This PDF file contains the editorial “Could Photonics Kill Optics, Part II?” for OE Vol. 39 Issue 07

This PDF file contains the editorial “Guest Editorial: Special Section Pushing the Envelope in Optical Design Software” for OE Vol. 39 Issue 07

Misalignment modes are combinations of rigid-body perturbations to the optical elements that make up an optical system. Comparison of misalignment modes associated with arbitrary metrology data and optical-system performance can be used to increase the sensitivity of those metrology measurements to specific optical-system performance specifications such as centroid distortion. Misalignment modes are compared using a mean squared error (MSE) metric. The MSE value can serve as a figure of merit by which metrology measurement parameters are optimized. Selection of measurable misalignment modes in the case of metrology and interesting misalignment modes in the case of optical- system performance is based on a determination of whether a mode can fit into a projection camera, given actuator-stroke and mirror-tilt bounds. As an example of this type of analysis, we find that in the case of an extreme ultraviolet lithography (EUVL) projection camera, exit-pupil wavefront measurements can be made more sensitive to centroid distortion if these measurements are collected at optimized field positions both inside and outside the ring field of view of such a camera.

Lens design software has been designed primarily to model conventional imaging systems. While interferometers do not generally fall into this category, lens design software nonetheless is well suited for analyzing a variety of aspects of interferometric systems. The general requirements for a ray-based model of interferometers are discussed, and a variety of examples are presented. The examples are designed to demonstrate both the power and flexibility of the approach proposed here for modeling interferometers.

A 60-deg-field-of-view optical see-through head-mounted display (HMD) using off-axis optics has been designed for 3-D medical imaging visualization. Two basic on-axis optical design concepts for seethough HMDs are reviewed first, to motivate the design of an off-axis optical form. An off-axis design is then presented. Because HMDs are typically designed from the pupil of the eye to the miniature display, it is common to assess final performance according to the display characteristics. Such analysis, however, does not provide information that is easily translated into task-based performance metric. Therefore, we present an analysis of the performance of the design from a usability viewpoint. For this analysis, the optical system is ray-traced from the display to the eye. Three key measures of performance—accommodation, astigmatism, and chromatic blur—are presented over the field of regard using customized graphical output.

A new optical design procedure for designing an optical device that will duplicate the wavefront errors of another optical instrument over the full field of view is described in this paper. We call this optical design procedure the inside-out method. It was devised to design an optical instrument that generated wavefronts simulating those of the Hubble Optical Telescope assembly (OTA). As is now common knowledge, during fabrication the OTA primary mirror was made using a defective null-test device; therefore the oTa wavefronts showed perceptible spherical aberration. With the inside-out method it was possible to design an instrument called Red-2 Stimulus that matched the OTA aberrated wavefronts to less than 0.0005? (638 nm) peak to valley on axis and 0.006 X maximum over the full field of view. The inside-out method is described, and two illustrative examples are given.

Tilted-component systems are known to be characterized by aberrations with unusual field dependences, such as decentered coma and binodal astigmatism. In this paper, the origin of binodal astigmatism in a multielement system from the contributions of individual surfaces is explained in an intuitive manner, as a logical extension of the ordinary aberrations known to all optical designers. The insight provided by this graphical model allows an understanding of why the astigmatism of any given system behaves the way it does, and what needs to be done to convert it to ordinary quadratic behavior, so that what remains can be corrected by a final, rotationally symmetric subsystem. Conditions under which ordinary behavior may be obtained from a system of tilted elements are explored. It is shown that a certain type of symmetry of the primary aberrations may be obtained (regardless of the tilt angles) if the initial, untilted system is monocentric. Graphical displays of the aberration types are used to explore the behavior of the system across the field, and direct optimization of the Zernike coefficients is used to guide the performance of the system.

Abstract. The authors describe a simple technique for the generation of well-corrected optical designs with both far-field and near-field requirements. Several design examples illustrate the use of this technique and its range of applicability.

