John Caulfield as editor of this journal has made a commitment to devote each January's special issue of Optical Engineering to the topic of optical computing. I belieive that this annual issue can become a significant factor in the advancement of the field.
Acousto-optic channelized receivers are modeled in terms of key performance criteria and critical component and system design parameters. Examples of current state-of-the-art receivers are presented that include an integrating channelized radiometer and a miniaturized parallel output channelized receiver.
Efforts to fabricate real-time optical correlators in compact packages are under way. In particular, a brassboard unit mounted inside a 15 X23 X28 cm enclosure featuring laser diode light sources and a liquid crystal incoherent-to-coherent image converter has been built; operational experience with this correlator is discussed. Further miniaturization is possible, and some suggested geometries are examined in detail.
A real-time optical image correlator is described. An acousto-optic device and an array of light-emitting diodes are the electronic-to-optical transducers for the input and reference images, respectively, in this architecture. The two-dimensional correlation is formed by temporal integration on a two-dimensional CCD detector.
Optical computation is usually performed on signals from the electronic domain, but it is especially powerful when applied to signals already in the optical domain. An example is shown here. Spatial filters representing the positive and negative parts of computer-designed linear discriminants operating in the spectral domain can be used to recognize and classify colors, species, and events. A demonstration in which colors are automatically sorted is given.
A novel high speed array processing optical architecture is described. A multichannel acousto-optic binary convolver is architecturally configured as a systolic array processor. The architecture provides a high speed means of matrix/vector multiplications using the digital multiplication via an analog convolution algorithm. This algorithm and a systolic acousto-optic implementation permit the speed of optics to be combined with the accuracy of digital computation.
Presented in this paper is a review of an electro-optical engagement-array architecture capable of performing a variety of matrix operations. The matrix information may be encoded and processed using either conventional optical analog techniques or the twos complement binary representation.
In some important image coding techniques, such as transform coding, an image is first divided into subimages, and then each subimage is coded independently. The segmentation procedure has significant advantages, but when used in a low bit rate scheme, an undesirable side effect can occur. Specifically, when the image is reconstructed at the decoder, a "blocking effect" can develop due to discontinuities between the subimages. Two methods are developed to reduce the blocking effect. The performance of these methods when applied to a discrete cosine transform image coder is discussed.
Aerosol size distributions measured with altitude in eight different maritime stratus regimes are used to determine the liquid water content, total volume scattering coefficient, and angular phase function for a wavelength of 0.525 um. The scattering coefficient and mean square angle of scattering were found to correlate with the liquid water content, which increased with altitude above the cloud base according to a power law relationship. Regression equa-tions relating the scattering coefficient and mean square angle of scattering to altitude within these clouds are then used to express the time broadening of a pulsed optical beam as a function of cloud thickness.
The use of multiple-cavity lasers in holography and photography is described. By splitting up a ruby laser cavity and using part of the crystal for different cavities, a number of interesting possibilities result. Eight basically different configurations are discussed to show how the normally oversized crystal in commercially available lasers can be used to advantage. These result in lasers that become ideal sources for some applications, such as ultrahigh repetition rate, stereo holography, multiple reference wave, simultaneous front and back lighting, multiple framing, and other techniques.
Application of image processing for the visually impaired is discussed. Image degradation in the low vision patient's visual system can be specified as a transfer function obtained by measurements of contrast sensitivity. The effectiveness of adaptive image enhancement for printed pictures is demonstrated using an optically simulated cataractous lens.
The classical spherical aberration structural coefficient for a plano-convex lens is compared to one derived via Fourier optics. The effects of these differences on diffraction are then studied in the maximum Strehl planes.
This paper describes characteristics of visual range attenuation in blowing snow in relation to motorist vision. Visual range V is related to wind speed U according to V = AU-5 with the A coefficient changing in response to snow availability. A lower limit for visual range is described with A =1.1 X108 m6 s-5. An operational monitoring system used on Interstate Highway 80 in Wyoming demonstrates how real-time computer analysis of photometric data can be used to determine the A value, interpret visual range in terms of vehicle operation, and provide automated traffic operations decisions.
We calculate the evanescent electric field and intensity profile of a circular Gaussian light beam when it is focused by a lens, enters a spatially finite prism, and is totally internally reflected at a planar interface of the prism and a dielectric substance. The electric field and intensity at the interface depend on the indices of refraction of the prism, the 1 /e2 width, wavelength, and polarization of the incident beam, the focal length and exit pupil of the focusing lens, the distance from lens to prism, the shape and size of the prism, and the beam's angle of incidence. Our calculations show the evanescent intensity profile to be well approximated by a Gaussian shape for many practical realizations of the lens and prism optical apparatus. The 1 /e2 width of the profile appears to depend primarily on the beam's 1 /e2 width, the ratio of the focal length of the lens to this width, and the incidence angle. The polarization of the evanescent electric field is approximately perpendicular to the incident plane of the beam center for an incident beam polarized in this direction. It is within the incidence plane, but rotates with time and position along the interface, for an incident beam polarized in the incidence plane of the beam center. The penetration depth (into the dielectric) does not have a strong dependence on position within the illuminated area. Experimental measurement of the intensity profile in one optical geometry agrees well with our theoretical predictions.
