Over the years SPIE has organized more than 3000 conferences and with few exceptions each conference produced a Proceedings volume. When combined with the articles published in this journal and its sisters, SPIE publications represent a major contribution to the optical engineering literature.

The results of detailed predictions are presented for the wavelength dependence of the blur circle of the proposed Stratospheric Observatory for Infrared Astronomy (SOFIA), which will have a 2.5 m diameter telescope in an open cavity aft of the wing on a Boeing 747SP aircraft. The analysis uses wind tunnel measurements of the turbulent shear layer (Rose Engineering and Research, 1996), computer calculations of the telescope cavity shear layer organized air density fluctuations (Atwood, 1993) and extrapolation of the cavity optical effects of the Kuiper Airborne Observatory (KAO) (Sutton, 1992) based on the expected SOFIA cavity noise (Rose Engineering and Research, 1996). The results indicate a visible wavelength blur circle diameter of many arc seconds, which then raised the issue of the use of the visible spectrum for natural guide stars to obtain the long term pointing accuracy of 0.14 arcsec. Detailed calculations of SNR indicate there are a sufficient number of stars having adequate visible brightness to achieve this objective. Optical tests in a quiet transonic wind tunnel and independent calculations of the air density fluctuations in the telescope cavity would confirm this result.

The analysis of an IR tunable filter that uses a transmission grating with variable spacing is extended to include incident spherical waves instead of plane waves and to illustrate specifically the effects of finite conductivity in the metal. The model is based on a rigorous vector solution of Maxwell’s equations implemented in a finite elements solver. The modeling results based on tabulated optical properties of nickel are compared with experimental measurements made with filters fabricated in permalloy by the lithographie, galvanoformung, abformung (LIGA) process. The theoretical and experimental results compare quite well and suggest that the resolution of the filter is severely reduced at the shorter wavelengths (less than 4 ?m). At longer wavelengths, the resolution is limited by the aspect ratio of the LIGA structures. A spectral resolution of 0.1 ?m appears to be practical.

A nematic liquid crystal (LC) grating is prepared using an amplitude-type grating electrode structure and a homogeneous initial alignment parallel to the grating electrode direction. The polarization properties of the LC grating are investigated by measuring Stokes parameters of diffracted light under differently polarized incident light. The electrically variable polarization modulations of diffracted light are verified for the first time, as well as the diffraction light intensity modulations. Moreover, the LC grating exhibits some unique optical properties, such as symmetric diffraction light intensities but antisymmetric polarization states in corresponding positive and negative diffraction orders when the incident light is linearly polarized parallel or perpendicular to the grating direction. A polarization grating model is proposed to clarify the unusual polarization properties, which suggests that it may be related to the antisymmetric LC molecular orientation domains induced by the grating electrode structure.

Carrier removal is a key step in the Fourier transform method (FTM) of fringe pattern analysis because it can give rise to significant errors in the recovered phase. The existing methods of fringe carrier removal in the FTM are reviewed and a comparison of three different methods, translation of the sidelobe to the frequency origin, leastsquares fit and subtraction of the phase of the undeformed carrier fringes, are presented. Computer-generated fringe patterns are used to determine the differences between the original and the retrieved phase distributions. Several figures of merit are proposed to assess the performance of the mentioned methods. Experimental fringe patterns, obtained by double-pulsed-subtraction TV holography, are analyzed by the methods considered here, and the retrieved phase distributions are also compared.

The effects of a thin lens on a real laser beam with an M2 factor larger than 1 is theoretically studied. The effect of weak lens aperture truncation is considered by introducing an increment to the M2 factor. A new thin lens equation is derived. This equation is a simple and effective tool applicable to real laser beams, and can be used to estimate the phenomena of focal shift and change in focused spot size caused by weak lens aperture truncation. Two examples of applications are described.

A simplified setup for an automatic film analysis system is described. It is composed of a digital image processing system and a film transmission mechanism. The film, which records the movement of an impact or explosion is driven by a film transmission mechanism and is converted to digital images. The signaled points with variable shape are matched by correlative matching in sequential images and data representing the movement is obtained. In this system, some practical methods such as subpixel location and distortion elimination for the camera lens are implemented to improve the measurement precision.

