Holograms existed in the nineteenth century, though they were not so called and were not very exciting or newsworthy. Going back even farther into the seventeenth and eighteenth centuries we find the diffraction of light being studied with keen interest by Grimaldi, Newton, Hooke, Delisle, Young, Fresnel and numerous others. All could have had holograms if they had had photography. All looked at the diffraction patterns but had no method of recording them except by drawings.
A survey is given of some of the basic hologram types: included are the in-line, off-axis, coded object beam, coded reference beam, incoherent light, and volume types. The salient properties of each are described.
The Fourier-transform relation between the spatial electric field-vector distribu-tion in a quasi-spherical unfocused wavefront, and the focused field near its center, was recognized at least as early as the begin-ning 1900's by A.A.Michelson (see e.g. refer-ences in G.W.Stroke,"Diffraction Gratings" pp. 426-754 in Vol. 29, Handbuch der Physik, Springer Verlag, Heidelberg 1967), and con-temporary operational image-formation rela-tions were used by him in his famous text "Studies in Optics" before 1927! Recently revived, thanks in particular to the work of Duffieux (1946), D.Gabor (1949), Peter Elias, D.Grey and D.Z.Robinson(1952),AiMarechal, E. L.O'Neill and others, Fourier-transform des-criptions of optical image formation and processing have become particularly powerful with their introduction into holography [Gabor(1949),Cutrona, Leith, Palermo and Porcello,(1960), B.J.Thompson and others(1964), etc.] especially since the introduction of the "lensless Fourier-transform hologram" by Stroke(1964,1965), who showed that such a hologram, formed by interference of the field scattered by the "object" and a reference field originating from a reference "point" situated near the object, had an intrinsic resolution gain which may exceed by more than three orders of magnitude the film-limited resolution capability of single-side band, "off-axis" Fresnel holograms. Among the several extensions and applications of the "lensless Fourier-transform hologram "arran-gement" is that recently exploited by J.Goodman for remote object holography,where the traversal by the point-source reference beam of an air path similar to that of the object beam permits to surmount atmospheric turbulence and path-difference variation effects. In addition to these and other 3-D imaging applications, the author (in part with D. Gabor) has proposed and demonstrated several operational image-processing and computing applications, among which the multiply-expos-ed holograms used for image-synthesis. interferometry and computer-generated holograms, and, most recently, and new type of "holographic Fourier-tranform division" [Stroke and Zech, Physics Lett. 25A, 89(1967)] which has now permitted to restore into sharp focus images from ordinary blurred photographs of three-dimensional objects, photographed in ordinary white light [Stroke et. ali., Physics Lett. 25A, 443(25 March 1968).
Fraunhofer holograms are defined as the interference pattern between a Fraunhofer diffraction pattern and a coherent background. The virtual image under these circumstances is located at an infinite distance from the real image and hence the deterioration of the image is minimized. This work started in 1963 and developed rapidly to field applications of pulsed laser holography for a range of applications involving the determination of particle size dis-tribution in the range from a few microns and upwards and detection of even smaller particles. The principles and application of Fraunhofer holography will be described by reference to various examples.
Two and three primary colors derived from an He-Ne gas laser and an Argon gas laser can be employed in recording and reconstructing holo-grams. For the tri-color case, it is possible to reconstruct a three-dimensional multi-color image which possesses almost all the natural hues of the original object. Each wavelength generates an independent fringe system that is recorded on a photographic film or plate. Due to the interrelationship between the three fringe systems during reconstruction, six ghost images are generated which are deleterious to the reconstructed image. Several techniques to control the ghost images were developed, some of which will be discussed. Emphasis will be placed on the technique which uses the three-dimensional properties of the recording medium. In the recon-struction, each fringe system diffracts light in a manner satisfying the Bragg relation for a particular reconstructing wavelength. If the reconstruction wavelengths are the same as the original wavelengths used to record the fringe system, the result is a multi-color reconstruction possessing few or no ghost images.
The ideas of holography, or wavefront reconstruction, preceded by some thirteen years the development of the laser. It is thus clear that lasers are not necessary for holography, and we should discuss why they are even helpful. Because holography is basically an interferometric process, the need for coherent radiation is clear; the laser represents a source of coherent radiation. Let us begin with a brief discussion of coherence in general, and the coherence properties of gas lasers in particular.
