaced vertically, when this is illuminated by an oblique collimated laser beam. The laser beam is focused onto the surface into a diffraction limited spot and is made to sweep the surface to be examined.. The horizontal position of the bright spot is continuously imaged onto a light detector array and the information about individual detectors that are activated is used to compute the corresponding horizontal shift on the reference plane. Simple trignometric calculations are used to relate the horizontal shift to the distance between the surface and a reference plane at each sampling point and thus a map of the surface topography can be built.

Interferometer sensors using optical waveguides have been shown to be suitable for sensing a vanety of physical items such as temperature, strain, humidity, electric and magnetic fields, posItion. In this paper we study an integrated optical microdisplacement sensor making use of a Michelson interferometric configuration. The two-beam semi-asymmetric X junction is composed of four single-mode Ti diffused LiNbO3 channel waveguides at ? = 633 nm. The design criteria stress the fact that the waveguide runs at a given angle with respect to the principal reference system. The electromagnetic field evolution is obtained both by mode-matching and beam propagation methods.

Nanoscopy is a term that we use to describe optical techniques using digital image processing that are capable of nanometric observation and measurement. Laser Scanning Tomography (LST) is used for defect analysis in the bulk of semiconductor wafers for revealing particles as small as mm and for measuring densities of uptolO13 cm3 The unusually high contrast of the system allows us to observe submicron particles which are more than three orders of magnitude smaller than the Rayleigh criterion for the optical system. Recent work using deconvolution of point image functions enables us to perform sub-micron optical serial sectioning, for determining the depth of defects. The best conditions for classical LST (using laser illumination perpendicular to the viewing direction) are when operating further than a few microns below the surface in semiconductor wafers; ie it is ideal for bulk defect studies. The study of imperfections inside epilayers in the top ijim layer requires a modified technique. Instead of illuminating at 900 the viewing direction, the infra red laser beam is introduced obliquely to the front suffe to illuminate the defects in the epilayer while still in the dark-field mode. Combining this method with high resolution sectioning will be the basis for a technique of three dimensional submicron defect analysis in epilayers. Results are given of defect studies in annealed GaAs and silicon to demonstrate the capabilities of LST for naiiometer analysis in bulk materials. Sub-micron depth measurement is shown for single particles using the PSF of the system for the out-offocus case. Some initial results are given of studies of defects in a highly lattice mismatched epilayer using IR transmission microscopy, phase stepping microscopy (PSM, used in surface profiling) and the new dark field oblique laser illumination technique. LST, and the new nanoscopy techniques are non-destructive, operate under normal room conditions, and give sub-micron observation and distance measurement of defects over large areas with the possibility of 3D image synthesis for defect analysis.

This paper deals with an autoiatic and precision alignient technique for proxiiity printing in x-ray lithography, using two pairs of iioire gratings, with moire signals from each pair being 180 out of phase with each other. The automatic and precision alignment experimental system which was constructed can measure both transmission moire signals and reflection moire signals at the same time. The automatic alignment was achieved using diffracted moire signals in transmission and also in reflection as control signals for a stage driver. The alignient position of the system was monitored by the difference signal in non-control signals. The drift characteristics of the alignment position were measured by operating voltage gain and/or offset voltage value of preamplifiers in the system. We concluded that the technique using diffracted moire signals is a usable automatic and precision alignment technique and the technique could be applied to one of the variable positioning techniques.

Moire effect produced by the two gratings under coherent illumination attracts a high level of interest in the mask and wafer alignment for X-ray lithographic system. In such reflection optical arrangement, the displacement signal is often affected by the light reflected directly from the mask grating. On the other hand, in the photoacoustic technique, the signal is generated by the absorbed light energy and thus it is not influenced by the direct reflection. We studied the detection method of the moire displacement signal by using the photoacoustic technique. Using the transmission and reflection (absorption) gratings with 25jim pitch, the lateral displacement was detected. The probe beam deflection method and the microphone method were used for sensing the displacement from the acoustic vibration and the acoustic pressure, respectively. The signals changed periodically by the pitch of the grating for the lateral displacement. The characteristics of the displacement signal were examined theoretically by the calculations based on the Fresnel diffraction theory.