A real-time visual approach for the calibration of phase step in image plane holography using BaTiO3 crystal as a recording medium is reported. In this method, the diffuse object is illuminated symmetrically with two collimated beams and a pump beam is added at the crystal plane. Each illuminating beam generates an individual interferogram at the observation plane when the object is subjected to load. A ? phase shift to one of the illuminating beams results in the complete disappearance of the fringes due to phase reversal between the individual interferograms, and at 2? phase change, the fringe pattern is restored. After calibration, a single-beam illumination configuration is used for phase measurement. We present the theory of the method and experimental results for a centrally loaded diaphragm rigidly clamped along the periphery.

An iterative approach that partitions an image into piecewiseconstant regions is presented. Each iteration consists of three steps. The first step extracts edges from the image. The extracted edges, which must exhibit high connectivity, are computed using a Laplacian-like morphological edge detector in the first iteration and a simple gradient thresholding in subsequent iterations. After the first iteration, the edge extractor operates on a piecewise-constant image for which the edge detection problem is well defined and well posed. In the second step, a fast averaging of connected pixels within closed boundaries defines the regions in the image. Finally, edge pixels (both true and spurious edge pixels) are each assigned to an underlying region. The iterative application of the three steps simplifies an input image in a diffusion-like manner. The advantages of this algorithm are that both local and regionbased information is incorporated, convergence occurs in a few iterations, and the component operations are relatively simple.

A fully automatic method for pilling evaluation in wear-andtear fabrics is developed from the image analysis of a set of standard photographs (Zweigle KG-741 Reutlingen). The method involves operations in both the spatial and frequency domains to segment pills from the textured background of the web. It calculates the total area of pilling in the sample image and assigns a degree of pilling according to the standard. Two mathematical descriptions are analyzed according to the underlying rule established by the standard images, using the visual estimation of the area of pilling performed by a group of observers. A logarithmic (in base two) approach, which is consistent with human visual perception laws and facilitates an optimization of the method, is eventually adopted.

The extraction of defective segments from fruit images is a critical step for on-line fruit defect detection using machine vision. To accommodate the gradient reflectance characteristics of objects with curved surfaces such as apples and to avoid the defect inspection error due to the natural brightness variability of fruit, a novel and practical brightness-invariant image segmentation method is developed. Experimental results from our vision-sorting system show that the proposed method is effective for large-scale multiple-line processing of massive numbers of apples.

The validity of a multichannel deconvolution technique where both within and between channel relations are used was demonstrated using stochastic and deterministic deconvolution filters. In either case, prior knowledge about the original image and the noise is required. A generalized regularized multichannel image deconvolution approach is proposed in which no prior knowledge of the variance of the noise at each channel or a bound on the high-frequency energy of the image are assumed. This information instead is estimated iteratively, resulting also in the estimation of the regularization parameters, based on the partially deconvolved image at each iteration step. The multichannel smoothing functional to be minimized is formulated to have a global minimizer with the proper choice of the multichannel regularization functionals. With this algorithm, the regularization functional for each channel is determined by incorporating not only within-channel information but also cross-channel information. The proposed multichannel smoothing functional is shown to be convex, and therefore, has a global minimizer. The algorithm not only does not depend on initial conditions but is also shown to be much more computationally efficient than existing algorithms. The validity and efficiency of the approach is demonstrated by applying it to color image restoration, the wavelet-based image deconvolution, and the reconstruction of the higher resolution image.

An algebraic method based on finding least-squares solutions from a set of simultaneous linear equations (SLEs) is applied to phase unwrapping by regions in optical interferometry. By transforming the phase unwrapping problem into a matrix equation, the characteristics of the phase map can be observed entirely from the properties, such as the rank or the null space, of the matrix that represents the phase unwrapping problem. The familiarity of solving SLEs using common numerical routines such as QR decomposition means that the algebraic method is easily formulated and computed. The robustness of the phaseunwrapping method is then tested computationally and experimentally and its enhancement is proven through the simulation and experimental results.

