What I want to cover in this paper is a set of fairly recent concepts in imaging and image processing that appear to offer the ability to achieve high resolution through atmospheric turbulence.

The optical transfer function (OTF) of the atmosphere has been measured over the ocean for four wavelengths from visible to far IR, using laser sources and a slit scanning telescope. The effects of diffraction, finite slit width, and aberrations have been removed by digital Fourier processing. The shapes of the curves of long term average OTF, and the image-centered (tracked) OTF, as well as the magnitude of the wander variance, all agree well with a theoretical model by Fried. Comparisons of the path-integrated values of Cn², obtained from the OTF, with the path-integrated values of Cn obtained from CT², indicate that, for nonuniform Cn2, the weighting of Cn² as a function of position on the path, behaves as predicted. This weighting heavily emphasizes the part of the path nearest the telescope for imaging devices, whereas scintillation emphasizes the path center. The weighting that applies to imaging devices also applies to beam-forming or projection sys-tems, with the heavily emphasized part near the projection optics. Measurement with a scanning telescope thus yields directly the properly weighted value for such systems.

A rapid and accurate approximation to the Lorentz line shape has been developed which significantly reduces the time of line-by-line atmospheric transmittance calculations. Computer arithmetic operations are minimized through the separation of quantities which depend on frequency and those which depend on atmospheric variables. The approximation and the standard methods of calculation were applied to the determination of absorption coefficients at five NIMBUS-6 HIRS center filter frequencies in the 4.3 μm spectral region for a 33 level vertical atmospheric path. Water vapor, carbon dioxide, nitrous oxide, and carbon monoxide were considered, involving an average of 1000 spectral lines per absorber. The approximation method was, in all cases considered, an average of nine times faster than the standard method. Absorption coefficients computed by the approximation method provided transmittance values which agreed with those computed using the standard method to four decimal places regardless of absorber type channel or atmospheric level. For atmospheric transmittance calculations over a band, time savings have been predicted by analyzingithe required number of arithmetic operations involved. For 1000 spectral lines over a band of 25 cm^-1 and a step increment of 0.02 cm^-1 , a reduction in computation time by a factor of at least 4.5 is anticipated.

Power spectra, histograms and cumulants of irradiance fluctuations of collimated laser beams (He-Ne 6328 A) were measured for folded transmissions by a corner-cube reflector and a plane mirror and the direct transmission. Measurements were performed by using an at-mospheric turbulence chamber (one way path length 2.54 m) and also in the outside field (one way path length 274 meter.)

The Air Force Weapons Laboratory (AFWL) has been involved in the study of laser propagation from airborne platforms. A Line Spread Device (LSD) capable of yielding direct measure of a laser beam's Line Spread Function (LSF) was developed and employed in propagation tests conducted in a wind tunnel by AFWL to examine optimal acoustical suppression techniques for laser cavities exposed to simulated aircraft aerodynamic environments. Measurements were made on various aerodynamic fences and cavity air injection techniques that affect the LSF of a propagating laser. Using the quiescent tunnel as a control, the relative effect of each technique on laser beam quality was determined. The optical instrument employed enabled the comparison of relative beam intensity for each fence or mass injection. It was found that fence height had little effect on beam quality but fence porosity had a marked effect, i.e, 58% porosity alleviated cavity resonance and degraded the beam the least. Mass injection had little effect on the beam LSF. The use of a direct LSF measuring device proved to be a viable means of determining aerodynamic "seeing" qualities of flow fields. It could also be applied to static atmospheric "seeing" measurements through various gases and pollutants.

Stability and height variations for C_T² results observed in the surface layer above ocean waves are compared with overland predictions. Evidence of the waves' influence is shown.