This paper presents an approach that uses implicitly defined optical surfaces for designing imaging and illumination systems. In the standard aspheric optical surface, consisting of a conic with an even- order polynomial, the surface sagitta (sag), z, is defined explicitly as a function of the coordinates x and y. This standard aspheric surface is deemed useful for describing surfaces with small departures from a conic surface. However, optical surfaces with large departures from a conic are sometimes useful for current applications, such as null certification, conformal domes and windows, luminaires, and condenser design. The approach in this paper, which uses implicitly defined surfaces to describe highly aspheric surfaces, can be more general and easier to use for some situations. The sag z of an implicit surface is not defined directly, as it is in an explicit surface. Instead, it is defined indirectly in a more general form, as a function of x, y, and z. Because implicit functions have a more general form than explicit functions, they can better describe a variety of surfaces that cannot be easily described using the standard explicit aspheric surface. We show some examples of current interest.

This paper discusses the use of ZEMAX's Image Analysis feature to verify and predict the performance of Digital Light Processingâ„¢ (DLP) projection-lens designs. The main goals are to visualize the effects of lateral color, axial color, and the remaining Seidel aberrations on the focus of small DLP pixels on the screen in actual use. In many cases there is a discrepancy between what the projector user would define as "good pixel focus'' or "sharp focus'' and metrics that would determine focus quality or resolution for the lens designer. ZEMAX Image Analysis is a valuable tool for lens design and visualizing lens performance before a prototype lens is built. In addition, the user-defined offset surface is discussed, which was developed to simulate the separate focus of red, green, and blue DMDsâ„¢ in three-chip displays. This feature is used to simulate the effect of interchangeable lenses, as each lens has a different axial color characteristic and the depth of field is small, or to verify the compatibility of a lens design on a projector prefocused using a different lens. As display pixel sizes shrink, this simulation technique becomes more useful for evaluating projection lens designs, manufacturing tolerances, and ergonomic concerns during assembly.

A new and simple method to simulate a multielement source or astigmatic source in standard optical design software is presented. The technique can be used to define a laser diode, an array of laser diodes, a screen pixel, or other kind of multielement or astigmatic source. This paper shows how the standard classical approach using an astigmatic point source distance model fails when simulating a laser diode. A new approach is proposed that defines the source as an object in a single plane. This allows the use of powerful optimization in the optical design software without restriction. Some examples are presented.

An effective method of assigning tolerances to a null lens is to determine the effects of null-lens fabrication and alignment errors on the end-use system itself, not simply the null lens. This paper describes a method to assign null-lens tolerances based on their effect on any performance parameter of the end-use system.

Changes in glass catalogs from the major manufacturers, Schott, Ohara, Hoya, Corning, and Summita, are a future certainty. The ongoing efforts of these companies to eliminate arsenic, lead, and other environmentally unfriendly materials may well have an additional effect on the size of their catalogs also. We should not assume a zero-sum game, however. Environmental concerns may not lead to permanently smaller catalogs, though many have speculated that in the near term this might be so. However, from the designer’s perspective, very small, abbreviated class catalogs, constructed for special purposes, can speed the glass selection process. Several examples will be discussed, based on derivative libraries suggested by Zhang, Shannon, and Walker. Streamlined libraries tailored for special purposes can be used effectively in the latest lens design software. Future software tools may speed this selection process by the use of algorithms that treat the problem as a ‘‘black box,’’ using logic tools derived from probability studies of the patent literature.

Windows and domes that are shaped to aerodynamic requirements can increase range and speed for the host platform. This class of optical systems is referred to as conformal optics. The solution discussed here is intended for conformal missile systems having gimbals that point the optical line of sight through different parts of the dome. A conformal dome induces large amounts of varying aberration, tens to hundreds of waves across gimbal angle, and therefore requires dynamic correction. Space is very constricted in missile sensors, and it is therefore highly desirable to limit the number of motors used for aberration correction. This paper describes the performance of a new class of optical systems that employ counterrotating phase prisms to correct conformal dome aberrations while gimbaling the optical system. The phase surfaces on the prisms are described by Zernike circular polynomials. Since the shear across the phase surfaces is rotational, the only aberrations that are generated are those without rotational symmetry, such as tilt, coma, or astigmatism. Using this approach, CODE V® was used to analyze and design a compact, high-performance conformal optical system.

