Charge carrier diffusion modeling, (CCDM), has been developed extensively during the past few years as a tool for performance evaluation of photovoltaic focal plane arrays. Some performance parameters which have been successfully predicted using CCDM include: line spread function and MTF; quantum efficiency; absolute responsivity; relative spectral response; junction effective optical width; and crosstalk. CCDM has also been used as an aid in evaluating related focal plane engineering topics, such as noise equivalent irradiance, production yield of acceptable array chips, array response to test set conditions, optimization of detector geometry, performance tradeoff analysis, and identification of performance degradation sources.
The development of a first principles computer simulation of a generic pyroelectric thermal detector is described. The formulation of the pertinent equations (based on a thorough literature survey) is presented. This simulation incorporates a finite difference treatment of the transient three-dimensional thermal response of composite focal plane arrays, with treatments of the signal generation, readout and processing, including all pertinent noise sources. A number of simplified problems having analytical solutions were treated to validate various portions of the simulation to generally much less than one percent in absolute temperature. Performance estimates were made for conceptualizing several configurations and materials. Results are given for the thermal crosstalk between adjacent detectors and the responsivities (for a Honeywell configuration) are compared well with Honeywell's predictions and measurements. Crosstalk predictions were within a few percent of Honeywell's while responsivities were within a factor of two. Reasons for these differences are discussed.
In a thermal imaging system each optical refracting surface can reflect internal radiation down onto the detector. Narcissus is the spurious signal generated by the composite effect of these reflections. A function is derived which represents the narcissus effect as an equivalent temperature differences in the scene. The function can be approximated by a simple model based on paraxial ray-tracing. The effectiveness of the model is demonstrated by four examples.
There is an increasing need for an objective figure of merit for specification of infrared imaging sensors in military applications such as manned airborne reconnaissance systems. This paper suggests a very simple modification of the subjective MRTD figure of merit used for predicting observer performance with infrared FLIRS. Objective MRTD can be used for modeling and control of both FLIR and infrared line scan sensor performance especially in those applications where the infrared sensor must interface with a total reconnaissance or weapon system. Such total systems, or sensor suites, usually consist of many diverse components each made by different vendors. Often interface must be established with equipment made in foreign nations. It is therefore imperative that the performance of a thermal imaging sensor be precisely measured in an unequivocal manner. Objective MRTD is suggested as an easily measured figure of merit for thermal imagers which is useful in meeting these objectives.
The problem considered herein is that of tracking a well-resolved object with an active direct detection sensor system. The sensed object image is corrupted by additive as well as multiplicative (speckle) noise. A steady-state object tracking filter based on a discounted weighted least-squares procedure is derived in closed-form, along with its error covariance properties. The filter, which is of a polynomial-in-time type, estimates the position, velocity, and acceleration of an object moving in space, based on noisy measurements of position offset in the focal plane array. The position offset measurements are obtained using the Fitts correlation algorithm. The error covariance matrix of the Fitts algorithm is derived in closed-form for both additive and multiplicative noise. The discounted weighted least-squares filter has the advantage of requiring only a single parameter, the discount factor, as opposed to three parameters in the conventional a-0-y filter. Most importantly, the discount factor guarantees a certain degree of stability in the filter. Simulation results are given for the case of a computer-generated speckled image. Tracking performance results are presented for the discounted weighted least-squares filter, the a-0-y filter, and the recursive Kalman filter, using the Fitts correlation algorithm for generating position offset measurements in all cases.