A novel setup for Lau fringe formation is proposed that employs a grating registered as birefringence modulation in a photorefractive crystal. The system behaves as a combination of an amplitude and a birefringence phase grating, which enables us to control the visibility and provide a contrast reversion of the Lau fringes by rotating an analyzer. A theoretical approach is presented and the predicted behavior is confirmed by results obtained under different experimental conditions.

A speckle correlation technique based on chirp modulation of the speckle interference pattern is proposed and demonstrated for the measurement of 3-D rigid body motion. The chirp modulation is a consequence of the out-of-plane displacement component associated with object motion. The Fresnel zone plate action introduced by chirp modulation causes the correlation peaks to focus at different planes along the longitudinal direction. The separation between the correlation peaks along the transverse direction gives a measure of in-plane displacement of object motion. Both simulation and experimental studies are performed to establish the proposed technique.

A speckle correlation technique is proposed for the measurement of small object tilt. Theoretical analysis is formulated for a general case of object tilt. Simulation and experimental studies are performed to verify the outcome of theoretical predictions and accuracy in measurement. Sensitivity of tilt measurement is studied and quantified from observations made from object position in various longitudinal planes.

A novel optical method for high-resolution angle measurement based on surface plasmon resonance (SPR) is presented. The phase difference between p and s polarization states when SPR takes place is very sensitive to the angle of incidence. Heterodyne interferometry is employed to measure this phase difference and thus to measure small angles. The feasibility of this method is demonstrated. The measurement resolution of 2×10-5 deg can be achieved by this method.

We describe a new method for microlens array generation based on the application of the imaging properties of the sampling filter. A binary amplitude mask corresponding to the transmittance of a lens is used as an input object. Its multiplied images, produced by the sampling filter with properly selected resolution, are characterized by the continuous distribution of intensity, which, when exposed in a phasephotosensitive material lead to the appearance of the microlens array. The specific, collective character of the imaging process results in great lens homogeneity. The technique offers wide design freedom and enables us to obtain lenses of long focal lengths. Numerical verification of the concept as well as experimental results are presented.

Moire´ magnification can be observed visually if an array of identical objects is viewed through an array of identical microlenses with a different period. Theoretical analysis and experimental results of the moire´ image obtained by a moire´ magnifier are presented. Conditions for erect and inverted moire´ magnifications are derived and interpreted. Virtual erect images are observed only when the period of the lens array is larger than that of the object array, while inverted images are obtainable in both cases. For equal periods, a uniform field of view results. The relation between the relative size of the periods and the distance between the object and the lens arrays are derived. Expressions for image magnification, orientation and size are deduced. The condition to obtain a demagnified moire´ pattern is deduced. Rotation of the lens array with respect to the object array results in rotation of the erect and inverted moire´ patterns in similar and opposite directions, respectively.

The probabilistic neural network (PNN) is based on the estimation of the probability density functions. The estimation of these density functions uses smoothing parameters that represent the width of the activation functions. A two-step numerical procedure is developed for the optimization of the smoothing parameters of the PNN: a rough optimization by the conjugate gradient method and a fine optimization by the approximate Newton method. The thrust is to compare the classification performances of the improved PNN and the standard back-propagation neural network (BPNN). Comparisons are performed on a food quality problem: french fry classification into three different color classes (light, normal, and dark). The optimized PNN correctly classifies 96.19% of the test data, whereas the BPNN classifies only 93.27% of the same data. Moreover, the PNN is more stable than the BPNN with regard to the random initialization. The optimized PNN requires 1464 s for training compared to only 71 s required by the BPNN.

A new method is proposed for the compensation of color distortion between the right and left sides of a projected image in an LCD projection system. The deviation of the incident beam angle causes the difference in spectral transmittance between the sides of a dichroic mirror, and the color distortion resulting from this difference is analyzed by the colorimetric method. In the compensation method, the variation of the transformation matrix between the Commission Internationale de I’Eclairge (CIE) XYZ tristimulus values and RGB primary values is linearly approximated under reasonable assumptions based on typical LCD projection systems. The proposed method can be simply implemented with an on-line scheme to be practically applicable to projection TV systems and its effectiveness is shown via experimental image examples.