Faceted reflectors are a ubiquitous means for providing uniform illumination in many commercial lighting products, examples being newer flashlights, department-store display lighting, and the faceted reflectors found in overhead projectors. However, the design of faceted reflectors using software has often been more limited by the tools available to design them than by the imagination of the designers. One of the keys to enabling a broader range of design options has been to allow more complex surfaces using constructive solid geometry (CSG). CSG uses Boolean operations on basic geometric primitives to define shapes to create individual facets. In this paper, we describe an improved faceted reflector design algorithm and use it to create a wide range of CSG- based reflectors. The performance of various reflectors is compared using a Monte Carlo ray-trace method.

A simple and inexpensive method is described for generating special null Ronchi patterns. These patterns null out the effects of spherical aberration in lenses in which this aberration is purposely left uncorrected. The artwork for the special test pattern is created using the standard spot-diagram feature available in most computer ray-trace programs. The program does not have to be modified in any way to generate the test pattern. A model of a straight Ronchi grating is set up in the computer using a series of decentered rectangular obscuration bars. A ray trace is made from an on-axis point source through the lens being tested and continuing through the focal plane of the lens to the model Ronchi grating. The key to generating the null Ronchi mask is to then trace backwards to the desired position of the mask. A dense spot diagram at this location becomes the properly distorted null Ronchi grating, which can be scaled down as needed and copied as a (positive) transparency. In the lens test a straight shadow-bar pattern will appear on a screen placed at the location of the model Ronchi grating in the ray trace if the lens matches the design. The test described in this paper is intended for a lens surface accuracy regime of >20?, where an interferometer is usually too sensitive.

The exercise given demonstrates the construction of a diffraction-limited transmission sphere for interferometry in a geometrical ray-tracing program, followed by an analysis of its performance in a physical-optics program.

We present a hybrid diffraction model for the efficient analysis of diffractive optical elements. This model uses a scalar-based approximation over those regions of the boundary that satisfy the scalar criteria and a vector-based solution over those that do not. In analyzing diffractive optical elements (DOEs) it becomes necessary to use a vectorbased model as the feature sizes within the DOE profile approach the scale of the illumination wavelength. However, in many instances only certain regions of a profile contain such small-scale features. In these cases it is inefficient to perform a vector-based analysis over the entire profile. Therefore, we have developed a method that allows for the concatenation of scalar- and vector-based solutions. This is achieved by simply assigning the surface field values according to the scalar approximation over those regions of the profile that satisfy the scalar criteria, and using the the finite-difference time-domain (FDTD) method to determine the surface fields over those regions that contain small-scale features. In combination these methods create a surface profile that can be propagated to any plane, or region, of interest. In the course of this paper we discuss the formulations of scalar diffraction theory, the FDTD method, and the method for propagating the concatenated boundary fields.

The spectral characteristics of a Fabry-Perot spectrometer (filter) formed by a pair of identical linearly chirped fiber Bragg gratings in optical fiber are studied numerically using the characteristic matrix method. The results indicate that based on available techniques and materials, one can fabricate such filters with a finesse as high as 104 and contrast as high as 109 by inscribing a pair of identical linearly chirped fiber gratings. Achieving such superfinesse and contrast in a conventional Fabry-Perot (FP) is very difficult because of fabrication complexities in achieving superflatness at ultrahigh reflectivities simultaneously and keeping them aligned. Our numerical simulation also indicates that the spectral characteristics of the fiber FP (FFP) filter can be approximated by that of a classical plane mirror FP (PFP) with a mirror separation of L + ?L, where L is the length of any one of the two gratings and ?L is the separation between the two gratings. This analogous characteristic enables one to estimate time domain and other behavior of an FFP from already established PFP analysis. Thus, miniature FFPs can be used not only to achieve ultralow, crosstalk in wavelength division multiplexing (WDM), but they can also be integrated into miniature, hybrid, spectral sensors (such as Brillouin and Raman sensors) where ultrahigh contrast with superresolution is required.