The variance of a bar pattern image can be used to measure the modulation transfer function (MTF) of an imaging system. This method is often simpler and more practical than conventional methods. The theoretical basis of the method is derived, and verification is provided by computer simulations. The computer simulations also are used to evaluate the accuracy of approximate formulas. The MTF of a computed tomographic (CT) scanner is measured to illustrate the technique.
The operation of an analog image processing technique based on a laser scanner system is described and illustrated with a number of experimental results. Point nonlinear and locally adaptive filtering are performed on-line. These operations use the deflection of a light spot on a mask with a transmit-tance mapping the desired nonlinear input/output relationship.
A proposed optical vector-matrix multiplication scheme encodes the matrix elements as a holographic mask consisting of linear diffraction gratings. The binary, chrome-on-glass masks are fabricated by electron-beam lithography. This approach results in a fairly simple optical system that offers both large numerical range and high accuracy. Simple holographic masks have been fabricated and tested.
The magnification of a single prism or a series of prisms is normally considered in a situation when collimated light is incident, which is equivalent to placing the object at a very large distance from the prism system. This is the usual situation considered and described in most of the literature. In such cases, the magnification is referred to as the anamorphic magnification of the prism system in collimated light. However, when the object is at a finite distance, the apparent magnification is quite different and sometimes is just the reciprocal of what it would be if the object were at infinity. These situations are analyzed for the case of Brewster prisms, and the results can be checked visually very easily in ordinary white-light illumination because of the mutual cancellation of the dispersion of the Brewster prisms.
Low concentration pollutants in the atmosphere can be detected by their infrared absorption spectra. We use a diode laser spectrometer in a dual-beam configuration for this purpose. The laser source is frequency modulated to provide the sensitivity enhancement associated with derivative spectroscopy. One of the laser beams is passed through a reference cell containing the gas to be detected in order to lock the laser frequency to the center of the absorption line. The other beam passes through a White cell with 64 m absorption path length. Sample air is sucked through this cell at a pressure of about 100 mbar. Although the pressure reduction reduces the density of absorbing molecules by a factor of ten, the increase in absorption cross section due to the narrowing linewidth nearly compensates this effect and drastically reduces interference from other gases. The absorption is observed as a modulation of the laser intensity at twice the modulation frequency. The intensity modulation is proportional to the second derivative of the absorption line. The spectrometer was used in a field experiment on board a research vessel in the North Sea for the measurement of HCI in the plumes of incineration ships. A HCI detection sensitivity of 50 ppbii-Lz was achieved.
A variety of optical coatings are discussed in the context of solar energy utilization. Well-known coatings such as transparent conductors (heat mirrors), selective absorbers, and reflective films are surveyed briefly. Emphasis is placed on the materials' limitations and on use of lesser-known optical coatings and materials. Physical and optical properties are detailed for protective antireflection films, cold mirrors, fluorescent concentrator materials, radiative cooling surfaces, and optical switching films including electrochromic, thermochromic, photochromic, and liquid crystal types. For many of these materials, research has only recently been considered, so various design and durability issues need to be addressed.
A new optical method to generate a family of color-coded images using holographic techniques is presented. This method involves the introduction of three primary colors. Thus, this technique overcomes an important limitation of previous holographic techniques. An image hologram of a black-and-white transparency is first registered. This hologram is put in contact with a beam splitter, and the sandwich is illuminated by three partially coherent waves. One wave produces a positive image, a second wave pro-duces an image with reversed contrast, the negative, and the third image corresponds to the product of the positive and the negative. Coding these waves in three primary colors and adding them at the final image plane results in a color-coded image. The principle of the method, theory, and some experimental color results are presented.
On page 610 of the September/ October 1983 issue the authors (Blouke et al.) state that the MTF of the CCD (as a discrete sample imager) ideally "should exhibit a sin(f/fN)/(f/fN) behavior, where f and fN are the spatial frequency of the scene and the spatial Nyquist frequency, respectively." Application of this formula would give values of 84 percent modulation at Nyquist and 45 percent at twice Nyquist (although the latter would be of no real significance). In fact, the relationship (for a fully-active pixel-area device) is a sinc function: