Today there is a recognized, integrated body of knowledge that constitutes optical science and engineering. It is an enabling area of science, technology and engineering that has a very significant impact on today's world. The scope of optical science and engineering will be explored to review how it is organized both from the educational point of view as well as the professional standpoint. The impact of national programs and international structures will be examined with some particular comment on the current initiatives in the United States through the Committee on Optical Sciences and Engineering formed under the auspices of the National Research Council whose report should be forthcoming next year.

The paper begins with the work of various pioneers in the use of the optical spark up to the early 1920s. These include Fox Talbot, Mach, Cranz, Boyes, Worthington and Bull amongst others. Mention is made of some of their ingenious solutions for problems of triggering and the problems to which they applied spark photography. These efforts began with single exposures and gradually turned to various multiple spark systems in order to incorporate a time history into their observations. There follows descriptions of spark source types, their backing circuitry, optical properties, general usage, later advances and applications including their use in Aeroballistic ranges. Finally spark source advantages and disadvantages are compared.

In 1943, there was no camera with the microsecond resolution needed for research in Atomic Bomb development. We had the Mitchell camera (100 fps), the Fastax (10 000), the Marley (100 000), the drum streak (moving slit image) 10^-5 s resolution, and electro-optical shutters for 10^-6 s. Julian Mack invented a rotating-mirror camera for 10^-7 s, which was in use by 1944. Small rotating mirror changes secured a resolution of 10^-8 s. Photography of oscilloscope traces soon recorded 10^-6 resolution, which was later improved to 10^-8 s. Mack also invented two time resolving spectrographs for studying the radiation of the first atomic explosion. Much later, he made a large aperture spectrograph for shock wave spectra. An image dissecting drum camera running at 10⁷ frames per second (fps) was used for studying high velocity jets. Brixner invented a simple streak camera which gave 10^-8 s resolution. Using a moving film camera, an interferometer pressure gauge was developed for measuring shock-front pressures up to 100 000 psi. An existing Bowen 76-lens frame camera was speeded up by our turbine driven mirror to make 1 500 000 fps. Several streak cameras were made with writing arms from 4 1/2 to 40 in. and apertures from f/2.5 to f/20. We made framing cameras with top speeds of 50 000, 1 000 000, 3 500 000, and 14 000 000 fps.

This paper discusses the processes involved in taking a new concept streak camera through the evaluation stages to the proof of performance point and demonstrates sub 400 fS resolution in single shot mode. Many designs of streak tube have been proposed in the period since the arrival of the laser as a fundamental research tool. Each iteration in design results some improvement in one more operational parameters pushing the time resolution and dynamic range barriers forward. The Femto-chron streak tube used in the FS 300 streak camera is the most recent design from a well known group in the field comprising of St Andrews University, Hadland Photonics and Photek Ltd and was part funded by a LINK award from the U.K. Government's Department of Trade and Industry. This design offers an advanced specification capable of both synchroscan and single shot operation and has the potential for operation in the sub 250 fS region. Prototype tubes have already shown sub picosecond performance in both modes of operation.

The multi-image framing camera based on gated MicroChannel Plates (MCPs), used in ICF programs, allows to obtain a sequence of 2D images in the x-ray range with gating times smaller than 100 ps and spatial resolution better than 20 pl/mm. After using such devices from different factories, we decided to build our own tube. Our choice is motivated by the great fragility of the tube, consequent to both bad vacuum during the experiment and moisture due to frequent air launching, and by the mastery of the technology which is necessary to repair rapidly such apparatuses. The characteristics of the tube, particularly shutter time and spectral response, must be known as exactly as possible. Consequently we describe in these pages the method used to measure a shutter time of about 75 ps and we appraise the errors due to the FWHM and the spatial unstability of the UV laser pulse employed to illuminate the front side of the MCP. The measurement of the spectral sensitivity, determined by experiments with synchrotron radiation on the storage ring SUPERACO in Orsay (France), shows a response in the energy band 2 - 5 keV different from the one measured with the gold vacuum-deposited photocathode on the MCP.

Applied program package ELIM\DYNAMICS\ is designed for simulation, optimization and computer-aided design of dynamic image tubes with electrostatic focusing and deflection. The package allows highly precise computation of basic electrical and electron-optical characteristics including spatial/temporal line spread functions, spatial and temporal resolution along the photocathode working area in static and streak modes, spatial and temporal scale distortion, as well as assembling tolerances associated with image errors due to weak disturbances of axial symmetry. One of the main advantages of the new version presented is that `through' computation of electron beam from photocathode to screen is available with regard to dynamic aberrations caused by scattering fields located nearby the edges of deflecting plates. Some examples of image tube computations made with ELIM\DYNAMICS\package are given.