Thin-film helicoidal bianisotropic medium (TFHBM) layers with large local linear birefringence are fabricated using the serial bideposition technique. The viability of axially excited TFHBM layers as circular polarization filters is experimentally demonstrated. The presence of index-matched cover and substrate media is shown to enhance the filter performance of a TFHBM layer. A thinner TFHBM layer is required for adequate performance when the local linear birefringence is higher.

The side-match vector quantizer (SMVQ) is a well known method for image coding. We propose the smooth side-match vector quantizer (SSMVQ). The SSMVQ improves coding quality by making smoother the transition of the gray levels of the adjacent blocks. Moreover, the smooth overlap-match vector quantizer (SOMVQ) is proposed to improve the performance of the overlap-match vector quantizer (OMVQ). Experimental results reveal that SSMVQ and SOMVQ have the higher peak SNR (PSNR) values, better visual perception quality, and lower bit rate than SMVQ and overlap-match vector quantizer (OMVQ), respectively.

We propose a new approach to adaptive image regularization based on a neural network, hierarchical cluster model (HCM). The HCM bears a close resemblance to image formation. Its sparse synaptic connections are effective in reducing the computational cost of restoration. We attempt to achieve adaptive restoration by assigning entries of a novel, optimized regularization vector to each homogeneous image region. The degraded image is first segmented and partitioned into smooth, texture, and edge clusters. It is then mapped onto a three-level HCM structure. An evolutionary scheme is proposed to optimize the regularization vector by minimizing the HCM energy function. The scheme progressively selects the well-evolved individuals that consist of partially restored images, their corresponding cluster structures, segmentation maps, and the optimized regularization vector. Experimental results show that the new approach is superior in suppressing noise and ringing at the smooth background while effectively preserving the fine details at the texture and edge regions. An important feature of the method is that the empirical relationship between the optimized regularization vector and the local perception measure can be reused in the restoration of other degraded images. This generalization removes the overhead of evolutionary optimization, thus rendering a very fast restoration.

This paper describes a new method of lossy still-image compression, called adaptively scanned wavelet difference reduction (ASWDR). The method produces an embedded bit stream with region- of-interest capability. It is a simple generalization of the compression method developed by Tian and Wells, which they have dubbed wavelet difference reduction (WDR). While the WDR method employs a fixed ordering of the positions of wavelet coefficients, the ASWDR method employs a varying order that aims to adapt itself to specific image features. This approach enables ASWDR to outperform WDR in a rate- distortion sense, and to essentially match the rate-distortion performance of the widely used codec SPIHT of Said and Pearlman. ASWDR compressed images exhibit better perceptual qualities, especially at low bit rates, than WDR and SPIHT compressed images. ASWDR retains all of the important features of WDR: low complexity, region-of-interest capability, embeddedness, and progressive SNR.

Semisupervised classification is one approach to converting multiband optical and infrared imagery into landcover maps. First, a sample of image pixels is extracted and clustered into several classes. The analyst next combines the clusters by hand to create a smaller set of groups that correspond to a useful landcover classification. The remaining image pixels are then assigned to one of the aggregated cluster groups by use of a per-pixel classifier. Since the cluster aggregation process frequently creates groups with multivariate shapes ill suited for parametric classifiers, there has been renewed interest in nonparametric methods for the task. This research reports the results of an experiment conducted on six Landsat Thematic Mapper images to compare the accuracy of pixel assignment performed by four nearest neighbor classifiers and two neural network paradigms in a semisupervised context. In all the experiments, both the neighbor-based classifiers and the neural networks assigned pixels with higher accuracy than the maximum-likelihood approach. There was little substantive difference in accuracy among the neighborhood-based classifiers, but the feedforward network was significantly superior to the probabilistic neural network. The feedforward network classifier generally produced the highest accuracy on all six of the images, but it was not significantly better than the accuracy produced by the best neighbor-based classifier.