A new particle sizing device is described for sizing liquid spray drops and "dry" particulates in situ. A pulsing ultra violet laser of short pulse duration is used to illuminate moving particles. "Stop action" images are formed on a vidicon and presented on a TV monitor. These images are stored in an instant replay recorder and replayed in slow motion for sizing manually or by automatic image analysis. This particle sizing device is now standard equipment at the National Bureau of Standards for sizing liquid spray drops. Excerpts from a number of contracts performed by the author are given via 35mm slides and by replay from the instant replay recorder. The mechanics of droplet formation, ligaments, hollow spheres, evaporation, agglomeration, combustion of liquid fuel droplets, impingement, impaction of insecticides on live insects, hairsprays, inhalation and weight respirable of some typical household aerosol and hand pumped sprays are explained, together with an explanation of the techniques of measuring particle velocity.

The purpose of this work is to determine whether satellite derived cloud imagery can be effectively analyzed by optical means. Fourier transforms of a number of cloud images are obtained optically and digitized for analysis. Scenes are chosen from a 20 March 1975 frontal system across the Midwestern United States. Analysis is limited to three image categories: straight edges (jet stream), uniform cloud cover with globular texture (the frontal band), and broken cover or isolated globular texture (cumulonimbus activity). These are analyzed for time development of features as well as for type distinguishing features. The ROSA system at Engineering Topographic Laboratories is used in this analysis and the imagery is taken from the satellite ground station at the Atmospheric Sciences Laboratory. The results so far indicate that: (1) certain cloud features can be identified but that uniqueness of the identification must be established on a larger statistical sample, (2) the contrast level in "visual" imagery is not ideal and improvements in identification are expected to result from appropriate enhancements which can be applied "on line", (3) the scan line structure of the imagery can be more effectively exploited, and (4) this method of analysis merits a closer look.

A new photomultiplier tube designed for the study of high speed and weak light phenomena is described. The tube has proximity electrode configuration to decrease the electron transit time spread and also incorporates MCP ( microchannel plate ) due to its excellent timing property. Sb-Cs photocathode was made directly upon the input surface of I:CP and photoelectrons emitted from the cathode were led into MCP by the aid of retarding field mesh arranged in front of iiCP. This configuration enabled to make photocathode in proximity type relatively easy. 10. -20. μA/ Lm range of photocathode current was obtained in tubes fabricated by this way. The tube showed that the gain was about 104 at total applied voltage of 1.8 Kv (1.2 Kv for chevron type MCP) and dark current was about 200 pA. Impulse response was measured by using short loser pulse of 26 psec FUHM. Tise time of 0.33 nsec was obtained. This value appeared to be quite reasonable when it was compaired with the value of 0.21 nsec estimated from electron transit time and its spread between photocathode and anode.

A two-beam holographic scheme consisting of a 1:1 relay lens and an integral hologram has been used with a Q-switched ruby laser to record high resolution 3-D images and interferograms of the combustion of small (1.5 x 3 x 6 millimeter; .06 x .12 x .25 inch) rocket propellant samples in a high pressure chamber. The holograms were helium-neon laser reconstructed. Images good to 2 micron resolutions were achieved with collimated illumination. Diffuse illumination holograms gave % 4 micron resolutions when viewed on a moving translucent screen. The reconstructed real images yielded particle size distributions. Double exposure holograms yielded either conventional holographic interferograms (first exposure prior to combustion), differential interferograms (both exposures during combustion), or sequential image holograms of the particles. The latter were improved by constructing a holocamera with two separate reference beams. One reference beam was used for one exposure. The other is used only for the second exposure. Such a hologram reconstructs separate 3-D images. From such a hologram, particle motions can be followed and velocities computed without confusion. Metallized propellant holograms showed time-averaged fringe effects when recorded with conventional 50 nano-second duration laser pulses. The dark fringes in the reconstructions were due to rapid gas expansion. Reducing the laser pulse duration to 10 nanoseconds (by pulse chopping techniques) gave relatively transparent reconstructions. Reflected light holograms (non-lens-assisted) of both the burning surface and the particle clouds were also recorded.

Methods for applying holographic interferometry in the study of turbulence are described. Examples of the effect of various types of turbulent cells on an optical wavefront are presented. Finally, a relatively new technique in which phase control on the two mixed reconstructed waves is discussed. It is concluded that the technique can ultimately provide an extremely high sensitivity for the viewing of phase objects and can provide the basis of automated data reduction based on heterodyne interferometry.