Theoretical investigation mainly on the physical temporal resolution of our newly designed femtosecond streak image tube has been made by means of numerical simulation, in which the influence of space charge effect on the nature and extent of intensity-dependent temporal broadening has been taken into account. The study includes simulating the emission of a great number of photoelectrons as a statistical sample from the position corresponding to the actual photocathode in terms of Monte Carlo method in proper probability, tracing these electrons trajectories by resolving the electron motion equation in the electron optics focusing system in the streak tube and dealing with the repulsion among these electrons i.e. space charge effect simultaneously, and finally analyzing the position distribution of the electrons statistically to come into conclusion. Through the agency of the step above can the photoelectrons' behavior inside the streak tube including the response of space- and time-dependent quasi-Delta functions in the focusing system be obtained principally. The result has shown that the space charge effect is the essential obstacle on the improvement of streak tube's performance both in temporal resolution and in dynamic range. Our theoretical estimate has indicated that considering the serious influence of space charge effect when streak tube works under illumination of the domain of few tens femtosecond the temporal resolution of less than 100 fs with acceptable dynamic range is expectable. Also the distortion of a waveform with FWHM is a few tenths picosecond is discussed which is due to the electron pulse's transformation dispersion and space charge effect and is more concerned in practice uses.

A variety of streak/framing/synchroscan image tubes are now under design and manufacturing in the Department of Photoelectronics, General Physics Institute. Among them are: a series of the well-known PV001 image tubes introduced into wide practice since 1978, a set of more advanced PIF001 tubes originally designed in 1979, specially developed femtosecond streak tubes of BSV-type, which were initially proposed in 1987, and finally a number of PF-type tubes placed in service last year. The whole set of these image tubes may cover the spectral range from 115 nm up to 1.55 micrometers , providing maximum sensitivity of 0.5 (mu) A/W at 1.55 micrometers (S1/IR) and up to 3 mA/W at 900 nm (S25/ER). Various input photocathode windows may be used: fiber-optics or borosilicate substrates which blue transparency starts at 350 nm. UV-glass windows (> 200 nm), MgF₂ input window (> 115 nm). All tubes with photocathode- accelerating mesh geometry have photocathode area of 6 mm in diameter, while the tubes in non-mesh configuration (PV and PF0 have a rectangular photocathode area of not less than 4 mm by 18 mm. The described tubes may be supplied with any type of phosphor screen (red, orange, blue, green) deposited onto fiber-optics faceplate.

Streak tubes design in General Physics Institute of Russian Academy of Sciences is traditionally supported by computer modeling. Numerical examples which illustrate the main phases of electron image formation inside such tubes are presented and discussed in this paper. Some factors resulting in spatial and temporal resolution decrease are examined in details. These factors include various geometric and chromatic aberrations of electron-optical focusing system, dynamic aberrations of deflection system, temporal instrumental function broadening due to Coulomb interaction effects inside the electron beam. Contribution of photoelectrons spread in energy and angles to spatial and temporal resolution is also studied for various photocathode types and streak speeds.

In-line Fraunhofer holography has been developed and implemented at the Los Alamos National Laboratory to measure particle distributions of fast moving particles. Holography is a unique diagnostic that gives unambiguous information on the size and shapes of particle distribution over a 3D volume. Currently, the capability of measuring particles two microns in size which travel many mm/microsecond(s) ec has been demonstrated in hydrodynamic experiments at the Pegasus Pulsed Power Facility. Usually, for setting up an in-line holography experiment for measuring particles a few microns in size, the holographic film would be placed less than one centimeter from the particles. However, due to the high kinetic energy associated with the dynamic experiment, an optical relay system is used to relay the interference pattern 35 cm so that the glass hologram will survive. After the hologram has been recorded the data must be extracted. A spatially filtered laser is used to reconstruct a real image which is a projection of the particles over a 3D volume. Planes of data from this volume are digitized via a CCD camera by moving the hologram with a three axis actuator. After the data has been digitized it is then analyzed with intelligent image processing algorithms.

A high speed data acquisition system has been developed for monitoring the speckle pattern produced by a fiberdyne interferometer. The system captures data from a 256 element, linear, CCD array with sample frequencies of up to 18.5 kHz. Data from the system is in the form of a 2D image showing the variation in pixel intensity as a function of time. The system is used to address the speckle pattern produced by a fiberdyne interferometer which is used as a vibration sensor. Information taken using this system is compared with the predictions of an existing theoretical model which indicates that all speckles should be affected by a vibration signal but, that the amplitude and phase of the resulting variations will be speckle dependent. Data taken using this system agrees with these predictions. All of the pixels in the CCD array show some variation when subjected to a sinusoidal vibration signal. The amplitude and phase of the pixel signals show a considerable variation in both amplitude and phase, indicating that the model is valid. By averaging the spectra obtained by performing a Fourier transform on each pixel it is possible to substantially reduce noise effects making the system more sensitive to changes in the acoustic signal.