Recently, the interference phenomenon known as grating imaging has been studied for some practical applications as well as for the theoretical interest. When the object grating, which is illuminated incoherently, is located in the vicinity of the transmission grating, the magnified image of the object grating is obtained at the image distances far from the gratings. Since the high magnification (%.lO2 to 1O) can be obtained, the image displacement is very sensitive to that of the object. We propose here a displacement sensing technique using the grating images detected by CCD image sensor. The images are processed with a computer by using a spatial phase detection method. The lateral displacement and the distance separating the two gratings can be obtained simultaneously and separately. The proposed technique will be useful as a compact two dimentional displacement sensing method in opto-mechanical systems.

A noncontact method of monitoring structural displacement is demonstrated. The technique uses a lightweight diffraction grating with a variable period that is attached to the structure of interest. The position of the grating is monitored optically and linear displacement information can be separated from the effects of simultaneous rotary displacement. The technique is demonstrated for both static and dynamic measurements.

Interferometric metrology in which transverse or in-plane measurements are made is discussed. This technique can measure the displacements of diffusely scattering surfaces with microinch resolution and accuracies on the order of 0.01 percent and the displacements of relatively coarse scales with submicroinch precision and an accuracy determined by the scale itself at speeds over 100 ft/sec. The measurement principle and geometry are described and the measurement of the interferometric phase is addressed. The system design is presented.

This paper describes a new approach for mask-to-wafer alignment and gap control using modified moir technique. The alignment marks are in the form of gratings. A laser beam is passed through the mask grating and reflected from the wafer grating. A relative displacement, say in X-direction, between mask and wafer gives rise to periodic intensity variation called a moire signal. In the present method, the alignment is automatically achieved in a region of higher slope instead of at peak or trough of the moir signal where the slope is zero. This gives greater alignment accuracy. This was done earlier by using two pairs of l800 spatially phase shifted gratings. We have simplified this technique by using only one pair of grating to obtain the moir signal. The other moir signal is the electronic inversion of the former. Difference between the two moir signals is zero at a point in the higher slope region, which is considered as correct alignment position. Automatic alignment can also be achieved by using difference between the moir signal and a d.c. voltage equal to its average value. At the aligned position the mask grating is approximately quarter pitch shifted with the wafer grating. In a similar way an orthogonal grating pair gives automatic alignment in the "Y" direction. It has been observed that the gap characteristics for a quarter pitch shift are best suited for non-contact gap (Z-direction) measurement. This can be conveniently done by simply counting the number of cycles of the intensity variation with gap. The gap control is obtained between any two of the Fresnel image positions by using electronically inverted moir signal similar to that for X and Y alignment. The estimated accuracies for X and Y alignment are 32 nm and for Z control is 500 nm.

Fringe scanning for projection moire to be used in measuring thermal deformation of space panels has been realized using a servo-controlled grating. This system can measure deformation of a 1.6 m x 1.6 m white panel with a resolution of 0.3 mm if point-by-point calibration is available. The cyclic thermal deformation testing of a space panel was successfully carried out using this system.

We describe the interference fringes produced when periodic and aperiodic (random) amplitude masks are illuminated with partially coherent light. Periodic transmission gratings and the recordings of laser speckle patterns have been used to constitute the masks. We will discuss applications to measure the motion of objects and in particular study the rotational motion measurement using random masks.

The modulation of the spectrum of a light beam is consider as a metrological tool. In particular, double spectral modulation of a Super Luminiscent Laser Diode (SLD) is used to analyze surface's profiles. Intensity and frequency modulation allows absolute measurements of the surface without any auxiliary phase shifting. Depth and lateral resolution is determined by the spectral resolution of the involved spectroscopic devices.

A new Zeiss interferometer, the Direct 100, is described. This interferometer contains important improvements that are especially applicable to measurements in the presence of vibrations, measurements requiring the highest spatial resolution, integration of testing into manufacturing, use of interferometry for optimum assembly of complex systems, high-precision measurements in air with sub-nm accuracies, absolute calibration of residual errors, testing of aspheres with partial compensation and electronic hologram, measurement of fast processes, and temporally resolved measurements. The measurement and evaluation principle of the Direct 100 are described along with its optics, mechanics, and electronics.

The new Zeiss interferometer works with a complex software package which can be run in two different modes : a "workshop mode" for simple pushbutton operation of standard measurement sequences, and a "master mode" well suited for the use in the laboratory with changing requirements for measurement and evaluation methods. Macros of arbitrary functional sequences can be created in the master mode and associated with function keys for the workshop mode. A description of the implemented features concerning data acquisition and display, control of the instrument as well as data evaluation and manipulation is given.