A detailed IR sensor simulation has been developed which models the IR scene and optical system for a variety of space-based military missions including SDI and ADI. The IR scene model includes a background radiance map, a foreground map (such as clouds), and a target map. This target map includes a hardbody radiance as well as plume radiances for a variety of boosting or afterburning targets. The scene generation portion of the software includes the combination of the various scene elements into a high resolution map which is then propagated through the atmosphere to the sensor. The atmospheric transmission and turbulence effects are estimated and included in the simulation. The actual sensor performance is then simulated using a two-dimensional Modulation Transfer Function (MTF) which includes errors due to low frequency aberrations, mid and high spatial frequency surface roughness, diffraction effects, and image motion. This sensor MTF is then multiplied by the Fourier Transform of the high resolution IR scene, and then transformed back to the spatial domain, to produce the high resolution blurred scene which is incident on the detector focal plane array (FPA). Finally, the software simulates the FPA and electronic components to arrive at the sampled, low resolution output from the detector. Detector parameters including gain, bias, crosstalk, quantum efficiency and electronics noise, photon noise statistics, and A/D conversion are all simulated in the model.
We derive an expression for the time dependent temperature distribution in a finite solid state laser rod for an end-pumped beam of arbitrary shape. The specific case of end pumping by circular (constant) or gaussian beam is described here. We discuss the temperature profile for a single pump pulse and for repetetive pulse operation. The particular case of the temperature distribution in a pulsed Titanium Sapphire rod is considered.
This paper addresses methods of modeling some of the dynamic effects that affect a laser beam propagation through the atmosphere. Among the components adn phenomena are adaptive optics, atmospheric turbulence with wind shear and beam steering, jitter, and integration of the energy on the target. The emphasis is on the development of simple but accurate models which are readily implemented in a physical optics code. The system of our immediate interest is described in a companion paper.' Figure 1 shows some aspects of space relay system.
Many optical systems have demanding requirements to package the system in a small 3 dimensional space. The use of computer graphic tools can be a tremendous aid to the designer in analyzing the optical problems created by smaller and less costly systems. The Spectra Physics grocery store bar code scanner employs an especially complex 3 dimensional scan pattern to read bar code labels. By using a specially written program which interfaces with a computer aided design system, we have simulated many of the functions of this complex optical system. In this paper we will illustrate how a recent version of the scanner has been designed. We will discuss the use of computer graphics in the design process including interactive tweaking of the scan pattern, analysis of collected light, analysis of the scan pattern density, and analysis of the manufacturing tolerances used to build the scanner.
CAD/CAM systems permit the manufacture of real physical parts on a milling, or other shop, machine which has received its instructions from the design system. However, a persistent problem has been the lack of a link between the optical CAD software and the mechanical CAD software. A past paperl described software which bridges this gap by producing an Interactive Graphics Exchange Standard (IGES) file from the neutral plot files of the optical model. This IGES file can then be directly read into the mechanical drafting and design programs, ANVIL and GEOMOD. Although the latest enhancements to this older interface will be addressed, the main thrust of the paper is the description of three completely new capabilities: 1) the ability to directly read NASTRAN-generated strain data for optical surfaces into the optical design program CODE-V, 2) the direct passage of a standard lens prescription in CODE-V to the program APART for stray light analysis, and 3) the entry of mechanical drafting data directly into APART via IGES files. This last interface is particularly exciting because it allows, for the first time, completely graphical entry of structural data into an optical analysis program. A description of the typical procedures used to accomplish the transfer of data is provided along with several examples. Plans are presented to use these new software links to achieve paperless manufacture of opto-mechanical systems. Since the current algorithms have been designed with an eye on eventual improvements, an assessment of current limitations and future enhancements is also included.
Optical system modeling is particularly important for simulation of laser beam propagation over large distances. Ground-satellite-ground propagation is such a system; it includes three stages: the beam preconditioning stage to ready it for propagation over large distances, beam interception at a relay mirror, (possibly including additional beam conditioning), and beam propagation to and detection at a ground-based target board. The target board is designed to measure the efficiency of power transfer in this experiment by integrating over the resulting redistribution of the laser power density. Key elements of the model are the ground-based laser; adaptive optics for aberration correction and correction of atmospheric effects; beam expander; propagation through a turbulent atmosphere; one or more space-based relay mirrors; and downward propagation to a ground-based target board. Laser beam propagation over large distances is best done above the atmosphere to reduce the effects of atmospheric turbulence on the phase of the laser beam. This paper describes the optical modeling of the propagation of a laser beam between two points on the surface of the Earth, using relay mirrors on one or more satellites. Because of the importance of the atmospheric and propagation effects, computer modeling is necessary to develop understanding of the first order design parameters. Some of the important design parameters are characterization of the adaptive mirror for the atmospheric aberration correction, size of the beam expander, apertures of the relay mirror, and acceptable levels of jitter. An example of a propagating laser beam is presented, demonstrating the fully three-dimensional nature of this model.