The idea of time-resolved electron microscope, which is composed of a thermal emission electron gun, focusing lenses, object lens, intermediate lens, projector lens, scanning deflector, multi-compensation deflectors, fiber plate based phosphor screen and image intensifier, based on an integration of transmission electron microscope and picosecond framing techniques has been proposed. The design methods, including the calculation of magnetic field, dynamic electric-magnetic field, electron trajectory tracing and spatial and temporal spread functions, are also involved. The operation mode of this time-resolved electron microscope, in order to obtain microscopic framing images for the studied object, are discussed. Finally, this paper shows the design results and concluded that 6 framing images with the size of (Phi) 10 mm, temporal resolution of 100 ps and spatial resolution of 100 angstroms can be achievable by this design.

This paper describes the method of measurement ultrafast electrical pulses with a gain switch semiconductor laser and a LiTaO₃ probe tip. The results are presented. The results are comparable to the digital sampling oscilloscope with 20 GHZ bandwidth. The minimum detectable voltage is 1.54 mv, the temporal resolution is 25 ps. The voltage sensitivity of 5.4 uv/(root)Hz. This system has compactness stability and broad application prospects.

A high-speed photographic camera in which a concept of Cranz-Schardin has been embodied and which has been manufactured on the basis of present requirements upon technique of recording fast events is described in this paper. The camera operates in a convergent ray beam and is intended for recording non-self-luminous processes, that is, transparent optical nonuniformities that arise in a gaseous medium being disturbed by shock waves or compressional waves from moving objects or explosions of charges of different origin as well as by shock waves that accompany electric and laser breakdown. A framing rate for this camera is defined by lighting pulse duration and lies in the range of 10⁴ to 5 X 10⁶ frames/s. A process is recorded on a 35 mm film. The number of frames is 6. A frame size is 15 mm X 24 mm.

The diagnostic complex presented includes two stigmatic spectrometers for the visible, close UV and IR region with electronic registration: (1) the high resolution spectrometer: wavelength range 200 - 800 nm; light throughput f/3; dispersion changed in the range 0.15 - 2.5 nm/mm; image intensifier in 5 X 10⁴ times; spectral channel number--512; spatial channel number--8; time resolution 1 - 100 msec; the best spectral resolution--0.01 n; dynamic region 4 X 10³; (2) survey spectrometer: full wavelength range 200 - 850 nm (for simultaneous registration: 380 - 850 or 200 - 400 nm); f/N--3; image intensifier in 5 X 10+4) times; time resolution 20 - 1000 msec; spectral resolution--0.2 nm; dynamic region 4 X 10³. This complex was used for diagnostic measurements at the TSP and T-11M Tokamak (Russia) and for investigations in the diverter region of DIII-D Tokamak (USA). The results of these measurements are presented.

MicroChannel Plates (MCPs) are used in high speed cinematography systems such as MCP framing cameras and streak camera readouts. In order to know the dynamic range or the signal to noise ratio that are available in these devices, a good knowledge of the performances of the MCP is essential. The point of interest of our simulation is the working mode of the microchannel plate--that is light pulsed mode--, in which the signal level is relatively high and its duration can be shorter than the time needed to replenish the wall of the channel, when other papers mainly studied night vision applications with weak continuous and nearly single electron input signal. Also our method allows the simulation of saturation phenomena due to the large number of electrons involved, whereas the discrete models previously used for simulating pulsed mode might not be properly adapted. Here are presented the choices made in modeling the microchannel, more specifically as for the physics laws, the secondary emission parameters and the 3D- geometry. In a last part first results are shown.

High precision and high dynamic range measurements are required to properly characterize the pulseshape in ICF laser systems. The dynamic range and precision of the measurements that can be made with streak tubes is determined by the number of photoelectrons that can be transported to the recording screen per channel and per temporal sample of the signal. They are limited by the overall current that the tube can deliver without distorting the signal. In order to build tubes with the large dynamic ranges required by this application, we need to understand at what current density level the tube response becomes nonlinear. We present results of experiments made by using a laser illuminating several streak tubes at various intensities. We show that charge depletion in the photocathode can occur in pulsed mode and limit the signal to levels well below that where space charge induced nonlinearities appear. The use of scientific grade CCD's for recording the streak traces has allowed the introduction of a new method to measure the absolute current in the tube. A 2D PIC electron optics code has been used to simulate the propagation of the beam in the input diode (photocathode/accelerating electrode) and in the drift region (accelerating electrode/screen) of a bilamellar streak tube. We compare the numerical results with the experiments. We conclude by comparing the bilamellar and classical electron optics tubes and show the advantages of the former tube design for this application.

In this paper, a new type of synchroscan streak camera which works in fixed frequency but can measure ultra-fast weak light phenomena with arbitrary repetition rate, is reported. The array square wave-form pulse generator for choosing synchronized photoelectron pulse has been developed. The pulse top width is 5 ns, bipolar voltage amplitude is +/- 50 volts and repetition rate is 1M Hz. The temporal resolution of this camera, which is independent on the stability of the repetition rate of laser pulse, is better than 10 ps.

