This PDF file contains the editorial “Finding the Frontier” for OE Vol. 41 Issue 12

We report a pump-controlled optical signal switch consisting simply of a length of dual-core erbium-doped fiber. It has the following attractive features: (1) wide bandwidth signal switching covering the wavelength range 1552 to 1585 nm, and (2) switching efficiency of nearly 100%.

This paper proposes an adaptive 2-phase level set image segmentation algorithm to improve the original level set-based piecewise constant Mumford-Shah model. With the introduction of a multiplicative gain field, the model is adaptive to intensity inhomogeneity, thus tending to obtain the actual boundaries of the objects, as well as a property-seeking and -driven classification algorithm.

We demonstrate how to make a twisted nematic liquid crystal display act as a programmable wave plate array. This is an unusual phase plate because the orientations of the fast and slow axes are rotated from one side of the display to the other.

This PDF file contains the editorial “Advances in UV Ground- and Space-based Measurements and Modeling” for OE Vol. 41 Issue 12

A tutorial review is provided of ultraviolet (UV) and visible radiation transport in the atmosphere-ocean system. Emphasis is placed on the basic physical principles involved, rather than on mathematical/numerical aspects. To illustrate the application of the theory, the effects of an ozone depletion on UV irradiance at the surface are discussed. A comparison of measured and predicted UV penetration into the ocean under the Antarctic ozone hole is also provided. Our ability to model the transport of UV and visible light in the atmosphere-ocean system seems to be limited to a large extent by the lack of information about inherent optical properties of marine constituents, especially in the UV spectral range. To make progress in this area, measurements are needed of spectral absorption and scattering coefficients and volume scattering functions. Identification of the marine constituents that give rise to these optical properties and their variability in space and time is also of crucial importance.

Satellite instruments currently provide global maps of surface UV irradiance by combining backscattered radiance data with radiative transfer models. The models are often limited by uncertainties in physical input parameters of the atmosphere and surface. Global mapping of the underwater UV irradiance creates further challenges for the models. The uncertainties in physical input parameters become more serious because of the presence of absorbing and scattering quantities caused by biological processes within the oceans. We summarize the problems encountered in the assessment of the underwater UV irradiance from space-based measurements, and propose approaches to resolve the problems. We have developed a radiative transfer scheme for computation of the UV irradiance in the atmosphere-ocean system. The scheme makes use of input parameters derived from satellite instruments such as the total ozone mapping spectrometer (TOMS) and sea-viewing wide field-of-view sensor (SeaWiFS). The major problem in assessment of the surface UV irradiance is to accurately quantify the effects of clouds. Unlike the standard TOMS UV algorithm, we use the cloud fraction products available from SeaWiFS and MODIS to calculate instantaneous surface flux at the ocean surface. Daily UV doses can be calculated by assuming a model of constant cloudiness throughout the day. Both SeaWiFS and a moderate resolution imaging spectroradiometer (MODIS) provide some estimates of seawater optical properties in the visible. To calculate the underwater UV flux, the seawater optical properties must be extrapolated down to shorter wavelengths. Currently, the problem of accurate extrapolation of visible data down to the UV spectral range is not solved completely, and there are few available measurements. The major difficulty is insufficient correlation between photosynthetic and photoprotective pigments of phytoplankton absorbing in the visible and UV, respectively. We propose to empirically parameterize seawater absorption in the UV on a basis of available datasets of bio-optical measurements from a variety of ocean waters. Another problem is the lack of reliable data on pure seawater absorption in the UV. Laboratory measurements of the UV absorption of both pure water and pure seawater are required.

Satellite instruments provide global maps of surface UV irradiance by combining backscattered radiance measurements with radiative transfer models. The accuracy of the models is limited by uncertainties in input parameters representing the atmosphere and the Earth's surface. To reduce these uncertainties, we have made enhancements to the currently operational TOMS surface UV irradiance algorithm (Version 1) by including the effects of diurnal variations of cloudiness, an improved treatment of snow/ice, and a preliminary aerosol correction. We compare results of the version 1 TOMS UV algorithm and the proposed version. We evaluate different approaches for improved treatment for average cloud attenuation within a satellite pixel, with and without snow/ice on the ground. In addition to treating cloud transmission based only on the measurements at the local time of the TOMS observations, the results from other satellites and weather assimilation models can be used to estimate atmospheric UV irradiance transmission throughout the day. A new method is proposed to obtain a more realistic treatment of the effects from snow-covered terrain. The method is based on an empirical relation between UV reflectivity and measured snow depth. The new method reduces the bias between the TOMS UV estimations and ground-based UV measurements for snow periods. We also briefly discuss the complex problem of estimating surface UV radiation in presence of UV-absorbing aerosols. The improved (Version 2) algorithm can be applied to reprocess the existing TOMS UV irradiance and exposure estimates (since November 1978) and to future satellite sensors (e.g., GOME-2, OMI on EOS/Aura, and Triana/EPIC).