The formation of any hologram involves the exposure of a light-sensitive material to the holographic pattern resulting from the interference of object and reference beams. This recording process and the subsequent reconstruction of the resulting hologram are the essential steps of holography. They were considered carefully by Gabor (Ref. 1) and have since been examined theoretically and experimentally by many authors. It is the purpose of this paper to present a broad survey of the effects of hologram recording media, and to summarize in compact form the potential and actually observed properties of the various types of holograms resulting from the use of the different recording media. We shall also summarize the effects upon the reconstructed image of the detailed characteristics of some particular recording media. We shall concentrate on results, with few derivations of mathematical formulae or descriptions of experimental procedure.
In the following, a description will be given of the existing field of microscopy. A survey will be made to determine what voids exist and where holographic microscopy can make an important contribution.
Holography has two facets which make it a revolutionary scientific instrumenta-tion technique. The first capability is that holography can be used to circumvent photography's depth-of-focus problem for small objects in motion. The second is that measurements equal, equivalent, or beyond those of an interferometer can be made with the technique. The latter can be done in terms of two holograms recorded on the same photographic plate (double-exposed holographic interferometry),lin terms of holograms on separate plates, or in terms of comparing the reconstruction from a hologram against the actual scene wavefront at a later time (stored beam holographic inter-ferometry). 2 In each case, the difference in phase between the two light wave patterns produce fringes, with each fringe tracing the loci of constant phase or equal optical path length. Neighboring fringes are normally separated by an optical path length of one full wavelength:3,4,5 The new ability to make optical interferometric measurements on and in hitherto impossible situations is an extremely valuable contribution. Of particular value are measurements on diffuse surfaces, which are impossible with classical interferometry.
A double pulse technique has been used to record holographic interferograms of a laser created plasma. The technique made it possible to obtain a sequence of interferograms corresponding to different times relative to the laser gas interaction.
Microscopes are a mainstay of biological research. Yet, in many ways we are more frustrated by their limitations than satisfied with their capabilities. These limitations become particularly acute when we wish to study dynamic biological events which are random in nature and located at unpredictable sites. Research on the capillary blood circulation is particularly limited by the small depth of field and field size associated with most high resolution optical systems.
In conventional holography the object must be held stationary to within a fraction of a wavelength during the exposure and hence only a very restricted class of objects can be recorded. however, for applications such as television and portraiture where reconstructions need not be faithful to wavelength accuracies a series of conventional pictures can be combined holographically to produce adequate three dimensional reconstructions.
If a diffuse object moves during the formation of a hologram, the scattered light will be doppler shifted in frequency with the amount of shift dependent on the direction in which the light is scattered. When a portion of this scattered light is interfered with a fixed frequency reference beam, the resulting interference pattern will in general not be stationary. Since the hologram exposure is an integrating process over a time large compared to the fluctuations of the interference pattern (for any significant notion of the object), the recorded fringe contrast necessary for bright reconstruction is not obtained. However, if the light scattered by the moving object can be "sorted" by frequency such that all the light striking any point of the hologram is of a single frequency, and if a reference beam of that same frequency is provided at that point, then the interference pattern will be stationary and high contrast fringes will be recorded. If this condition is met at all points of the hologram, the fringes for any point of the object will be maximun (as good as for no object motion and the usual uniform frequency reference beam) over the entire hologram. It can be shown that such a hologram will reconstruct the point with no blurring (motion is "stopped") and since the entire hologram contributes to the reconstruction, the brightness of the reconstructed point will be the same as if no motion occured. (Ref. 1)
The beam of light emitted by a conven-tional pulse laser is a summation of a number of transverse and axial resonance frequency modes. This type of laser is not suitable for making holograms except with a very specific arrangement of the laser light, object, and photographic plate (Ref. 1-3). In this arrangement, the object is illuminated from behind. Thus, an observer looking through the hologram plate at the reconstructed image sees only the outline or shadow of the object. Obviously, this technique is inadequate when the observer desires to discern surface detail on the holographed object.
Considerable research has been directed toward applying hologram techniques to the recording of dynamic events. Single beam Fraunhofer holograms have been successfully used to study dynamic aerosols (Ref. 1-3). Two beam interferometer experiments have re-corded, shock waves, bullets and other moving objects, (Ref. 4, 5). This paper describes a hologram technique and instrument designed to record a large size range of moving objects over a deep sample volume. Several restrictions guided the approach taken. First, a Q-switched ruby laser was required to stop the object motion; this implies that coherence requirements be minimized. A simple optical system with as few elements as possible was desired so that the instrument could be readily engineered and operated. Photographic film, not plate, was required so that holograms could be recorded in rapid succession. Since film in general is of lower resolution than plate resolution, requirements on the film must be kept as low as possible. A large format was necessary to achieve a large sample volume; large objects as much as a centimeter in diameter at the near end of the volume and 100 m objects at the far end must be recorded simultaneously.