As an attempt to achieve realistic image representation, an efficient segmentation algorithm is proposed. The proposed method aims to represent homogeneous visual objects with few regions while preserving the semantic contents of an image as much as possible. This strategy is quite useful for a typical ''head and shoulder" video sequence, since visually homogeneous objects occupy large portions of the image. For this objective, we adopt a bottom-up approach using spatial domain information only. Initially, precise initial image segmentation is performed using an efficient marker extraction algorithm. Then, we classify the initially segmented regions by gradients, and apply an ordered region-merging algorithm to each class to reduce the number of regions. Finally, we eliminate redundant small regions by considering their neighborhoods. Experimental results show that the segmentation result can be a well self-contained representation of an image since it preserves most of the perceptually important image components.

A Hausdorff distance (HD), which is defined between two point sets in two-dimensional (2-D) binary images and does not require us to establish correspondences, is one of common measures used for object matching. The proposed hierarchical HD (HHD) matching algorithm, based on coarse-to-fine matching, reduces the computational complexity greatly by using the pyramidal structures that consist of a distance transform (DT) map pyramid and an edge pyramid. in the proposed HHD matching, an automatic thresholding method is presented to select candidate positions at each level, in which the threshold value is determined based on the property between adjacent levels of a DT map pyramid. Computer simulation with real images and noisy images shows that the proposed HHD algorithm yields matching results similar to those achieved by the conventional HD algorithm with greatly reduced computational complexity.

The Dark HORSE 1 (Hyperspectral Overhead Reconnaissance and Surveillance Experiment 1) flight test has demonstrated autonomous, real-time visible hyperspectral detection of military ground targets with real-time cuing of a high-resolution framing camera. An overview of the Dark HORSE 1 hyperspectral sensor system is presented. The system hardware components are described in detail, with an emphasis on the visible hyperspectral sensor and the real-time processor. Descriptions of system software and processing methods are also provided. The recent field experiment in which the Dark HORSE 1 system was employed is described in detail along with an analysis of the collected data. The results evince per-pixel false-alarm rates on the order of 10-5/km2, and demonstrate the improved performance obtained by operating two detection algorithms simultaneously.

The resolution of many viewing and staring systems is often restricted by the spatial limited resolution of its sensing device rather than by diffraction limits related to the optical system. This spatial resolution of the sensor, when limited by the pixels' dimensions, is coined hereby "geometrical resolution." We suggest a technique for overcoming this limit, thus obtaining "geometrical superresolution." The proposed approach is based on capturing a set of images, interlacing their pixels and applying special filtering over the interlaced image. The number of the captured images N corresponds to the desired resolution improvement. Each image in the set is captured after a lateral shift of the sensor by a subpixel distance of ?x/N, where ?x is the size of the sensor's pixel. The main contribution is in designing a special energetic efficient mask that is attached to the sensing plane. Without this mask, unrecoverable resolution loss prevents a qualitative reconstruction to be obtained.

Erbium-doped optical fibers are designed using the refractive index difference, fiber core diameter and Er concentration as parameters in the numerical calculations. The fibers are fabricated with modified chemical vapor deposition and solution doping techniques, which are described in detail. Preform and fiber parameters also measured at several laboratories participating in the European Community's COST 217 project are given and compared with the fiber design values. The produced Er fiber is utilized in prototype fiber amplifiers. The differential efficiency of an Al-doped Er fiber is measured to be 77%.

Optical end of line metrology (OELM) is a new method to measure relative line shortening effects using conventional optical overlay instruments. in this technique, a frame that has two adjacent sides that are constructed of lines and spaces is imaged onto a wafer. Since gratings below 0.5 ?m cannot be resolved using conventional optics, the alignment tool sees the sides composed of lines and spaces as solid edges. The purpose of this paper is to characterize the errors implicit with this approach. OELM results are compared with scanning electron microscopy (SEM) line shortening measurements, which showed that optical measurements exaggerated the physical effect. A simple aerial image analogy predicts that optical line shortening measurements are pitch and duty cycle dependent. OELM measurements require calibration as a function of pitch and line width.