Various aspects of interferogram analysis are reviewed. Fixed interferogram evaluation, fringe measurements, the global interpolation of interferograms, the application of Zernike polynomials, Fourier analysis of interferograms, and phase-shifting interferometry are described. Under the latter heading, the methods of zero crossing, phase lock, phase stepping, integrating phase shifting, and sub-Niquist and two-wavelength interferometry are considered.

We outline a fringe analysis algorithm that is completely automatic and has been demonstrated to give individual pixel resolution with good accuracy for very general types of fringe patterns, but is slow and quite sensitive to noise in the raw fringe image.

A new method, based on a combination of interferometric grating projection and 2D digital FFT-based phase measurement, is presented for 3D shape measurement and surface inspection. This technique can generate a grating pattern focusing everywhere in its illumination space; thus, it has infinity focus depth. This capability makes it possible to improve measurement sensitivity by increasing the projection angle without losing the grating focus, making it more suitable for surface inspection of a large object. The sinusoidal intensity distribution of the grating, the monotonic phase nature of the fringe pattern coded on the object surface, and the phase information spectra clustered in the frequency domain make data deduction by this technique not only possible, but more accurate.

The theory and practice of phase-measuring interferometry (PMI) are reviewed. The phase measurement algorithms are discussed and phase modulation and wavefront reconstruction are examined as part of the implementation of PMI. The performance limitations of PMI are addressed.

Moire is a common technique for contouring large scale features on diffuse surfaces. Typically, this is accomplished by projecting a high-frequency line grating onto the object and viewing its image through another high-frequency line grating. The resultant pattern, produced by aliasing between the two gratings, shows lines of surface contour. The surface height between these contour lines is determined by triangulation between the illumination and observation beams. Although moire is really a geometric ray technique, it is typically referred to as interferometry because its contour patterns resemble interferograms; Moire contouring can be classified as either shadow or projection moire. Shadow moire uses the same grating for both illumination and observation, while projection moire uses separated gratings. Another surface contouring technique is projected fringe. This approach is similar to moire in that it projects a line grating and uses triangulation to measure surface height, however, there is no second grating or aliasing. Rather, the projected pattern is.viewed directly. Although it is not really moire (because it does not use aliasing with a second grating), projected fringe is typically considered a moire technique. This paper has four parts. First, it reviews the theory behind moire contouring. Second, it outlines how to phase modulate a moire 'interferogram'. Third, it discusses practical considerations associated with performing moire contouring such as projection and observation geometry, imaging, and illumination. And fourth, it presents two examples of objects that were contoured using a white-light moire phase-measuring interferometer.

A novel two-dimensional Fast Fourier Transformation (FFT) based phase measurement technique for fringe pattern analysis is presented in this paper. The technique uses two fringe patterns whose phase are slightly shifted to eliminate any phase ambiguity which exists in a fringe pattern whose phase is not monotonically increasing. It is capable of extracting an accurate phase distribution from a noisy fringe pattern without assigning fringe orders and interpreting data in the regions between fringes. The technique is free from error caused by the high frequency components, the variation of background intensity, and the modulation of amplitude on fringe patterns. This technique is particularly useful for processing the high-frequency noise corrupted sinusoidal fringe patterns.

A new technique for phase extraction of a fringe pattern is presented and verified in this paper. The technique makes use of the orthogonality of the trigonometric functions. It acquires the wrapped phase of a fringe pattern at a pixel point with the use of intensity values at that pixel point and its neighbors. It is capable of obtaining the unwrapped phase of a fringe pattern at every pixel point automatically provided that the phase of the fringe pattern is monotonically increasing. In contrast to the Fast Fourier transform (FFT) based phase measurement technique, this technique not only eliminates the difficulty of choosing a passing-band in the frequency domain but also provides great flexibility for processing an arbitrary number of pixel points at every row of the image.

Techniques for the quantitative determination of shock position in supersonic flows using direct and indirect methods is presented. A description of an experimental setup is also presented, different configurations of shock position sensing systems are explained, and some experimental results are given. All of the methods discussed are analyzed to determine the ease of technology transfer from the laboratory to in-flight operation.

It is shown that the power spectrum of a specimen surface in the form of a laser speckle pattern can be characterized by the fractal dimension which reflects change in surface texture due to plastic strain. The fractal dimension method can be used to evaluate surface roughness and plastic strain. A number of measure-scale pairs are tested, and the intensity-frequency pair is found to be best at relating fractal strain to plastic strain or surface roughness.