The design and experimental performance of the femtosecond synchroscan camera are presented. By utilizing the electrical shutter with variable durations we have evaluated the influence of phase noise in the deflection signal and other limiting factors on the achievable temporal resolution of the camera. The best temporal resolution less than 500 fs was obtained with integration time of 0.1 ms.

Current GPI status on femtosecond cameras and diffractometers research and development is overviewed. Discussed are the key components of the femtosecond diffractometer which is now under design. The first experimental prototype of 20 - 40 kV, femtosecond photoelectron diffractometer is computer designed, manufactured and tested.

An understanding of bubble flows is important in the design of process equipment, particularly in the chemical and power industries. In vapor-liquid processes the mass and heat transfer between the phases is dominated by the liquid-vapor interface and is determined by the number, size, and shape of the bubbles. For bubble flows these characteristics are often controlled by the generation mechanisms and, since bubble flows are often generated at an orifice, it is important to determine the controlling parameters which dictate how bubbles grow and detach. For bubbles growing at orifices the liquid displacement is an important feature and affects the pressure distribution acting on the bubble and the heat and mass transfer that may occur at the bubble interface. Therefore, in this study, the characteristics of the liquid velocity field are studied experimentally using Particle image Velocimetry (PIV) during growth, detachment and translation of a bubble being generated at an orifice supplied with a constant mass flow rate of air. The process is transient and occurs over a period of approximately 50 msecs. In order to map the transient flow field a combination of high speed cine and cross correlation PIV image processing has been used to determine the liquid velocity vector field during the bubble growth process. The paper contains details of the PIV technique and presents several of the velocity vector maps calculated.

This paper re-examines the collapse of a single cavitation bubble in front of a boundary. A range of different diameter cavitation bubbles were generated by focusing a pulse from a Q-switched Nd:YAG laser of varying energy. High speed shadow photography and Schlieren techniques are used to visualize the collapse process and the resulting bubble deformation at a distance s from the boundary with high temporal and spatial resolution. A sequence of Schlieren images taken with nanosecond time resolution displays the evolution of the cavity at the distance s from the solid surface. Additionally, the pressure distribution of the oscillating bubble especially in the region of the first collapse was recorded by a thin film transducer which is bonded to a pmma block. Both transducer output signals and Schlieren images confirm the formation of a liquid jet and also indicate the importance of the resulting fluid flow to stresses induced in the solid. In an attempt to visualize the fluid flow a flexible rubber film was placed near an oscillating bubble in free space. As expected jet formation in this case is not observed, however unusual fluid flow effects around the bubble can be seen.

Sheets of metal can be bulged by impulsive high pressure produced by the underwater spark discharge. The following two matters are considered as the cause of the impulsive pressure. One is generation of a spherical shock wave that results from rapid evaporation of water by heat of an arc and another is that of a rapid water flow called a microjet that arises from the movement of a bubble containing water vapor produced by the arc. It is supposed that the pressure produced by the latter is not less effective than the shock wave on bulging sheets of metal. In order to make the effect of the water flow on the forming clear, we have bulged each aluminum plate (thickness equals 0.2 [mm]) set up on a die with a hole (diameter equals 3 [mm]) in the center of it by the underwater spark discharge and examined how the height of bulge of each aluminum plate varies with variable distance between the spark gap and the aluminum plate. Besides we have observed the behavior of the bubble by a high speed camera. As a result of this study, it is reported that there is the most suitable distance between the spark gap and the aluminum plate for bulging, because the pressure caused by the rapid water flow, that is, the microjet operates most effectively on the aluminum plate at that distance.

The high-speed phenomena caused by the focusing of a pulsed laser beam in liquid nitrogen is studied experimentally with an image converter camera. The focusing of a laser beam elevates the energy density at the focus. When the energy density exceeds a critical value necessary for optical breakdown, there appears luminous plasma, and a shock wave is emitted from the plasma into the bulk liquid. The initial shape of the vapor bubble and the shock front is nonspherical because of reflecting that of plasma. While the plasma relaxes to a vapor bubble, however, several compression waves are emitted following the preceding shock wave, so that both the bubble and the shock front become spherical. The resulting spherical bubble grows under the dominant effect of liquid inertia, and collapses after reaching a maximum size. The high vapor pressure retards the collapse process from the growth and makes the bubble surface rough. While the bubble repeats the growth and collapse processes a few times, that distribution over its surface is amplified through every rebound, eventually breaking up the bubble itself.

