Having known for thousands of years how to carry water through a channel or conduit, and for over 100 years how to carry electrical current through a cable, man's discovery of a way to pipe light took a remarkably long time to develop. It was in 1951 that the first true optical fiber was described and not until 1956 that there was any commercial activity to speak of.

A sudden increase in the availability of low cost fiber optic tapers has given rise not only to many new and interesting applications, but also to a myriad of measurement and specification difficulties. As with all fiber optic components, the method of specifying and measuring the optical parameters has not been easy. These difficulties, however, are compounded in the case of fiber optic tapers. This paper will endeavor to explain why a variation in fiber diameter along the length of the taper greatly complicates the measurement gathering process for distortion, transmittance and resolution. In addition, measurement techniques based on the fiber diameter variation will be explained.

Through the improvement of quality and manufacturing process of glass device named "SELF001W, which was introduced at SPIE of January, 1970, the several kinds of new SELFOOV lenses, in which SELFOCO lenses having different view angles, SELFOd lens of large diameter, SELFOOt4 lens with low chromatic abberation and high quality SELF0a) lens array are included, have been developed. We also succeeded in developing fused fiber plate and flexible image bundle of high quality and low cost by improving the manufacturing process. At the same time, we developed several types of unique endoscopes whose image transmitting devices were composed of the optical glass devices mentioned above. In those unique endoscopes, there are an arthroscope having the thinnest inserting needle in the world and flexible endoscopes for industrial and medical purposes of which costs are very low in spite of their high performances.

In recent times it became obvions that there is a strong interest in using fiber optics for the solution of problems in research and industry in which the use of ultraviolet light is required. As in the case of regular fiber optics the UV-fibers are of a primary interest if their flexibility and the opportunity of guiding light to inaccessible areas appear to be definitely advantageous to classical optical solutions. However, the real issue is concerned with the fact that particular problems cannot be solved, unless the spectral range of the used light is extended to the UV. There are essentially three types of applications that typically require UV-light:

Fiber optics is known to constitute a highly desirable transmission medium for a flexible, versatile laser retinal photocoagulator system. However, it had not been possible to deliver the high average power and high radi-ance required for photocoagulation through fibers due to their inability to sustain high transmittance under intense laser radiation. The development of a special core glass for fiber optics has now made it possible to reli-ably transmit CW power densities of 5 x 104 W/cm2 and maintain output radiance of 106 W/cm2/sr in the blue-green region of the spectrum. This performance has made it possible to design a compact argon laser retinal photocoagulator system which can be coupled to a direct or indirect ophthalmoscope delivery system, or be provided with a light-weight slit lamp adaptor for use with standard slit lamp instruments. Two incidental advantages of fiber optics transmission are increased uniformity of irradiance over the exposed area in the eye, and a convergence cone angle within the eye which is virtually independent of spot size.

With the introduction of communications theory and allied studies of random data and the engineers' increasing understanding of these fields the demand for wideband multichannel correlators is increasing rapidly (Ref. 1). Correlation techniques most commonly used today employ either dig-ital computers, coherent-optical correlators or an application of one of the nonlinear characteristics of semi-conductor devices. At present, however, no wholly satisfactory method of simulataneously correlating a large number of wide-band signals has been found; systems developed for this purpose suffer from one or more draw-backs, viz: phisical instability, slow operation, narrow bandwidth or lack of isolation between channels.

The shape of particles in a powder system, slurry or aerosol affects many of the properties of the particles in these systems, e.g., sintering, heat transfer, settling, in particular the problems associated with inhalation of fibrous dust pose many difficult problems in industrial hygiene. The determination of a shape parameter for a particle however has proved to be a difficult task. The classic studies of shape parameters were carried out by Heywood using large particles which were easy to handle, eg., rock fragments (Ref. 1) Heywood's work was extended by Hausner to metal powders (Ref.2). The ratio of the difference in size of a particle as measured by two different techniques, e.g., microscope measurement and sedimentation measurement has also been used as a shape parameter. Medalia has carried out extensive work on shape characteristics of carbon black flocs but the method has the disadvantage of requiring many computer calculations (Ref. 3) Most of the work on shape characterization of small particles deals with assessing in some way the two dimensional projection of the particle profile. An interesting development has been to generate a wave from a particle profile by rotating a radius vector about some point within a particle profile as illustrated in figure (1). (Ref. 4) The wave is then characteristic of the particle profile and the wave itself may be described by its Fourier components. The centre of rotation chosen is the centroid and the disadvantage of this method is the problem of locating the centroid and performing the measurements of the lengths of the radius vector. Kaye has suggested a simple way to overcome both of these problems by plotting angle of rotation against Feret's diameter. (Ref. 5) This parameter however has to be interpreted with care for re-entrant particle profiles.