A new design method using a color appearance model is proposed for the dichroic filters in LCD projection systems. The dichroic filters used for color separation/composition play a dominant role in the performance of color reproduction so that its spectral transmittance should be designed to have optimized color performance. In the proposed method, a reproducible color gamut in the 3-D color space is used as a performance index, and we find the optimal half-power wavelengths of dichroic filters, which are applicable to the development of LCD projection TV systems. Considering diverse viewing conditions with moderate ambient light, the design parameters are optimized by maximizing the volume of the color gamut in an apparent color space for each condition. Here, the RLAB color space developed by Fairchild is used as a color appearance model and the white balancing method is applied to restore the designated color temperature of peak white. The optimal wavelengths are compared with the design based on the CIELAB color space without considering nonstandard viewing conditions, and both results do not show as much difference as expected due to the white balance.

A system for shape recognition of simple objects (a cube, a plate, a rod) falling in the field of vision by extracting their image contours is proposed. Recognition is carried out from three projections. The minimal time of image processing is 1 ms. The idea of the simultaneous extraction of an image contour and the parameters characterizing its shape is proposed based on the properties of human vision. The sensor is an optoelectronic functional converter representing a system for searching and extracting an image contour by the radius-vector scan with a dissector as a photodetector. It realizes the fastest method of object image analysis to recognize image shape because of the linear functional connection between an electronic parameter (a brightness level as an argument) and a geometric parameter (for example, a radius as a function) of an image in the process of following its contour with a photodetector aperture. As a result, the instantaneous video signal amplitude (which is proportional to the brightness level at the point of the aperture location) corresponds to the current length of the image radius, and its variation during the tracing of the contour adequately characterizes the image shape, including such features as the extreme dimensions and the angles. Simultaneously the coordinates of the contour points of the image and its center of gravity are extracted. The highly informative nature of the video signal also enables us to determine the perimeter, the area, the orientation of the image, and the coordinates of its brightness extremum. The radius-vector scan is the only sensor currently available whose signals carry direct information concerning the image shape in real time.

Recently we published theoretical expressions for the eigenvalues and eigenvectors for a twisted-nematic liquid crystal spatial light modulator (LCSLM) as a function of the twist angle and birefringence using the Jones matrix formalism. These polarization eigenvectors are of interest for newer video graphics adapter (VGA) resolution (640 3480 pixels) LCSLMs that are thinner and consequently have a lower birefringence. Although these devices can produce excellent amplitude modulation, they cannot achieve phase-only modulation using linearly polarized light. In these cases, the polarization eigenvectors are required to obtain phase-only modulation. We present experimental measurements for both the transmission and phase shift for these polarization eigenvectors. These results are in excellent agreement with theory and show that phase-only modulation can be obtained for these highresolution LCSLMs.

In a continuous presence multipoint control unit, the multiple video sequences of the high spatial resolution are decoded and downconversed, then composed into an output video sequence. When the bandwidth to transmit the composed sequence is fixed, if the bit rate for each low spatial resolution sequence within the composed sequence is independently allocated, then the degradation of subjective picture quality among the low spatial resolution sequences is severe due to the different video characteristics. An adaptive bit-rate control method to maintain the uniform picture quality among the low spatial resolution sequences is presented. We first find a relationship of bit rate and quantizer parameter for the video manipulation and composition environments. Then we obtain the bit rate for each low spatial resolution sequence to have same quantizer parameter among the sequences by adaptively using the coded results of low or high spatial resolution sequences while considering the scene change of each sequence. We show by simulation that the proposed adaptive bit-rate control method can maintain almost uniform picture quality among the low spatial resolution sequences and almost constant picture quality within the composed sequence in comparison to an independent bit-rate control method.

Recently, several photonic delay lines (PDLs) based on various optoelectronic technologies have been proposed. PDL applications include phased array antennas1 laser radars2 astronomy3 ultrasound4 and two-photon memories5.