The fluid instability resulting from the impulsive acceleration a thin, heavy gas layer embedded in a lighter gas is characterized through high-speed flow visualization. The shock-accelerated event occurs over the course of 1 ms. A laser sheet illuminates a plane of the gas layer and its evolution is recorded by a camera observing high-angle scattered light. Two sets of diagnostics were evaluated. These derived their time resolution either from gating the camera or pulsing the illumination source. The arbitrary inter-exposure timing associated with the gated camera was found to be significantly advantageous over the fixed intervals of the pulsed laser. The intensified, gated camera also placed more manageable requirements on the light source and provided sufficient data to measure the instability growth rates. However, the pulsed laser arrangement produced superior spatial resolution and dynamic range primarily due to the lack of an intensifier.

A method is described for controlling and measuring the pulse width of electrical gate pulses used for optical shuttering of image intensifiers. The intensifiers are coupled to high frame rate Charge-Coupled-Devices (CCD) or Focus-Projection Scan vidicon TV cameras for readout and telemetry of time resolved image sequences. The shutter duration or gate width of individual shutters is measured in real time and encoded in the video frame corresponding to a given shutter interval. The shutter information is updated once each video frame by strobing new data with each TV camera vertical sync pulse. This circuitry is used in conjunction with commercial video insertion/annotation equipment to provide the shutter width information in alpha numeric text form along with the time resolved video image on a frame-by-frame basis. The measurement technique and circuitry involving a combination of high speed digital counters and analog integrators for measurements in the 1 ns to 1024 ns range are described. The accuracy obtained is compared with measurements obtained using high speed DSOs. The measured data are provided in 10-bit Binary (Bi) and four decades of Binary Coded Decimal and also displayed on four digit seven segment displays. The control circuitry including digital and analog input means for gate width selection are described. The implementation of both measurement and control circuitry into an Intensified Shuttered CCD radiometric system for recording fast shuttered images at RS-170 to 4 KHz frame rates is presented.

The combination of two complementary technologies developed independently at Los Alamos National Laboratory (LANL) and Sandia National Laboratory (SNL) has demonstrated feasibility of target detection and image capture in a highly light-scattering medium. The technique uses a compact SNL developed Photoconductive Semiconductor Switch/Laser Diode Array (PCSS/LDA) for short-range (distances of 8 to 10 m) large Field-Of-View (FOV) target illumination. Generation of a time-correlated echo signal is accomplished using a photodiode. The return image signal is recorded with a high- speed shuttered Micro-Channel-Plate Image Intensifier (MCPII), designed by LANL and manufactured by Philips Photonics. The MCPII is gated using a high-frequency impedance-matching microstrip design to produce 150 to 200 ps duration optical exposures. The ultra fast shuttering produces depth resolution of a few inches along the optic axis between the MCPII and the target, producing enhanced target images effectively deconvolved from noise components from the scattering medium in the FOV. The images from the MCPII are recorded with an RS-170 Charge-Coupled-Device camera and a Big Sky, Beam Code, PC-based digitizer frame grabber and analysis package. Laser pulse data were obtained by the system but jitter problems and spectral mismatches between diode spectral emission wavelength and MCPII photocathode spectral sensitivity prevented the capture of fast gating imaging with this demonstration system. Continued development of the system is underway.

The development of an advanced instrument for real-time radiometric imaging of high-speed events is described. The Intensified Digitally-Controlled Gated (IDG) camera is a microprocessor-controlled instrument based on an intensified CCD that is specifically designed to provide radiometric optical data. The IDG supports a variety of camera- synchronous and camera-asynchronous imaging tasks in both passive imaging and active laser range-gated applications. It features both automatic and manual modes of operation, digital precision and repeatability, and ease of use. The IDG produces radiometric imagery by digitally controlling the instrument's optical gain and exposure duration, and by encoding and annotating the parameters necessary for radiometric analysis onto the resultant video signal. Additional inputs, such as date, time, GPS, IRIG-B timing, and other data can also be encoded and annotated. The IDG optical sensitivity can be readily calibrated, with calibration data tables stored in the camera's nonvolatile flash memory. The microprocessor then uses this data to provide a linear, calibrated output. The IDG possesses both synchronous and asynchronous imaging modes in order to allow internal or external control of exposure, timing, and direct interface to external equipment such as event triggers and frame grabbers. Support for laser range-gating is implemented by providing precise asynchronous CCD operation and nanosecond resolution of the intensifier photocathode gate duration and timing. Innovative methods used to control the CCD for asynchronous image capture, as well as other sensor and system considerations relevant to high-speed imaging are discussed in this paper.

This paper describes a high speed, low noise analog integrated circuit which has been designed to interface with charge-coupled device (CCD) arrays in high speed CCD camera systems. This IC performs the analog signal processing functions required between the CCD output and analog-to- digital converter input. Channel gain can be adjusted from 2.7 to 12 in 16 steps as specified by a 4 bit digital word. The chip operates from power supply voltages of +/- 5 V, dissipates 380 mW/channel, and has an input referred noise of 260 (mu) V rms.