The ultrafast laser scanner microscope at the Optical Sciences Center, University of Arizona, was designed as a prototype for a new class of optical microscopes to serve directly as an optical data-acquisition peripheral for a computer. The design called for the ability to digitally record large specimen areas at high spatial resolution within a short period of time. These design parameters imply very high data rates and very large data volumes. Optical systems with the required point-spread function, such as conventional high-aperture microscope objectives, typically offer a field-of-view diameter of approximately 250 jam. Because this field is small, areas of frame overlap must be rescanned. Consequently, in the design of the laser-scanner objective, a very wide field of view was specified. The laser scanner design work began in 1981, and the instrument was completed in 1986. It has since been transferred to the Department of Pathology, where it is being used to digitally record histopathologic images. The instrument design implements a number of novel features. The following is a critical review of design and instrument performance features, particularly the optomechanical aspects of the instrument. Previous papers cover the design evolution of the ultrafast laser scanner microscope.1-4
A computer simulation of a phased array imaging system has been written. The OTF of an imaging system with a specified number of annular apertures can be computed directly by a convolution if there are no phase aberrations. If phase aberrations are to be included (piston, two axis tilt, focus, and uncorrelated aberrations for each subaperture), the OTF is computed via the point spread function. A two-dimensional scene is created from elementary shapes and transformed to the spatial frequency domain by a two-dimensional fast Fourier transform (FFT). The spectrum is multiplied by the optical transfer function (OTF) and by a function which accounts for pixel size and shape. After being properly scaled by a radiometry analysis the scene is transformed back to the spatial domain. Pixel fluxes are then computed and Poisson noise and detector noise are added.
The finite element method has become the nearly exclusive province of the structural mechanics community. However the method has broad applicability to many other fields of analysis as well. One very useful application of the finite element method is in optical systems, especially where the optical performance depends upon suitable behavior of the structural support system. Rigorous analogs allow an analyst to model many different optical systems and phenomena in finite elements. This paper explores the nature of optical analogs in finite element analysis and attempts to demonstrate how their quantitative rigor can lead to very accurate analysis of otherwise difficult problems. The discussion includes examples of both geometric optical analogs and physical optical analogs so the reader may appreciate the broad applicability of the method.
A mock ray tracing procedure for modeling complex optical systems with "blackbox" lens modules is described. The procedure permits the inclusion of third order aberrations, as well as axial and lateral chromatic errors. Examples of f-theta scanning lens simulation and zoom lens design with lens modules are given, using the CODE V lens design program.
A simple method for calculating encircled energy is presented which is valid for circular extended objects imaged by systems with circular or annular apertures. The system may be diffraction limited or have an MTF which is circularly symmetric. Previously published papers have required nested double integrals to obtain results. This new method takes advantage of symmetry and properties of the Fourier Transform to reduce the problem to one single integral. Because the MTF of an optical system always has a cut-off frequency the integration limits of the remaining zero order Hankel Transform are finite speeding the calculation further. With this new algorithm computation times have been reduced significantly while accuracy has been improved.
The general mixed optical aberration function produced by a centered lens is calculated using geometrical raytrace of three dimensional rays in conjunction with linear interpolation. This technique has proven to be an efficient tool in lens design applications where the complete aberration function is required to assess the overall dependence of image quality on the lens construction parameters. Our approach does not use the Zernike circular polynomials which limits quality assessment to specific modes. We demonstrate the efficiency of this approach by presenting numerical results characterizing the quality of a number of practical lenses. Optimum lens construction parameters are indicated by the Point Spread Function (PSF) of the optical system at the focus and the focal shift, both, calculated from the aberration function. We examine the effects of various types of lens construction parameters and materials on image quality.