A spatiotemporal phase unwrapping method is proposed that combines the dynamic optic fiber interferometric fringe projection method and the phase-shifting technique. A fringe projection fiber optic phase- shifting interferometer was set up, in which the fringe spacing and rotation can be easily adjusted. In the first step of the spatiotemporal phase unwrapping method, a large effective wavelength can be chosen so that the phase jump at the discontinuous profile is less than ?, then the spatial phase unwrapping method can be applied. Several intermediate phase maps can be obtained by changing fringe pitch and fringe orientation, reducing the effective wavelength step by step, and unwrapping each pixel along the time axis. In the final step, a high precision result can be obtained. A minimum number of steps can be chosen to obtain the required accuracy according to the conclusion presented. An experimental result is presented for the measurement of a discontinuous object and shows the validity of the proposed method.

The Gabor expansion of an arbitrary optical signal in terms of Gaussian signals that form an NxM truncated von Neumann lattice in the time-frequency plane is studied. The accuracy of the expansion for various values of (NxM) and various widths of the Gaussians is examined. The accuracy of the method is also demonstrated by comparing and contrasting the Wigner function for both the original and reconstructed signals. Practical implementation of the method through ''Gabor analyzers" that directly measure the Gabor coefficients of optical signals is also discussed.

A compact, dual-probe, fiberized Sagnac ultrasound sensor (SUS) capable of detecting ultrasonic waves of various types is described. The main advantages of the proposed device are (1) the ability to detect ultrasonic waves on the surface (extrinsic mode) as well as in the interior of structures (intrinsic mode); (2) improved SNR, which has been achieved using an optical frequency shifting technique for biasing to quadrature and for elimination of parasitic interference between the desired sampling beams and other nonessential beams in the interferometer; (3) the ability to detect ultrasonic signals on rough surfaces by the use of an electromechanical speckle-hunting technique (in the extrinsic mode); (4) dual-probe configuration; and (5) directional sensitivity to ultrasound when the system is operated in the intrinsic mode or on a polished surface in the extrinsic mode. Several applications of the SUS for nondestructive characterization of material properties and testing of structures for flaws are presented. Bulk longitudinal waves, Rayleigh waves and Lamb waves are measured using the SUS to (1) characterize piezoelectric ultrasonic transducer beam profiles, (2) study the scattering of ultrasonic waves by flaws, (3) identify and size surface breaking flaws in aircraft wheel components, and (4) monitor in real time the cure process of an epoxy resin.

To analyze aqueous ammonia, we develop a simple and inexpensive spectrometer that consisted of a diode laser, optical fibers and a photodiode detector. The concentrations of ammonia are determined by the indophenol blue method. Preliminary results on the performance of the system in terms of the stability of output intensity, linearity of a calibration curve, and the determination of ammonia in environmental samples are described. The measured concentrations are also compared with those on a conventional spectrometer. The results indicate the potential to develop a portable spectrometer useful for on-site monitoring of environmental samples. Diode lasers are favored due to the monochromatic and high intensity radiation, and the reduction of heavy and expensive components in available commercial spectrometers. For the tested on-site analyzer, a photomultiplier tube, a monochromator, and complicated optical alignments are not required.

A compact and mobile optical spectrometer for the detection of total atmospheric ozone is described. The ozone column is determined by differential absorption of solar radiation at two adjacent UV wavelengths (328 and 329 nm). Measurements made during an 18 month period in South Germany confirm previous satellite data during the periods of October 1996 to February 1997, April 1997 to February 1998 and May 1998, but exhibit some ozone decrease in March 1997, as well as in April 1998.

We present a novel multiresolution block-matching scheme- based on multistage motion vector quantization (MSMVQ) for videocoding. The MSMVQ performs the block matching on the coarse-to-fine basis where rough results of block matching are provided at the earlier stages of MSMVQ with fast computation, and accurate results are given at the later stages with additional arithmetic complexity. The multiresolution block matching process at each stage utilizes the results of its previous stages to further enhance the performance of MSMVQ. Simulation results show that the algorithm can be effectively applied to the applications where video decoders having different demands on the computational complexities and resolutions for block matching are desired to share the same video encoder.