\Lfl ultra high frame rate CCD type solid state image sensor has been developed, capable of frame rates up to a million frames per second. Referring to Figure 1 and Table 1 ,each macro-pixel of the 360 x 360 pixel array ontains a photo-detector and a 30 element CCD type memory array. The image acquisition frame rate is the rate at which photoelectrons are read out of the photo-detector into the pixel' s memory array. After the pixel memory fills, further exposures replace the oldest stored images in a "first in- first out" mode of operation. Thus the imager continuously stores the last 30 images acquired. This allows a pre-trigger mode of operation, making it possible to capture poorly predicted and spontaneous transient events. Readout of the stored images is similar to a frame transfer CCD readout and is at a comparatively slow pixel rate, consistent with low readout noise and inexpensive, PC based data acquisition systems. During the readout, the thirty images are demultiplexed and assembled in the computer's memory for display. A fast framing camera system employing this unique image sensor is currently being evaluated for a variety of applications. This paper presents these evaluation results which include high frame rate image sequences of hypersonic gas turbulence and fluid cavitation processes.

A novel video stroboscope for periodical and nonperiodical repetitive high-speed events has been developed, constructed and tested. The stroboscope does not utilize a flash light. The object can be illuminated by a standard CW light source such as a halogen lamp or by day light. Hence, self-luminous objects can also be examined. The system consists of an asynchronously shuttered progressive scan CCD camera, a frequency-independent digital phase shifter and a PCI-bus frame grabber. Both, the phase shifter and the frame grabber are located in the same Pentium computer. Optionally an universal counter and frequency synthesizer for control purposes can be added. The image of the stroboscopically investigated event is displayed live on the PC monitor. The shutter of the camera is triggered by the event via the phase shifter which can be programmed to change the phase shift slowly. In the case a slow motion effect can be realized in real time. The apparent speed of the displayed movement does not depend on the real event frequency. Both the phase shift and the slow motion period are selectable independent of the real event frequency. Non-periodical but repetitive events can be observed in an alternative time- delay mode of the phase shifter. The camera exposure time can be varied between 1.25 microsecond(s) and 16 ms. It allows the stroboscopic observation of very fast oscillating (above 10 kHz) and rotating (above 100,000 rpm) objects. Sequences of up to 300 frames each can be stored on the hard disc derive for documentation, evaluation and playback. Each frame has a maximum resolution of 659 X 494 pixel at 256 gray levels.

A digital high speed photodiode array (PDA) camera had been developed in NINT (Northwest Institute of Nuclear Technology). The high speed digital camera was designed compact with high performance that could be applied to conduct area scan. It was constructed by using a high speed PDA as image sensor, which has a fiber optic faceplate coupling to the sensing surface of the device. The camera had a single channel video format that allows video rates up to 1,000 frames per second with exceptional noise immunity and more than 100 times optical dynamic range. Because of the uses of digital signal processing, the quality of image could be made very clearly. We have also adopted the technology, Field Programmable Gate Array, as timing block which makes the camera better work with high reliability. The camera has a wide-range output voltage up to 18 Vp-p, which is suitable for long distance transmission. Advanced features such as frame reset and single work status allows the camera was synchronized to external events. This work details the essential design features and the applications of the high speed PDA camera.

A technique is presented, which consists of a computer program and ancillary instruments: (1) to adjust displacement of images taken by two or three image sensors and (2) to calibrate relationship between incident light intensity and corresponding digital output signal simultaneously. The pixel-based displacement adjustment is made by corresponding a pixel of one image sensor with four pixels adjacent to that point on another sensor. The weighted average of outputs for the four pixels are then multiplied by a coefficient calculated from a pixel-based sensitivity calibration.

Dynamic behavior of the imploding detonation in C₂H₂-O₂ mixtures is monitored using a high speed image converter camera (Imacon 790) with frame intervals of 50 to 500 nsec. Excellent cylindrical symmetry is maintained until the imploding shock wave is confined as small as 0.2 mm in diameter. Temperatures behind the reflected shock waves are determined from the spectral intensity distribution of the plasma and a theoretical model for the strong reflected shock wave of the imploding detonation is examined by comparing the experimental results. Line spectrum of calcium ion emitted from the CaF₂ window is also identified as a source of the very strong radiation with a long life time at the center of the implosion.