A novel optical method of measuring 2-D surface strain is proposed. Two linear strains along orthogonal axes and the shear strain between those axes is determined by a variation of Yamaguchi's laser-speckle strain gage technique. It offers the advantages of shorter data acquisition times, less stringent alignment requirements, and reduced decorrelation effects when compared to a previously implemented optical strain rosette technique. The method automatically cancels the translational and rotational components of rigid body motion while simplifying the optical system and improving the speed of response.

A design for a birefringence transducer for high-temperature applications is described. The spring element and the readout instrumentation are addressed. A pressure transducer based on the concept has been built and successfully tested at temperatures up to 600 C.

Speckle correlation, in real time, is used to detect the growth of the oxfdation layer in copper at room temperature.

Nuclear contaminated waste must be isolated for up to 10 , 000 years before it is rendered harmless and, at present, bedded salt is the burial in&lium of choice. Bd&1 salt is not, however, considereI to be an erineerir material ; its lorj-term creep properties are recondite. The prcble facir analysts is that new defomation models require experimental verification for which traditional point-by-point methods , such as strain gages , provide inadequate coverage. Fortunately, wholefield optical tecbniques, speckle photography in particular, are amenable to this measurement prcblem. The use of 'speckle' in this novel application

Yeast and bacteria growth has been detected by observing light diffracted from specially prepared diffracting screens upon which these organisms were grown. The screens were fabricated by impressing patterns on the surface of a growing medium. As the microbes metabolized, they caused severe changes in the lightdiffracting properties of the patterns as they ate into them. Such changes were detected within twenty minutes for bacteria and ten minutes for yeast.

A brief review of the application of acoustic speckle as an information carrier for uses is surveyed first. A two-dimension acoustic speckle pattern (ASP) as a whole is shown for the first time by liquid surface relief method, and the basic principle of taking ASP due to a random gridded-like "acoustic diffuser" is presented. Also, the use of ASP for measuring displacement of underwater object is given, and the physical linking of ASP and the sun glitter pattern on random sea waves is described as a possible new application of ASP.is described.

Angle-resolved light scattering (ARLS) was used to investigate the roughness of coatings on glossy paper. Angular spectra were measured for laser light scattered from several glossy paper samples and from uncoated paper. These spectra are compared to those calculated using the Beckmann model of a random surface that is isotropic and rough in two dimensions. Such a surface is characterized by its rms roughness and autocorrelation function, which are determined from surface profiles measured with a stylus instrument. There is very good agreement between the measured and the computed ARLS spectra. The surfaces are too rough to produce a specular beam large enough to provide an accurate value of the rms roughness, but ARLS provides information about the coating roughness when the measured spectra are cornpared to computed ones.

The intensity correlation function of' the light scattered by the diffuse object under fully developed speckle pattern illumination has been theoretically and experimentally investigated. The scattered light is called a doubly scattered light. When the patially coherent light illuminates the diffuse object over enough wider region than the coherence area, the doubly scattered light obeys Gaussian statistics. Then the surface roughness o the diffuse object does not depend on the speckle contrast of the doubly scattered light, but depends on the speckle size (decay rate of the intensity correlation function). From these properties, two roughness parameters of the object have been determined.

Reported here are a method and some experimental results for measuring the scattering of optical coatings in high power lasers. This method is practical and nondestructive and is to be used in absolute measurements.

A new technique to obtain the refractive index and thickness of a thin film simultaneously is presented. The reflectances of p polarized light and s polarized light are measured at various angles of incidence, and by a numerical procedure, the film index and thickness are extracted from the measured reflectances. The measurement and numerical procedure to determine the film index and thickness are simple and the obtained values are accurate. As an example, we made measurements on a single layer film (SiOJSi) and a double layer film (SiON/Si02/Si), and confirmed the obtained values were consistent.

Techniques for the quantitative determination of shock position in supersonic flows using direct and indirect methods is presented. A description of an experimental setup is also presented, different configurations of shock position sensing systems are explained, and some experimental results are given. All of the methods discussed are analyzed to determine the ease of technology transfer from the laboratory to in-flight operation.