All materials interact with light in some manner, but few record a view-able image. Very few materials record an erasable, viewable image. Thus, the invention by W. H. Armistead and S. D. Stookey (Ref. 1) of photochromic glass offers the possibility of devices using image write-read-erase capability. Photochromic glass is formed by precipi-tation of silver halide crystals of 50 to 100 A diameter spaced 500 to 1000 A apart in a glass matrix as described by G. P. Smith (Ref. 2). The possibility of fabricating large pieces and of using the many glass manipulation pro-esses gives photochromic glass advantages over crystalline materials for practical applications. G. K. Megla (Ref. 3) has described the optical properties and an appropriate system of units to characterize the features of photochromic glass. He has also described experiments demonstrating that photochromic glass is free of fatigue, at least to 300,000 cycles of operation. It may be added that no evidence of fatigue in photochromic glass has been encountered in our work. G. K. Megla and D. R. Steinberg (Ref. 4) have described a time-sharing computer terminal using the image storage CRT incorporating the photochromic fiber optic plate discussed here. This use of photochromic glass places many requirements on the material. Considerable effort has been expended both to configure the system to minimize material requirements and to measure material performance ance relative to system operation.

Derek Morgan of Rank Precision Industries presented a paper to the SPIE meeting last year on Traffic Engineering. This paper reports on advances in the technology since that time. Review of Previous State of the Art. A. We use glass fiber optic bundles that will withstand the high temp-eratures of high intensity light sources. B. Fiber optic bundles are produced in the form of a harness; 1 input, up to 250 output tails. C. Arranging the output tails in a pegboard allows fixed messages to be brightly displayed as dots of light.

As mentioned in other NVI, papers given this session, faceplate development was con-cluded quite a while ago and recent measure-ments on current faceplates are scarce. It was also mentioned that a problem area has recently been uncovered concerning the input or cathode faceplate of an inverter or wafer tube. This discussion will deal with the nature and extent of this problem including recent test data bearing on the problem. To adequately discuss this, a review of the function of a faceplate would be in order. Figure 1 shows the objective lens, faceplate, and photocathode as employed in Night Vision Systems. The fiber optic faceplate's primary function, and the one in which we are interested, is to transfer an image formed on the input face of the plate through the thickness of the plate to the photocathode, which is deposited on the output face of the faceplate. If the transfer process degrades the image, the photocathode starts off with a blurred and low contrast image. To this are added or multiplied other system degradation factors and the final image is then definitely not state-of-the-art. To achieve a good image transfer through the fiber optics plate, stray light, which spills out of each fiber or which initially fell between the fiber core areas, must be taken care of. To do this, the industry developed various methods of placing black glass around the fibers or between the fibers. This black glass was to absorb any stray light which was wandering about outside the fiber walls, which lead to the name EMA, or "extra mural absorption." This black glass, EMA, or interstitial opaque coating as it is sometimes called, then is somehow interspersed throughout the plate and outside of the fiber cladding to absorb much of the stray light formed by imprefections which serve as scatter centers in the fibers, by light entering at too great an angle for the total internal reflection necessary for propagation down the fiber, or by the incident light falling outside of the fiber core cross section.

The slow-scan vidicon television camera is one of the principal instruments used by NASA in the unmanned exploration of the planets. Cameras of this type have already mapped the surface of Mars and will soon be sent to photograph Venus and Mercury. The unaided vidicon, however, lacks the sensitivity required for further improving the surface resolution at these inner planets, or for maintaining the present level of performance when missions are flown to planets further from the sun. Therefore, we have recently considered the use of a first generation image intensifier as a means for increasing the sensitivity of these cameras.

Fiber optic components which are currently being used in Night Vision Systems are discussed in reference to their critical performance parameters and the test methods employed to evaluate the fiber optic components' performances in terms of these parameters. Specifications for the 18, 25, and 40mm fiber optic image inverters are given.

Fiber optic image inverters were developed under the sponsorship of the Night Vision Laboratories of the United States Army Electronics Command for use with second generation proximity focussed "wafer" type intensifier tubes. Figure 1 schematically shows the main components of this type of image tube. These include the input fiber optic window which supports the photo-cath-ode, the microchannel plate which provides the amplification and the image inverter which forms the rear window of the tube. The reason for using the fiber optic inverter is that a proximity focussed tube does not electronically invert the image, and some method of optical erection is needed to compensate for the inversion caused by the objective lens. The alternative to a wafer tube with a fiber optic inverter is an electronically inverting second generation tube, but it is not proposed to compare the two methods in this paper.

A three-dimensional data acquisition and display system using fiber optics is described. The system consists of a range-gated pulsed laser illumination source and a direct view image intensifier or remote view camera tube sensor. Such sensors have been effective in locating, ranging, and observing targets which may be located behind a scattering medium such as clutter, haze, or camouflage. The ability to control the viewing time of the sensor allows the back-scattered light to be gated out, thereby providing enhanced detection and recognition capability. The three-dimensional information is obtained by range-gating the different z-planes in time. The display system consists of fiber optics slugs of varying thicknesses, and each frame in the object plane is projected on a different fiber optics slug. This entire data acquisition and display occurs at a speed faster than the flicker frequency of the eye, thus penetrating a real-time, three-dimensional view of the object. A prototype of the three-dimensional display system is explained and various alternate approaches for data acquisition, data recording, and data display is described.

I want to thank the Society for inviting me to present this material to you. I hope that it will give you some idea of what we can do now, what some of the technical problems are, and perhaps, stimulate your interest and solicit your help.