The free piston shock tunnel (FPST) is one of the most useful ground test facilities for generating hot hypersonic flows. the piston motion depends upon its characteristics. However, the piston speed has never been measured continuously. In this study, a piston speed was measured continuously and precisely in 50 mm dia. and 2,000 mm length compression tube simulating FPST, using a piston motion velocity interferometer, which was developed specifically to measure a piston speed inside a compression tube. Then piston trajectory was obtained by integrating the measured speed. These experimental results were compared with the numerical predictions of a fourth order non-oscillatory scheme. Good agreement was obtained between the experimental and the numerical results. On the other hand, flows around a model suspended in hot hypersonic flows were visualized using a double-exposure holographic interferometry, because real gas effects of high enthalpy flows and driver gas contamination hamper quantitative comparison between measured and numerical results. Then driver gas contamination severely limits test time. For the initial installation of the FPST, the variation of shock stand-off distances also indicated if driver gas contamination was present. The duration time of steady-state flow in the test section was estimated from variation of the fringes in the shock layer, the shock stand-off distance and shock angle.

This paper presents an experimental and numerical investigation of 3D shock wave reflections over a corner of two wedges intersecting perpendicularly in a shock tube. Experiments were conducted in a 100 mm X 180 mm diaphragmless shock tube equipped with double-exposure diffuse holographic interferometry in which the time interval between the first and second exposures was set to be 1 microsecond(s) . This arrangement clearly visualized complex configurations of 3D shock wave reflections. A numerical study was also carried out for interpreting these holographic interferometric observations by using the Weighted Average Flux method to solve the 3D unsteady compressible Euler equations. It was found that along the line of the intersection of these two wedges, two Mach stems intersected each other resulting in the formation of a Mach stem which was leaned forward.

This paper describes a compact laser, which produces high power, wide-angle emission for a near-field, range-gated, imaging system. The optical pulses are produced by a 100 element laser diode array (LDA) which is pulsed with a GaAs, photoconductive semiconductor switch (PCSS). The LDA generates 100 ps long, gain-switched, optical pulses at 904 nm when it is driven with 3 ns, 400 A, electrical pulses from a high gain PCSS. Gain switching is facilitated with this many lasers by using a low impedance circuit to drive an array of lasers, which are connected electrically in series. The total optical energy produced per pulse is 100 microjoules corresponding to a total peak power of 100 kW. The entire laser system, including prime power (a nine volt battery), pulse charging, PCSS, and LDA, is the size of a small, hand-held flashlight, System lifetime, which is presently limited by the high gain PCSS, is an active area of research and development. Present limitations and potential improvements will be discussed. The complete range-gated imaging system is based on complementary technologies: high speed optical gating with intensified charge coupled devices (ICCD) developed at Los Alamos National Laboratory and high gain, PCSS-driven LDAs developed at Sandia National Laboratories. The system is designed for use in highly scattering media such as turbid water or extremely dense fog or smoke. The short optical pulses from the laser and high speed gating of the ICCD are synchronized to eliminate the back-scattered light from outside the depth of the field of view (FOV) which may be as short as a few centimeters. A high speed photodiode can be used to trigger the intensifier gate and set the range-gated FOV precisely on the target. The ICCD and other aspects of the imaging system are discussed in a separate paper.

A compact high-intensity pulsed light source has been developed in order to match a microdynamic test facility for high-speed motion analysis of micromechanical components. The test stand encompasses a universal microscope Zeiss Axioplan, the new light source and an electronic ultra high- speed multiple framing camera Hadland Imacon 468. The light source consists of a narrow cylindrical Xe-filled discharge tube, thus providing a locally stable emission. Since the small-size flashlamp easily fits into a standard microscope lamphousing, it allows to maintain the advantages of Koehler illumination as well as switching to other types of lamphousings. The flash tube is operated via an artificial asymmetric transmission line and delivers a square light pulse with a flash duration of 110 microsecond(s) FWHM and a peak intensity of 50 Med. The light source illuminates the object uniformly within the interesting time window; image shuttering is provided in the camera by gated micro-channel- plate intensifiers. To test the efficiency of the total system for various standard visualization methods (transmitted light, reflected light and differential interference contrast), microscopic still images have been taken at magnification up to 500X and with exposure times down to 10 ns. In addition, two microscopic darkfield methods which provide a high contrast but a low light intensity of the image, have been selected to test their applicability down to an exposure time of 100 ns. Two examples for real-time cinematography of high-speed phenomena in microactuators are shown: the bouncing behavior of an electro-magnetic microrelay and the bubble/jet formation of a thermal ink jet printhead.

A flashlight system has been developed to generate high intensity luminance which would allow high speed imaging up to 10,000 frames per second in the field where there was no power supply or in a room with restricted power capacity. It gives higher luminance than continuous light sources thus making it possible to set shorter exposure times and use a smaller aperture lens at the same framing rate. We have obtained many successful color images by using a combination of this light source and a 16 mm rotating prism camera. This paper outlines the features of the flashlight system, and describes some applications that make best use of this system.