A worldwide forecast of the global solar erythemally effective UV irradiance, also called UV Index, under clear skies is presented. The forecast was established to inform the public about the expected amount of erythemally effective ultraviolet radiation for the next day. The forecast enables the user to choose the appropriate sun protection tool and to adopt the personal behavior to reduce the risk of health damage. The forecast as well as the radiation model are validated using measurements from a Robertson-Berger meter made at four continents (Antarctica, Australia, America, and Europe). The measurements cover the latitudinal range from 67.4°N to 60.1°S. Since the UV Index was determined to quantify UV radiation protection, the goal of a forecast scheme is to avoid underestimation. Therefore the frequency of underestimation was used as quality criteria also. The Austrian forecast model shows less than 7.3% underestimation in all cases, whereas 5.0% result from the radiation model and 2.3% from the used value of the total ozone content of the atmosphere.

Comparison of measured UV irradiance with estimates from satellite observation is potentially effective for the validation of data from the two sources. Summer data from ten Canadian Brewer sites were compared in this study with noon UV irradiance estimated from total ozone mapping spectrometer (TOMS) measurements. In general, TOMS estimates can successfully reproduce long-term and major short-term UV variations. However, there are some systematic differences between the measurements at the ground and satellite-retrieved UV irradiance. From 3 to 11% of the Brewer-TOMS difference can be attributed to the Brewer angular response error. This error depends on the solar zenith angle and cloud conditions, and is different from instrument to instrument. When the angular response of the Brewer instrument is considered and applied, the Brewer data are still lower than TOMS-estimated UV irradiance by 9 to 10% on average at all sites except one. The difference is close to zero at one station (Saturna Island), possibly due to its much cleaner air. The bias can be seen in clear sky conditions and at the 324-nm wavelength, i.e., it is not related to local cloud conditions or absorption by ozone or SO2.

Ground-based global solar ultraviolet-B (UV-B: 290 to 320 nm) irradiance has been measured by a narrow band UV-B radiometer at Tokai University, Hiratsuka, Japan (35°21'N, 139°16'E) for a 10-year period from October 1990 to September 2000. A precise calibration of the UV-B radiometer was periodically performed, and the yearly decay in sensitivity was found to be –3.7%. Using this decay rate, the measured UV-B irradiance was corrected, and the long-term trends of the UV-B irradiance were estimated. When the seasonal variation was eliminated by taking 12-month moving averages, an increasing trend in the UV-B irradiance was demonstrated to be 1.57% per year. Moreover, to remove quasibiennial oscillation (QBO), 26-month moving averages were applied to the UV-B irradiance normalized by the global total (300 to 3000 nm) solar irradiance. An increasing trend in the normalized UV-B irradiance was found to be 1.22% per year. In winter, the clearly increasing trend of the UV-B irradiance was statistically significant, although the increasing tendency of the UV-B irradiance in other seasons (spring, summer, and fall) is not clear. A significant inverse correlation was confirmed between the UV-B irradiance normalized by the global total solar irradiance and the effective ozone amount defined as total ozone amount ×secθ, where θ is the solar zenith angle. These findings give supporting evidence for a direct relationship between solar UV-B irradiance and the stratospheric ozone amount. In conclusion, the increasing trend of global solar UV-B irradiance, especially in winter, was confirmed in response to stratospheric ozone loss in mid-latitude Japan in the 1990s.