The visualization of sound fields has recently attracted the attention of scientists and engineers of a wide range of disciplines. Consequently, a brief introduction to the physical principles and practical methods involved in light and sound holography will serve to point out the similarities to and diversities from each other. The two proven methods for forming acoustical holograms, the "free surface" method and the "scanning" method will be discussed in detail. Other less perfected methods will also be included in the discussion. Most of the latter methods require rather high acoustical intensities, and this will lead to a number of non-linear effects, which result finally in phase distortion. Various applications of acoustic holography are enumerated.
In this paper we are concerned with the application of holographic techniques to optical systems, and in particular to the problem of imagery from a distance. There are of course many cases of practical interest in which an object is located at some distance from the observer, and for one reason or another the observer is unable to closely approach that object for a detailed inspection. One possible recourse is the use of a suitable optical system to form a high-resolution image of the object from a distance.
This paper describes a thermally compensated, servo-stabilized, single frequency argon-ion laser. Operating in a single transverse and single axial mode, the laser can supply i350 millixatts of optical power at 5145Å. or 4880Å with coherence lengths of 1000 meters. This combination of high power and high coherence makes it particularly useful for deep-field holography. Using this laser, the requirement for approximately equal path lengths for the reference beam and image beam is removed, giving the holographer a new degree of freedom.
Hologram microscopy at the present time is not only faced with serious problems, but also with unique and exciting prospects for the future. Its principal area of difficulty, well recognized by researchers in the field, consists in the diffraction noise generated by edges, out of focus details in the specimen, lens imperfections, dust particles in the system, and "speckle noise", in the granularity associated with stationary interference patterns in space whenever coherent light is trans-mitted or reflected in diffuse fashion.
Scaling of holographic stereograms in the ratio did' can be accomplished by using a spacing, d, between component photographs and a different spacing, d', between corresponding holograms. Scaling down of scenic holograms is desirable in order to increase visual impression of 3 dimensionality. Up-scaling of stereograms is called for to decrease the perspective of very small objects. Our analysis of resolution and depth of field shows the projector lens diameter to be the crucial factor in down-scaled stereograms, while the camera aperture is crucial in the up-scaled case. In did' scaling of a scene which extends from s1 tow , the optimum projector aperture is given by PA51/(d/e)VII; the component photographs, should be taken with an aperture of Pits10, which may be altered by the root of the scale factor, by [d/d']2,without seriously degrading the scaled-down stereogram's resolution. In the text numerical examples are given to illustrate the wide applicability of ordinary photographic apertures in making these down-scaled stereograms. Both linear and two-dimensional arrays of scaled-down scenic holograms have been made and these are described.
The theory of particle size assessment via Fraunhofer holograph-ic techniques has been developed by several researchers, (Refs. 1-4). These methods allow determination of size and location within a sample volume of opaque particles ranging from four to several hundred microns in diameter. This paper will present a method of applying these methods to the determination of three-dimensional flow-field velocities.
Although holographic image reconstruction has been applied prin-cipally at optical wavelengths, such reconstruction is also of in-terest in the radar microwave regime, in which recognition of an object from the characteristics of scattered radiation is a classical problem. The application of holography to centimeter wavelength radiation directs attention to certain problem areas that need not be of such importance in the optical case. In particular, the large aperture needed to obtain reasonable resolution at centimeter wave-lengths requires that the hologram function be sampled at discrete points rather than recorded on a continuum. Further, the thermal noise introduced by the microwave receivers used to sample the hologram introduces a noise background in the reconstructed image. The use of sampling requirements determined, and calculations have been made of the effect of receiver noise in producing image background noise.
Q. By what means can the laser speckle that limits hologram resolution be reduced? Dr. Leith: Speckle reduction can be accomplished by a number of methods. Introducing a phase diffuser such as a diffraction grating in a plane near the object can pro-duce a number of discrete planes in object space free from speckle. Dr. Thompson: However, the grain size and resolution of the hologram is limited by the grain size of the diffuser.