In the past photographic "taking" lenses and, in particular, those for the motion picture industry i.e. cinematographic lenses have had a mixed career due to inconsistencies between the processes of lens design, manufacture, testing and calibration and practical assessment in the customer domain. Usually these inconsistencies can be attributed to differences between the comparison of, a lens design "scientifically" made and final evaluation in an "artistic" manner. The following paper addresses the processes of lens design, manufacture, testing and calibration using a combination of acquired practical experience and modern test and calibration methods. Various performance aspects are separately addressed and considered in terms of different means of measurement.
Commonly used methods of tolerance generation for optical systems are reviewed briefly; the most well known being that of inverse sensitivity. Another method is then discussed which will minimise a 'cost' function, subject to multiple performance constaints. The optimisation will allow either the worst case values of the effects or the RSS values to fulfill the specifications. Examples of such a set will be given, and their effects on the variation of MTF illustrated. In addition, practical upper and lower tolerance limits may be imposed.
A technique has been developed for analyzing the effect on image quality of various optical surface errors. The technique can be applied to surface errors typically seen in optical fabrication, such as rolled edges or zonal errors or discontinuities. The technique allows rapid assessment of surface error effects for known errors as well as for typical errors, and gives insight into the general problem of optimizing aspheric surfaces.
A new variant of our general reflector analysis program includes special computations to simulate the behavior of light after striking a diffusing surface. Each ray is converted at that surface, into a group of rays whose angular distribution is dictated' by the surface properties. The program accepts an experimentally determined "map" of the reflected beam profile for a given material, and uses this map to generate a group of reflected rays for each incident ray. Computed light distribution is compared with experimental values for a simple example of diffusely reflecting bead-blasted aluminum.
A methodology for determining the relative cost of optical components, given the basic specifications of these components such as dimensions, tolerances, materials, coatings, etc., is developed here. Each of the specifications for an optical component must be translated into a manufacturing process which has the capability to achieve the required specifications. To achieve the required specifications, the use of both labor and materials is involved. The actual amount of labor and materials required to complete the process is not derived, however the hypothetical relationship of specification to process is shown. An actual mathematical relationship of specification versus process will be derived in future papers.
Grinding is modelled as a packing of particles connected by interparticle bonds that are characterized by a given force-displacement relation for the nomal and tangential components. Particles on the surface of the packing may be break or sheared off if the forces between particle contacts exceed the strength of the bonds. The forces between contacts due to surface fraction are calculated based on the principle of equilibrium and the compatibility between particles. The parking of particles is generally random and the surface friction including normal and shear stresses be a function of time (i.e. variable speed of grinder). The roughness of the surface is smoothed due to the "BREAK OFF" of surface particles. The mechanism is modelled microscopically to provide more insight to this phenomenon. The calculated results are compared and disscusseed with existing theories such as Prinston Equation to show its validity.
Shrinkage causes both the appearance & dimension defects of the injected plastic lens. We have built up a model of state equations with the help of finite element analysis program to estimate the volume change (shrinkage and swelling) under the combinations of injection variables such as pressure and temperature etc., then the personal computer expert system has been build up to make that knowledge conveniently available to the user in the model design, process planning, process operation and some other work. The domain knowledge is represented by a R-graph (Relationship-graph) model which states the relationships of variables & equations. This model could be compare with other models in the expert system. If the user has better model to solve the shrinkage problem, the program will evaluate it automatically and a learning file will be trigger by the expert system to teach the user to update their knowledge base and modify the old model by this better model. The Rubin's model and Gilmore's model have been input to the expert system. The conflict has been solved both from the user and the deeper knowledge base. A cube prism and the convex lens examples have been shown in this paper. This program is written by MULISP language in IBM PC-AT. The natural language provides English Explaination of know why and know how and the automatic English translation for the equation rules and the production rules.