We show the comparisons between ground-based measurements of spectrally integrated (300 to 380 nm) UV irradiance and satellite estimates from the total ozone mapping spectrometer (TOMS) total ozone and reflectivity data for the whole period of TOMS measurements (1979 to 2000) over the Meteorological Observatory of Moscow State University (MO MSU), Russia. Several aspects of the comparisons are analyzed, including effects of cloudiness, aerosol, and snow cover. Special emphasis is given to the effect of different spatial and temporal averaging of ground-based data when comparing low-resolution satellite measurements. The comparisons in cloud-free conditions with different aerosol loading have revealed positive TOMS bias with interannual variations within +5 to +20%. Reprocessing the TOMS data by accounting for boundary layer aerosol absorption (single scattering albedo ~0.9) removes the bias for all-sky conditions, except for completely overcast situations, where the TOMS bias remains large (15 to 17%). The single scattering albedo was independently verified using sun photometer (CIMEL Electronique) sun and sky-radiance measurements at MO MSU in September 2001. The analysis of interannual UV variations shows quite similar behavior for both TOMS and ground measurements (correlation coefficient r0.8). No long-term trend in the annual mean bias was found for all conditions. Both TOMS and ground data show positive trends in broadband UV irradiance (300 to 380 nm) between 1979 and 2000. The UV trend is attributed to decreases in both cloudiness and aerosol optical thickness during the late 1990s over the Moscow region. However, if the analyzed period is extended to include the preTOMS era (1968 to 2000 period), no trend in ground UV irradiance is detected.

At mid-latitudes, ozone variations are mainly caused by horizontal and vertical air transport, while the intensity of sources and sinks play a minor role. A short-lived episode of low ozone values over central Europe on 29 November 2000 was observed at some Italian Brewer stations. The total ozone was reduced by an episodic intrusion of tropical air having an ozone column substantially lower than normal. This reduction was not caused by photochemical reactions, which are responsible for the ozone reduction over Antarctic regions. The investigation was based on the O₃ tendency equations for determining the dynamic cause of the low ozone event. Ultraviolet (UV) irradiances variations were also analyzed to determine how weather patterns may lead to enhanced levels of UV radiation within a few days, which, however, was not found to be biologically harmful.

The Alps are one of the regions where the highest UV levels are measured in Europe. Solar UV radiation increases with altitude mainly due to decreasing amounts of air molecules, ozone, aerosols, and clouds in the atmosphere as well as due to snow covered surfaces. The altitude effect plays an important role in the understanding of the UV radiation field in mountainous terrain. Therefore simultaneous measurements of erythemal UV radiation (broadband) on three different heights were performed in the Alps over more than four years. Under clear-sky conditions, the altitude effect of daily noon-time yearly mean values of direct, diffuse, and global erythemal UV radiation results in 17.4%/1000 m (direct), 8.5%/1000 m (diffuse), and 10.7%/1000 m (global). Seasonal variations of the altitude effect are mainly influenced by changes of solar elevation, albedo values, and turbidity levels during the year. Measured altitude effects are also compared to calculated values obtained by the MODTRAN radiative transmission model.

At present the United States Department of Agriculture (USDA) Reference Spectroradiometric Network consists of three sites: Table Mountain, Colorado, Lamont, Oklahoma (the ARM program SGP site), and Beltsville, Maryland. At each site we deploy and continuously operate a 1-m cascaded additive-double Czerny-Turner scanning monochromator with a bialkali photomultiplier and photon-counting detection. Lambertian fore-optic errors are less than 1% over the range of zenith angles from 0 to 80°. The instruments use photon counting and make measurements at 290 nm not affected by stray light under typical conditions. The basic performance specifications of the instrument were demonstrated by a prototype at the 1997 North-American UV Spectroradiometer Intercomparison. Data shown here demonstrate that these are met in routine operation. The fundamental instrument performance specifications are: Optical resolution: 0.1 nm FWHM, triangular slit-function. Wavelength reproducibility: ∓0.0025-nm 2 with 296-nm Hg retrace-scan corrections applied, ∓0.007 nm 2 over typical diurnal variability, without correction. Wavelength accuracy: Limited by calibration systematic errors. Believed to be 0.005-nm worst case. Stray light: <10–7 at 4 FWHM, 10–10 at 20 nm, slit-scattering function versus 325 nm HeCd. Angular response: less than 1% error from cosine over the range of zenith angles from 0 to 80°. Signal linearity: The instrument uses a photomultiplier with 2-ns rise-time and photon counting detection. The dual-threshold discriminator has a 700-Mhz synchronous signal counting limit. The maximum counting rates seen at the longest wavelengths are less than 10 MHz; less than 1/5 of the frequency where nonlinearity can be detected, as tested for the 1997 Intercomparison. 2000 was the first full year of operation of our instrument at the NOAA Table Mountain site (140.177 °N 105.276 °W, 1900 m asl) for which the operational and calibration frequencies justify making the data accessible to outside users for scientific application. We show performance in routine operation and issues of calibration over the period April 2000 to 31 December 2001.

