The past several years have, in a sense, been "shakedown" years for holography. Very little in the way of original concepts have been introduced; rather, the many ideas proposed in previous years have been developed. In the process, many have been discarded. On the whole, holography has shown a rather remarkable viability; it now appears certain that holography will have a secure, although perhaps modest, niche in our technology. Additionally, it must be said that practical successes in holography have thus far not come easily; and most of the major commercial prospects require long-range development. Several of the most promising ones are included in our survey.
The paramount importance of the hologram recording process has been recognized since the pioneering work of Gabor (Ref. 1). It is evident that information lost in the recording step of holography cannot be recovered subsequently. Therefore, it is essential to carry out this step in an optimum manner, retaining as much of the original object information as possible. Since its critical na-ture was widely appreciated, it is not surprising that the effects of this recording process have been the subject of well over one hundred published papers, innumerable talks, and much unreported research. As a result of this substantial effort, many aspects of hologram recording materials and practices are now well understood. However, there remain numerous questions, a few of which will become evident in the course of this paper.
When you make an ordinary hologram the wave propagation takes place in exact agree-ment with Maxwell's equations, even if you don't understand Maxwell. But if you simulate wave propagation in the computer you must understand exactly what you are doing since the computer cannot help you as nature does in an actual experiment. An unprogrammed computer is absolutely dumb, nature is not. Thus, producing computer holograms is an educational exercise. You gain intuitive insight into the fabulous sampling theorem. And you can also study systematically why it is sometimes good to use a diffuser in contact with the object. In other words, computer holography lets you explore by simulation. This subject will come up 'again at the end of this paper.
Optical memories based on holo-graphic storage have the potential of large volume, rapid access and relative-ly low cost. Estimates of up to 108 bits of storage, random access times of a few microseconds and per bit costs of a few hundredths of a cent are reasonable. Yet to our knowledge such a compelling combination of cost and performance does not exist in a marketable system. What then is the problem, or problems? In this talk we will explore both the promise and several of the problems encountered in the development of a read only holographic optical memory.
The accuracy with which the spatial senses and the brain mechanism associated with them succeed in providing a veridical account of the environment is such that the extent of agreement between the physical world and our experience of it is seldom brought into question. The eye senses information carried by electromagnetic waves, the ear as sound receptor serves to pick up information propagated from distant sources on mechanical waves. In certain cases, information may be transmitted by tactile stimuli. Accordingly, the sensory receptors embodied within the retina, the basilar membrane and tactile end-organs are actuated by stimuli of quite different kinds with widely differing ranges of energy sensitivity. Nevertheless, apparent congruence between the various stimuli is achieved as a result of a series of extremely complex information processing mechanisms performed simultaneously by the peripheral and central nervous system.
The holographic process has been used for many scientific applications and is expected to be extensively applied in modern optical memories, computer applications, and optical information processings, as well as coherent optical communications. In order to efficiently utilize all of these applications it is extremely important to effectively use the holographic channel. Thus, the optimum technique for the holographic process will be given in this paper.
Since the early use of a pulsed ruby laser in holography, significant technical advances have been made recently to increase the coherence length with temperature -controlled multi-etalons and to improve the beam uniformity by growing better ruby crystals. (Ref. 1,2) The pulsewidth obtainable has been reduced to 2.5 nanoseconds without the loss of coherence properties, thus providing shorter holographic exposure times. The ruby laser can be readily multipulsed with pulse separations down to two microseconds. Larger diameter ruby crystals now available have made it possible to obtain higher energy outputs without reaching the damage threshold. Environmental "0" ring sealed enclosure designs have freed today's pulsed ruby laser from the laboratory and have introduced it into the severe industrial environment.
A disadvantage of optical holog raphy has been that the sensitivity to movement and the coherence requirements for hologram recording ordinarily lead to rather large and massive systems and/or lasers. In this paper we describe a portable hologram micro scope which uses a small, reliable pulsed ruby laser light source to overcome this disadvantage. This instrument was designed to record images of rough, diffusely reflecting surfaces with a resolution of 5 pm or better.