A new way of precise measurement of moving object is described in this paper. Using Laser-Fourier Transformation-CCD System and computer (for real-time processing) We realized the dynamic measurement of tungsten filament etc.under the real-time si- tuation, and measuring results have been achieved of linear-diameter range 1O-2OOm, the precision 0.5% and the velocity O-l5OmImin. Though the static precision measurement of fine wire diameter has been well re- solved in scientific research and industrial applications, the precision measuremeit of a moving fine wire is still a rather difficult problem. There are two kinds of dynamic measurement of fine wire now: One is the light spot scanning method, with a parallel beam scanning across a wire and a phototube receiving the light singnal op- posite, the diameter of wire from 5Opm to several millimeters can be measured; the other is to measure the shadow of fine wire with a CCD ( charge coupled device ), the measuring range is about 0.5-30mm. These two methods can measure fine wire dy- namically, but their accuracies are 1,ow and measuring ranges are limited. They are not suited to measure thinner wires for too big error. It is often necessary to measure dynamically wires thinner than 50 Itm with high precision in industry practice, such as tungsten filament using in bulb factories, which diameters are from 8 to 40 im. The above methods aren't suitable. Now in some factories the weight method is used to measure fine wire diameters but it can only obtain the mean value but not the real value of diameters, so it affects seriously the quality of bulbs. The new method of dynamic precision measurement we studied is to measure the power spectrum of a fine wire with a CCD, and process data with a micro-computer at real time. Satisfacory results have been achieved.

This report describes a simple automated system for measuring interfacial tension using the pendant or sessile drop method. The size and shape of a transparent or opaque drop of one fluid immersed in a second, transparent, fluid is recorded with a CCD video camera and digitized and stored by a computer-controlled system. Custom software determines various droplet shape factors and computes the interfacial tension. A limited number of video frames can be stored on disc, or longer runs can be stored on video tape for later digitization. Alternately, only the shape factor and interfacial tension data are stored to reduce demands on the storage medium. The first application of the system was measurement of the interfacial tension of crude oil interacting with various bacterial agents in aqueous suspension. Some of these agents can greatly influence the effective interfacial tension of the crude oil and potentially improve recovery rates from oil reserves, particularly of the "heavier" or more viscous oils.

The characteristics of an already reported E0 sensor have been taken into consideration in order to extend its application to various metro- logical and industrial set-ups. The inclined white light fleck mci- dence and differential photo-detection based optical triangulation sensor is highly sensitive to all types of polished and rough surface targets. The effect of the various design parameters on the perfor- mance of the E0 sensor has been studied theoretically in the present paper.

This paper presents a simple and low-cost range finding system based on the absorption of light by a dispersive medium. The object to be digitized is immersed in a colored liquid and the 3-D data is obtained by processing a pair of greylevel images taken at different wavelengths. Calibration, acquisition, visualization and manipulation of the 3-D data are implemented through menus and graphic windows on a SUN-3 workstation.

A concept - holotag, of an automatic holographic-aided identification label is presented with comparative advantages. Technological challenges met on the concept experimental implementation are reviewed. The label carried information is first transposed with a 2-D interleaved self-correcting spatial coding into a plane binary object patch pattern. Then recorded in the form of a computer generated Fourier hologram, spatially multiplexed- holotag, onto an optical recording material-Digital Paper, laminated and protection covered. Image processing techniques followed by decoding procedures regenerate the original data despite process incurred degradations.

A new method of isoclinic measurement is presented. The isoclinics in photoe- lastic stress analysis can be doubled or quedrupled by the phase-shifting and the digital image processing techniques. The centers of the isoclinics are easy to be drawn out.

The use of the phase shifting interferometric technique is discussed to make quantitative surface profiling using the Nomarski differential interference microscope. Lateral shift of the Nomarski prism introduces mutual phase shift between interfering two wavefronts with small amount of shear. Since the analyzed phase distribution corresponds to the differential of the surface profile under test, integration of the phase distribution gives the correct surface topography. The procedure for an analysis method and experimental results are presented.

We describe an optical 3-0 sensing system by PMP for 3-D shape measurement of bottoihole pattern.The principle,system construction and experiaental results are described briefly.

Yeast and bacteria growth has been detected by observing light diffracted from specially prepared diffracting screens upon which these organisms were grown. The screens were fabricated by impressing patterns on the surface of a growing medium. As the microbes metabolized, they caused severe changes in the lightdiffracting properties of the patterns as they ate into them. Such changes were detected within twenty minutes for bacteria and ten minutes for yeast.