Side-by-side comparisons of Langley-type Dobson AD double wavelength pair direct sun observations between SBUV/2, SSBUV flight models, and the NOAA world primary standard Dobson spectrophotometer 83 and 61 show that the SBUV/2-type instruments yield column ozone amounts that are 2% higher than the NOAA Dobson spectrometers. Similar results have been obtained with a radiometrically stable multifilter spectroradiometer (MFS) under less than ideal conditions. A new approach based on a modeled table look-up method for using zenith sky radiances to derive total column ozone from zenith clear sky conditions has been tested using SBUV/2, SSBUV flight models and the MFS equipped with narrowband interference filters at the Dobson AD wavelength pairs. These clear zenith sky observations yield total column ozone that is in good agreement with ozone from the NOAA Dobson spectrophotometer direct sun observations. New model calculations on the relationship between satellite nadir radiances and surface-based zenith clear sky radiances suggest that the combination of a very radiometrically stable surface-based spectroradiometer combined with a compact high performance double monochromator could be used to derive a common radiometric scale among satellite-borne ozone monitoring instruments.

In so-called compact two-mirror focusing and afocal systems the secondary mirror is situated at the paraxial focus of the spherical primary mirror. The peculiar conelike shape of the secondary is chosen to eliminate the spherical aberration. Thus the systems are stigmatic, but they have a great offence against the sine condition (OSC), creating uneven distribution of energy on the surface of the outgoing wavefront. There is a substantial concentration of energy toward the optical axis, similar to apodization of the pupil. Calculations of the point spread function (PSF) show that in such systems, the Airy disk is slightly enlarged (at most 20%) and the second maximum (the first bright ring) can be completely suppressed. The most effective suppression of the second and higher maxima in the diffraction pattern is obtained with an imposed central obscuration ratio of 0.5 and a primary mirror focal ratio of 2. Compact two-mirror afocal and focusing systems have intrinsic apodization, enabling modification of the diffraction pattern without the use of a variable transmission filter or coating at the pupil.

We propose a new method for fabric defect detection by incorporating the design of an adaptive wavelet-based feature extractor with the design of an Euclidean distance-based detector. The proposed method characterizes the fabric image with multiscale wavelet features by using undecimated discrete wavelet transforms. Each nonoverlapping window of the fabric image is then detected as defect or nondefect with an Euclidean distance-based detector. Instead of using the standard wavelet bases, an adaptive wavelet basis is designed for the detection of fabric defects. Minimization of the detection error is achieved by incorporating the design of the adaptive wavelet with the design of the detector parameters using a discriminative feature extraction (DFE) training method. The proposed method has been evaluated on 480 defect samples from five types of defects, and 480 nondefect samples, where a 97.5% detection rate and 0.63% false alarm rate were achieved. The evaluations were also carried out on unknown types of defects, where a 93.3% detection rate and 3.97% false alarm rate were achieved in the detection of 180 defect samples and 780 nondefect samples.

The experimental implementation of the diffractive phase element (DPE) that synthesizes three-color psudo-nondiffracting beams (PNDBs) is described. This DPE is designed with the amplitude-phase retrieval method and fabricated by using optical contact lithography and reactive-ion etching (RIE). Measurements demonstrate that the fabricated DPE has the desired function, i.e., it forms a six-segment PNDB over a finite axial region and is monochromatic in each segment.

We introduce a process for applying directly UV photopatternable materials and processing methods for the fabrication of binary diffractive optical elements. We design and model a binary axicon—an optical element that produces an almost diffraction free beam at a specified distance from the element. We also synthesize sol-gel-based hybrid glass materials and tailor their processing parameters to fit to the demands of the axicon design. A grating periodicity of 2 μm, an 850-nm structure depth, and certain morphological properties are required to meet the design parameters. The materials are synthesized using zirconium(IV)isopropoxide, methacrylic acid and methacryloxypropyltrimethoxysilane as deposition material precursors. We determine the morphological and shape characteristics of the fabricated axicons as a function of the lithographic exposure parameters. The optical characteristics of the axicons are measured in terms of the axial and radial intensity profiles. The difference between the modeled and measured results is explained.