Photopolymer materials have been studied for several years at Hughes Research Laboratories because of their potential application to coherent image recording (Ref. 1-5). These materials are particularly attractive for application to holographic interferometry because they are self-developing and can be optically fixed. Thus, a photopolymer hologram may be reconstructed immediately after exposure without chemical processing or moving the hologram. This paper will briefly review the characteristics of the Hughes photopolymers, discuss several fixing techniques, and present some preliminary results from signal-to-noise studies. Recent experiments with dry photopolymers will be briefly mentioned.
The invention of the kinoform (Ref. 1) as a form of on-axis computer-generated holo-gram has stirred considerable interest. The basic observation was that a photographic plate of optical density D(u,v) could be bleached to give an amplitude transmission of eiD(u,v); and therefore, if we contrived to achieve D(u,v) = 0(u,v), we could gene-rate the wavefront ei0(u,v). For most interesting wavefronts there is no way of generating D(u,v) optically, so it is usually recorded by a computer-plotter arrangement.
Coherent optical systems have the capability for performing two-dimensional data processing operations at high speeds. Applications of these systems have been made to pattern recognition, character recognition, spectral analysis, and the processing of a wide variety of optical and electrical signals. The full potential of optical processing has, however, not yet been realized. To increase the flexibility and, therefore, the applicability of optical process we need (1) an input interface device which converts electrical signals or incoherent optical signals into coherent optical signals, (2) an in situ recording material, which is reversable, for constructing spatial filters in real-time, and (3) an interface device which converts the optical output of the processor into an elec-trical signal.
Greatly sharpened images may be extracted from photographs which have been blurred either by accident or deliberately (for instance when coded in view of special image processing applications). In simple words, it has recently become truly possible to turn a bad photograph into a good image in a great number of situations, notably those of interest in space imaging. Examples in the first category (accidentally blurred) include photographs blurred by motion, imperfect focus, instrumental defects and by atmospheric turbulence, among other causes. Examples of the second category (deliberately coded) include the synthesis of multiple-pinhole camera X-ray photos, for the purpose of S/N increase. The results obtained illustrate new extensions of the basic "holographic Fourier-transform division method" first described by G. W. Stroke and R. H. Zech in 1967, for the first category, and of the method described by G. W. Stroke in 1968, for the second. A review of image deblurring principles and of methods used will be given.
The basic concept of acoustical holography is to form (and if necessary record) a hologram using coherent sound waves and to reconstruct the hologram using coherent light. There are many variations on this concept including reconstructing the hologram with a computer and Bragg diffraction imaging. During the 5 years since the first publication on acoustical holography a wide variety of applications and methods have developed. After a brief statement of some of the theory influencing the general principles, this paper discusses some of the methods which have recently developed as useful methods in practical situations. Among the methods discussed are the liquid surface levitation method and Bragg imaging method applied to nondestructive testing and biomedical imaging; including array synthesis and scanning being applied to the underwater and seismic areas.
During the past three years several methods have been found for extracting useful information from laser speckle patterns, either visually or by photographic or photoelectric correlation techniques. Speckle interferometers can be used to detect movement or vibration, or to measure inplane strain of a loaded specimen. It is possible that they can be used for desensitised comparison of a succession of diffusely reflecting components against a master shape. As an example of a noninterferometric speckle device, a visual surface depth probe will be considered.
The technique of pulsed ruby laser holo-graphy has been adapted to record microscopic particles at large distances. The developed hologram reconstructs three-dimensional dark field records of scattered light images frozen during the ~50 x 10-9 second ruby laser pulse. Reconstructions may be analyzed to determine particle density. This is accomplished by measuring variations in absolute intensity of the reconstructed images and relating particle density to image intensity. The ability to determine particle density from absolute light intensity information contained in the holographic image was made possible with an improvement: the holocamera utilized two reference beams instead of one, each at a different angle relative to the scene beam. The second reference beam permits one to make absolute intensity measurements and thus direct comparisons between successive holograms. Separate tests have, in addition, indicated the extreme sensitivity of the three-beam holographic technique. Scene beam signals 10-° less intense than the second reference beam have been recorded, reconstructed and measured.
Holographic interferometry has been known since the original work by Brooks, Hef linger, and Wuerker (Ref.1); Powell and Stetson (Ref. 2); and Haines and Hildebrand (Ref. 3) in around 1965. Numerous general investigations using holographic interferometry have been performed since then. My purpose here is to bridge the gap between the many fine original investigations and the wide audience of possible users of holographic interferometry who would like to know how these techniques can help them with their particular problems. I will show what range and accuracy to expect for the two classes of motion, deformation, and translation, which are most amenable to measurement by holographic interferometry, and "how to get the numbers" out of the interferograms.