A recently developed particle camera operated on the South Dakota School of Mines T-28 weather research armored aircraft has been used to photograph cloud particles in hail precursor and hail producing clouds for the National Hail Research Experiment. The 35 mm camera, located under the port wing, photographs particles within up to nine liters sample volume per exposure at rates up to about 20 exposures per second. Particles of approximately 50 micrometers diameter and larger are imaged at 0.16 magnification with ice/water distinction possible for particles larger than about 100 micrometers. Design features of the camera include: (1) image motion compensating rotating mirror with speed electronically servoed to true airspeed; (2) ratio control for varying the frame rate at typically between one per four seconds up to 20 per sec at 100 m per sec true air-speed; (3) accumulate feature for multiple exposures of up to four per frame and (4) choice of magnifications of 0.16 or 0.6x as appropriate for the size range of particles being studied. Examples of photographs from the 1976 field season are shown. Techniques for camera calibration and automatic data reduction are briefly discussed.

When a particle dimension is comparable to the laser light wavelength, scattering calculations require the use of Mie or similar infinite series. The terms in the series may be an infinite set of complex determinants. Despite the power of modern computers these equations are difficult to compute with, because of the lack of analytical criteria for terminating the computations. Commonly, the series computations are stopped when the value of the series is changed by an amount less than a predetermined factor as a result of the addition of the computed term. This procedure has led to large errors. These series are easily transformed into the continued fraction form of the Fsad6 approximates. Complex continued fractions are more stable, computationally, than infinite series and frequently converge more rapidly. Further, an anlytical test exists for determining the circular domain within which the value of the continued fraction must lie. This permits an accurate and precise delineation of the maximum residual error. There is a disadvantage in the use of continued fractions. If the order of the creates convergent is not great enough for the desired accuracy, the computation must be repeated. This is not a serious handicap in actual use.

The measurement of atmospheric properties from which visual range and thus "visibility" can be inferred reliably is a difficult task. One approach to instrumental measurement of atmospheric properties that can be interpreted in terms of visual range, with appropriate assumptions, is to use an optical transmissometer. Such an instrument provides a measure of the atmospheric extinction coefficient averaged over the propagation path and bandwidth of the instrument. When visual ranges are to be measured, the instrument baseline must necessarily be a significant fraction of the maximum visual range to be resolved and both micro-scale and macroscale beam effects are evident in the measurements. To realize the desired measurement of transmittance due to extinction by aerosol, it is necessary that the errors in measurement of atmospheric transmittance due to gaseous absorption and the macroscale components of atmospheric transmittance due to stochastic and deterministic refraction be minimized. This paper describes the field testing on an 18 km folded baseline of a long-path laser transmissometer parametrically designed for visual range measurements under good visibility conditions. Design parameters, instrument performance, measurements and data interpretations are presented. It is concluded that with large collecting optics, aperture averaging and adjustment of beam divergence, stochastic and refractive effects can be sufficiently minimized to make the transmissometer a useful instrument for visual range measurements.

Optical techniques continue today as our most extensive methodology in sizing particles. Imaging and light scattering methods are used most extensively. Recent work in light scattering aerosol sizing methods involve Active Scattering with light collection over a 211 steradian solid angle, an off-axis technique with application to in situ measurements at concentrations up to 107cm ^-3, and a single particle multiprobe wherein five separate measurements are performed on a single particle. Imaging systems useful in sizing particles larger than a few microns have also been improved; for instance, an Optical Array Spectrometer has been expanded to provide multilevel isodensitometry with an imaging bandwidth above 1 GHz. The use of an image dissector with multiple apertures is shown to have both light scattering and imaging potential. The evaluation of sampling techniques for special testing include wind tunnel evaluations of aircraft mounted isokinetic sampling probes, ground aspirators for large airborne particles, hot sampling sections for high temperature particle sources, and an in situ particle size spectrometer for in-stack measurements.