The standard assembly technologies for vertical-cavity surface-emitting laser (VCSEL) to fiber coupling systems involve the integration of discrete elements with demanding requirements for alignment effort and time. We present a method for the monolithic integration of diffractive microlenses on the chip level. This process is based on a UV-casting replication technique using ORMOCER® (a registered trademark of Fraunhofer-Gesellschaft) materials [hybrid organic-inorganic polymers (Streppel et al. 2001)] and offers the capability to be extended to a wafer-scale process. A mathematical description for the propagation of the laser modes through the system and the resulting fiber coupling efficiency is presented. We use a model for the source characteristics of the VCSEL based on a step-index fiber model for the simulation of the mode-field propagation. A model for the estimation of the diffraction efficiency of the lens is developed. Finally the simulations are compared with first experimental results of single replicated elements. Experimental coupling efficiencies for a multimode fiber [50/125, numerical aperture (NA) = 0.20] better than 0.7 over the entire operation range of the VCSEL are achieved. Losses below 0.5 dB (10%) are observed within lateral fiber displacement tolerances of ∓10 μm.

The results of fracture testing are usually reported in terms of a measured strength σ_M = sigma;i∓&Deltaσi where σi is the average of the recorded peak stresses at failure, and ;&Deltaσi represents the standard deviation. This "strength" does not provide an objective measure of the intrinsic strength since σ_M depends on the test method and the size of the volume or the surface subjected to tensile stresses. We first clarify issues relating to the application of Weibull's theory of brittle fracture and then make use of the theory to assess the results of equibiaxial flexure testing carried out on selected IR-transmitting materials. Since equibiaxial flexure testing has now been adopted as the preferred method for measuring the strength of optical materials, we describe the failure-probability distribution in terms of a characteristic strength σ_C—i.e., the effective strength of a uniformly stressed 1-cm² area—which enables us to predict the average stress at failure in a concentric ring configuration if the Weibull modulus m is available. A Weibull statistical analysis of flexural strength data thus amounts to obtaining the parameters σ_C and m, which is best done by directly fitting estimated cumulative failure probabilies to the appropriate failure-probability expression derived from Weibull's theory. Ring-on-ring fracture testing performed at four mechanical test facilities on five lots of Raytran^TM materials [AlON, chemically vapor deposited (CVD) diamond, ZnSe, standard ZnS, and multispectral ZnS], on sapphire specimens cut in the a-, the c-, and the r-plane orientations, and on Oxyfluoride^TM glass validates the procedure and demonstrates that, in many instances, the wide divergence of measured strengths can be attributed to the size of the stressed area.

Block truncation coding (BTC) is a well known lossy compression scheme. Due to its low complexity and easy implementation, BTC has gained wide interest in its further development and application for image compression. Based on simple thresholding, BTC retains sharp edges and thus leads to artifacts such as the staircase effect. The second problem encountered in BTC is the splitting of homogeneous regions, which produces false contours. In this work a fuzzy approach of BTC to avoid truncating homogeneous blocks and to preserve smooth edges in two-cluster blocks is proposed. Each image block, viewed as a fuzzy set, is segmented into two clusters using a fuzzy clustering algorithm. The block is then encoded by modified fuzzy weighted means of the two clusters. Initialization strategies of the fuzzy clustering algorithm and a contextual quantization method are proposed. Experimental results show an improvement of visual quality of reconstructed images and peak signal-to-noise ratio when compared to BTC, economical BTC (EBTC), absolute moment BTC (AMBTC), and a minimum mean square error quantizer (MMSEQ). Computation time required by AMBTC, EBTC, and fuzzy BTC methods are reported.

We propose a joint source-channel coding system for fractal image compression. We allocate the available total bit rate between the source code and the channel code using a Lagrange multiplier optimization technique. The principle of the proposed unequal error protection strategy is to partition the information bits into sensitivity classes and to assign one code from a range of error-correcting codes to each sensitivity class in a nearly optimal way. Experimental results show that joint source-channel coding with fractal image compression is feasible, leads to efficient protection strategies, and outperforms previous works in this field that covered only channel coding with a fixed source rate.

Active contour models, otherwise known as snakes, are extensively used in image processing and computer vision applications. However, although the approach is popular for detecting the contours of smooth convex objects, it is much more problematic in handling images containing an object with concave parts or sharp corners, or multiple objects. We further develop our segmented snake approach to contour detection and illustrate its flexibility by showing how it can be adapted to yield a dividing snake algorithm for use in multiple object segmentation. We also introduce a snake relaxation technique that can improve the convergence of the snake contour onto the object boundary.