At the present time, the most common method for the real-time visualization of steady state mechanical vibrations by holographic interferometry is that of exactly superimposing the reconstructed image on the vibrating object and observing the resulting fringe patterns under stroboscopic illumination. By introducing an initial family of interference fringes by th static rotation of the object before vibration begins, information is available which is not possible from the above method.
A self referencing holographic system capable of compensating for the effects of a fluctuating random medium between object and detector is described. Due to the manner in which the reference beam is derived, the resultant imagery is that which would appear in the absence of the fluctuating medium. Experimental dependence of the resolution on system parameters for the cases of air and water turbulence are presented.
Camera motion, atmospheric turbulence, lens aberrations, image blurring, and diffraction limiting all degrade the quality of photographically recorded imagery. The photographic emulsion further distorts the image by introducing film-grain noise and attenuating high spatial frequencies. Although the objective or mathematical quality or the image becomes fixed on exposure, one can improve its subjective or visual quality by optical spatial filtering or electronic digital processing. As noted below, the success or failure of such techniques depends to a large degree on receiving sufficient exposure from the object scene.
The design of diagnostic apparatus for our large plasma physics experiments involves some unpleasant constraints. In addition, to seeking accurate quantitative results, one seeks to design reliable and flexible apparatus Which can perform in the hostile environment of a megajoule capacitor bank. discharge. The holographic interferometer described below has successfully met these requirements. Except for the high quality lasers the system consists of ordinary optics and components mounted on small tables or exist.- ing framework- We record interferograms on essentially every plasma, shot and operate for weeks without realigning. There is no doubt that holography provides much greater ease and flexibility than previous Nach-Zehnder interferometry.
The sensitivity of holographic techniques to motion in the recording system is often considered to be a disadvantage of holography. Motion in the apparatus, or in the object being recorded, causes phase changes at the recording plane during the period of exposure.1,2 These phase changes cause a reduction in the contrast of the recorded fringes and thus, in turn, a reduction in image brightness. Ordinarily, holographers take great pains to minimize motion in the system so that motion does not darken or eliminate the desired holographic image. However, this sensitivity to motion becomes an advantage in studying certain kinds of motion, for example motion within biological specimens.3 If the motion is not spatially uniform within the specimen, the effect of motion on image brightness is equivalently nonuniform. Thus the hologram provides a permanent record of the motion in a living, changing specimen. The specimen, which may be continually changing, is recorded as it exists during a finite interval on time.
This paper discusses the quality of the imaging of depth information in a reconstruction from a hologram made through a microscope. One of the properties of holography which was at first the most difficult to explain to people new to holography was depth information. Little note has been paid to this property even though its existence is a necessity for most of the useful applications of holography.
Neodymium oxide doped laser glass, with compositions commonly used in our laboratory, are sometimes observed to fracture in use. Similar fractures can be produced by illuminating unpumped glass with a laser pulse with sufficient energy and power densities. This is commonly done to establish the durability of laser glass before fabricating finished laser rods from it. (Ref. 1)
Holographic interferometry has been the most immediate, significant application of laser holography. For the first time, optical interferometric measurements can be made on non-optical surfaces; such as castings, pipes, panels, composites, etc., as well as through diffusely illuminated transparent scenes. Like classical interferometry, holographic interferometric measurements are sensitive to path changes down to one tenth of a wavelength of light (0.07 micron or 3 microinches). The technique has made available a completely new method of non-destructive testing: automobile tire testing, testing of adhesive bonds, pressure vessels, microwave antenna testing, the determination of resonant frequencies of mechanical structures, transient deformation of surfaces, and the study of aerodynamic phenomena.
A great deal of research has been reported in the past five years on the applications of holographic interferometry to the nondestructive testing of materials and products. The majority of this research has been concerned with measuring either the object's surface displacements or its internal refractive index changes (or both). Relatively few researchers have pursued the fact that, in addition to measuring such changes in the object, holographic interferom-etry can be used to produce a contour map of the object's shape. The lack of interest shown to holographic contouring is easily explained by the relative lack of practical applications.