Some dielectric materials become birefringent when subjected to an external force, such as an electric field or a mechanical force. For more than a decade, our research group has been engaged in developing two-dimensional optical measurement techniques for the dynamic measurement of charge distributions on a dielectric surface using the electro-optic Pockels effect, the dynamic measurement of electrical field distributions in a liquid using the electro-optic Kerr effect, and the measurement of birefringence vector distributions in plastic plates using the photoelastic effect. The system nonuniformity and the system reliability are the inevitable problems in two-dimensional measurement systems, and make two-dimensional measurements quite different from point measurements. The common image-processing techniques that have been specially developed to overcome these problems are analyzed and summarized; detailed mathematical analysis is avoided.

Modern trends in the design of semiconductor lasers are addressed. Nanoscale coherent inclusions of a narrower bandgap semiconductor in a wider gap semiconductor matrix, or quantum dots (QDs), are designed as a new type of active medium for injection lasers. QDs provide a possibility to extend the wavelength range of heterostructure lasers on GaAs substrates to 1.3 μm and beyond and improve their device performance. We realize 330-mW cw 1.3-μm single-mode continuous wave edge emitters and 1.2-mW cw vertical-cavity surface-emitting lasers (VCSELs). Long operation lifetimes and other competitive device parameters are demonstrated. Novel device designs are proposed. In one concept, high-order mode filtering in structures with periodically modulated refractive index containing a "defect" enables realization of stable narrow-beam divergence fundamental mode lasing in both edge emitters and VCSELs. In a different novel design, light propagates at some angle with respect to multilayer interference reflectors (MIRs), and the MIRs and the cavity are calculated for the tilted photon incidence. A "tilted cavity" laser (TCL) gives wavelength-stabilized operation in edge and (or) in surface direction and does not require materials having a high refractive index difference. New generations of semiconductor optical amplifiers, photodetectors, optical fibers, etc. may become a reality.

Three major sources of nonlinear error in the common-path laser heterodyne interferometer (CLHI), which include misorientation of incidence beams relative to the Wollaston prism, elliptical polarization of laser beams, and misorientation of two linear orthogonal polarization beams relative to metal-coated mirrors, are theoretically investigated. The results show that the error from the Wollaston prism is mainly second harmonic, and the error from the elliptical polarization of lasers is first harmonic. We discover that the misorientation of the metal mirror can also generate nonlinear errors, which has rarely been studied before. When the polarization directions of the two linear orthogonal incidence beams are not parallel (or vertical) to the normal plane of the metal mirror, the reflecting beam will become an elliptic polarized beam with orientation of the two ellipses having nonorthogonal and nonequal eccentricity. The frequency mixing may occur when the beam reflecting from the mirror passes through a Wollaston prism. The error caused by the frequency mixing is mainly first harmonic. Additionally, error compensation methods are also discussed to improve the accuracy of the interferometer.

We present an evaluation of the continuous wavelet transform method when it is used to measure the phase distribution encoded by electronic speckle pattern interferometry (ESPI) fringes. The evaluation is performed using computer-simulated fringes, an approach that allows knowing precisely the phase map contained in the pattern. It is shown that only ESPI fringes that verify the stationary phase approximation and its analytic asymptotic limit can be analyzed with the continuous wavelet transform method. The influence of the filtering process to smooth the ESPI fringes and the method used to extend the fringe pattern edges is also analyzed. Finally, additional drawbacks that emerge when this phase evaluation method is applied are discussed.

An optimization algorithm is proposed for choosing the parameters of a Fourier flat-top interleaver. The requirement for identical response in both the bar and cross states requires that the intensity coupling ratio for the first coupler must be 1:1. The second and third couplers will depend on the linewidth of each channel. Calculations for typical situations are carried out and the insertion loss for input spectra of different linewidths are also investigated for an interleaver optimized for a typical application.

We design a laser welding machine for fabricating butterfly (BFY) modules with an optical coupling setup that is well suited for rapid prototyping and small-scale production of InP and GaAs-based waveguide-fed photonic components such as diode lasers, semiconductor amplifiers, optical modulators, and optoelectrical ICs (OEICs). The light of these devices is transferred very efficiently into single-mode fibers using integrated tapered lensed ends. In our setup the mechanical adjustment of the fiber-chip coupling is done semiautomatically by a newly designed coupling machine while laser welding performs the fixing of the fiber-chip connection. We found a yield of 47% within a shift of 1.0 μm at the initial welding spot. With additional weld adjustment, it was possible to extend the yield of 67%. We introduce, to our knowledge for the first time, an adjustable welding technique, which we call "strain-reducing" welding, which has several benefits in comparison to the well-known "laser hammering" technique.

A number of planar silicon light-emitting devices are designed and realized in standard 1.2 and 2-μm complementary metal oxide semiconductor (CMOS) integrated circuitry. The devices yield optical power intensities of up to 0.2 μW/cm2 (up to 0.2 nW per 100 μm²) at operating voltages from 4 to 31 V and at currents of 0.1 to 10 mA, respectively. The devices emit light in a broad spectrum from 450 to 800 nm with characteristic peaks at 500 and 650 nm. The emitted intensity of the devices is three to four orders of magnitude higher than the low-frequency detectability limit of integrated Si optical pn detectors utilizing similar areas on chip as the light sources. Initial investigations indicate that the devices have a very fast inherent modulation bandwidth capability. The devices show potential for on-chip electro-optical communication and chip-to-chip electro-optical communications.

We present a detailed study of the intensity variations of the joint Fourier spectrum of two input objects when one of them is rotated while the other remains fixed by determining the 2-D average contrast of the spectrum as a function of the angle of rotation. The contrast is found to be object dependent, decreases rapidly for small rotations, and has an almost flat response for larger rotations. We explain this behavior in both theoretical and numerical ways for two simple rectangular objects and only numerically using the fast Fourier transform, for binary and grayscale test objects whose analog rotations are acquired with an optomechanical system that uses a high-precision rotatory stage. As an application of the 2-D average contrast analysis a pair of similar cross section images of a biological organ are used, to obtain the lowest rotation angle between them, and as a consequence a quantitative aligning method for fine angular movements is presented that uses a 2f optical-digital coherent processor to obtain the Fourier transform of the input objects. Experimental results are presented.

We present the improvement of color pattern recognition by the optimization of a new parameter, the polychromatic discrimination capability. We use the polychromatic correlation obtained as the addition of the three correlation intensities obtained with the RGB components of the input color scene and the corresponding components of the color target. The polychromatic discrimination capability is defined in this fused correlation plane. We propose a method to optimize this performance parameter, which represents a global criterion for the optimization of the polychromatic correlation. The approach involves the design of regions of support to be used with the matched phase-only filters to optimize the polychromatic discrimination. We present simulation results illustrating the improvements obtained after the optimization process. We also include experimental results obtained with an optical real-time correlator that uses two spatial light modulators. Experimental data confirm the numerical results and show the capacity of the proposed technique to improve color pattern recognition.

Microbent optical fibers are potential candidates for evanescent wave sensing. We investigate the behavior of a permanently microbent fiber optic sensor when it is immersed in an absorbing medium. Two distinct detection schemes, namely, bright-field and dark-field detection configuration, are employed for the measurements. The optical power propagating through the sensor is found to vary in a logarithmic fashion with the concentration of the absorbing species in the surrounding medium. We observe that the sensitivity of the setup is dependent on the bending amplitude and length of the microbend region for the bright-field detection scheme, while it is relatively independent of both for the dark-field detection configuration. This feature can be exploited in compact sensor designs where reduction of the sensing region length is possible without sacrificing sensitivity.

We demonstrate the feasibility of operating 1-D lateral effect photosensitive position sensors by measuring the time of delay between the two terminals. This mode has the advantages of a linear response, avoiding the singularities associated with signal normalization, and thus the possibility of achieving higher SNR over the voltage difference mode of operation. This mode has the potential to simultaneously detect the centroids of several light spots modulated at different frequencies using frequency-division multiplexing techniques.

A study of integration for different sensing functions is described. Three different noncontact measuring functions, including a micrometer, a roughness meter, and a profilometer, are integrated into one new sensor. In this sensor, one common laser source, optical system, and photodetector are shared. Based on the theory of laser scattering and interferometry, this sensor can be used to measure the surface roughness and microprofile with the measurement range of R_a = 0.001 to 3 μm. Based on the theory of triangulation this sensor can be used to measure the microdisplacement with a measurement range of 0.5 to 300 μm. The accuracies of this sensor are similar to each existing single sensor for the same function. A typical application of this sensor is to complete the measurements of the diameter, surface quality, and form error of a cylinder or cannon in operation. This integrated sensor is applicable to design a virtual instrument with multiple functions.

We investigate the influence of motion sensor errors on the calculation of a modulation transfer function (MTF) and its implementation in image restoration. The goal of this research is to describe an automatic system for restoration of pictures blurred by vibration, and to consider its possible disadvantages. We present an analytical approach for estimating the vibration MTF from the measured system MTF according to the frequency response of motion sensors and their noise data. Our method is based on point-spread function verification by the data of motion sensor characteristics. We build an analytical model of the sensor and analyze influences of errors caused by system noise and incorrect axis direction of the restoration device. Some image restoration of degraded images is presented based on improvements using the original wiener filter. Performances of inverse and wiener filter operations are compared, and the dependence of restoration quality on the signal-to-noise ratio and angle between the restoration axis and true vibration direction is considered. We also demonstrate the axis rotation sensitivity of the system, and describe improvements over the initial method as seen from our simulation. The key to the restoration is the determination of the improved modulation transfer function unique to the image vibration and sensor characteristics.

A more accurate methodology to measure the reflectance of optical disks, taking into consideration the scattering losses at the surfaces of the disk, is proposed by the National Institute of Standards and Technology (NIST). The proposed method is a modification of the current standard, as described in Annex D of the Standard ECMA-267, second edition, for 120-mm DVD—Read-Only Disk (Dec. 1999, http:/www.ecma.ch/). The experiment shows that the optical losses could be, in some cases, as much as 2% of the incident light power. By accounting for the losses, as in the proposed method, the accuracy of the measured reflectance is significantly improved. Thus, the primary reference disks produced with this method have a higher accuracy in reflectance measurement for the use of optical disk industry.

An unsupervised change detection problem can be viewed as a classification problem with only two classes corresponding to the change and no-change areas, respectively. Thanks to its simplicity, image differencing is a widely used approach to change detection. It is based on the idea of generating a difference image that represents the modulus of the spectral change vector associated with each pixel in the study area. To separate the "change" and "no-change" classes in the difference image, a simple thresholding-based procedure can be applied. However, the selection of the best threshold value is not a trivial problem. We investigate and compare several simple thresholding methods. The combination of the expectation-maximization algorithm with a thresholding method is also performed for the purpose of achieving a better estimate of the optimal threshold value. As an experimental investigation, a study area damaged by a forest fire is considered. Two Landsat TM images of the area acquired before and after the event are utilized to detect the burnt zones and to assess and compare the mentioned unsupervised change-detection methods.

The focus of this work is optimizing recognition models for synthetic aperture radar (SAR) signatures of vehicles to improve the performance of a recognition algorithm under the extended operating conditions of target articulation, occlusion, and configuration variants. The recognition models are based on quasi-invariant local features, scattering center locations, and magnitudes. The approach determines the similarities and differences among the various vehicle models. Methods to penalize similar features or reward dissimilar features are used to increase the distinguishability of the recognition model instances. Extensive experimental recognition results are presented in terms of confusion matrices and receiver operating characteristic (ROC) curves to show the improvements in recognition performance for real SAR signatures of vehicle targets with articulation, configuration variants, and occlusion.

We propose analysis and algorithm implementation for a novel harmonic wavelet joint transform correlator (HWJTC) for image noise reduction and automatic target recognition (ATR) applications. We investigate useful properties of the HWJTC and apply our novel correlator in ATR for both noiseless and noisy synthetic aperture radar (SAR) images. Performance is demonstrated in reducing both simulated speckle noise and realistic clutter noise for SAR images. We also compare its performance with that of a well-established image denoising technique such as wavelet packet (WP) processing. The test images used in our simulation are obtained from a CAD simulation model such as Xpatch and Defense Advanced Research Projects Agency (DARPA) Moving and Stationary Target Recognition (MSTAR) program's synthetic aperture radar (SAR) image databases respectively. Simulation results show improved noise tolerance for HWJTC.

Based on the metals property of modifying the phase of an incident beam, a simple method for identifying the fast axis of a wave plate independent of a standard retarder is demonstrated.

The optogalvanic effect (OGE) of CO₂ waveguide lasers and CO2 lasers stabilized by the OGE and tuned by the compound cavity are investigated. The laser operates at 60 emission lines and the maximum single line power is 8 W. The stability of the laser in long-term operation is better than 0.8%.