The vacuum spark is an excellent source of pulsed X-rays (also called flash X-rays) suitable for high speed photography. In the vacuum spark concept a capacitor is discharged through two properly shaped electrodes, made of selected materials, in a vacuum. This high current (over 1 kA) discharge produces intense pulsed hard X-rays with a pulse width of about 10 ns (FWHM, Full Width at Half Maximum). The measured source size by pinhole photography is smaller than 0.5 mm. Efforts have been made to reduced the total inductance (below 200 nH) and to use a relatively small capacitor (just a few nF), so as to increase the X-ray intensity. A vacuum spark X-ray source (VSX I) has been under routine operation at ALFT and has logged over 1,200 shots during X-ray tests carried out with Los Alamos National Laboratory. The radiation head was designed and built by ALFT and the remaining components are all commercial, off-the-shelf products. An external signal of 10 V, 1 ns rise time and 500 ns width triggers the machine at rep-rates up to 10 Hz, and higher rep-rate operation of the vacuum spark is being studied at ALFT.

This paper describes a tunable fiber soliton laser. This laser is configured as a ring with a silica optical fiber gain medium. A length of 300 m conventional fiber provides the negative group velocity dispersion, self-phase modulation and Raman gain effect. The time-dispersion tuning technology is used for wavelength selection, because the Raman gain of silica fiber has a wide spectrum. The different wavelengths have different optical lengths, when they propagate in a optical fiber. We use an active modelocking Nd:YLF laser which has wavelength 1.313 micrometers with pulse width of 70 ps for pumping source and silica optical fiber of 300 m for gain medium and group velocity dispersion compensator. The pulse width (FWHM) of 195 fs has been got, the laser wavelength is 1.395 micrometers and tunable region of 47 nm near 1.397 micrometers .

This paper presents a Ti-sapphire-based laser amplification system in which the technique of a chirped- pulse- amplification is used. It consists of a laser oscillator, a pulse stretcher, a regenerative amplifier and a pulse compressor. The laser oscillator is a home-made self-mode- locked Ti-sapphire (3 mm long) oscillator. The oscillator pumped by 5 W of cw all line Argon laser can generate Fourier-transform-limited sub-20 fs duration pulses at a 91 MHZ repetition rate, a center wavelength of 800 nm, a FWHM bandwidth of a spectrum of 80 nm. The seed pulses from the oscillator are stretched by a factor of 1000 with a four- pass grating stretcher at first. After stretching, these pulses are seeded into a Ti-sapphire regenerative amplifier pumped by a frequency-doubled Q-switched a Nd:YAG laser at 5 KHZ. We have obtained single-pulse energy of 100 (mu) J, pulse duration of 50 fs at the repetition rate of 5 KHZ.

Prior to the investigation of the photographic process over 150 years ago, the analyses of rapid motions were limited by the dynamic efficacies of the human eye, which has a temporal resolution of approximately 1/10 sec and a maximum information acquisition rate estimated at 10³ to 10⁴ bits/sec. At high rates of object motion, only the simplest actions can be resolved, comprehended and retained in human memory. Advances in the field of high-speed photography drastically changed all this by providing us with the ability today to capture permanent images of transient events at acquisition rates in excess of 10¹² bits/sec. As remarkable as these improvements in temporal resolution and image retention may be, the final step in correctly interpreting any image still rests largely upon the analyst's ability to process visual data. Those who enter the field of image analysis soon learn how capricious the eye can be in this task. It is incumbent upon anyone performing important image analyses to have at least a basic understanding of the eye's performance characteristics, especially its limitations and capricious anomalies. Exemplary data presented in this paper are drawn from the scientific literature and the author's forty years of experience as a researcher, author and educator in the field of high-speed imaging.

The Hubble Space Telescope started out on its mission of infamy and is now providing unparalleled information on the universe. Shortly after its launch in April 1990, the HST was found to have a problem with the primary mirror. After extensive analysis, it was concluded that serious spherical aberration was present caused by an improper manufacturing process and measurement of the surface. A servicing and repair mission was launched on Dec. 2, 1994 to correct and compensate for the primary mirror problem and also to replace other portions of the support system on the HST. Since its repair, the HST has been providing information of astronomical events never previously considered possible and far exceeding any ground based telescope systems. This paper describes the telescope, its problems, and the corrections to compensate and improve the imagery from the HST.

This paper describes the contents of a unique introductory, applications oriented, high speed photography course offered to Imaging and Photographic Technology majors at the Rochester Institute of Technology. The course covers the theory and practice of photographic systems designed to permit analysis of events of very short duration. Included are operational characteristics of intermittent and rotating prism cameras, rotating mirror and drum cameras, synchronization systems and timing controls and high speed flash and stroboscopic systems, and high speed video recording. Students gain basic experience not only in the use of fundamental equipment but also in proper planning, set-up and introductory data reduction techniques through a series of practical experiments.

Some technique is presented for IR photocathodes manufacturing on the basis of In_0.53Ga_0.47As/InP heterostructures with Shottky barrier. Discussed is a method for ultra-high vacuum transfer of the photocathodes into vacuum devices. It is shown that the sensitivity of photocathodes in a sealed out device at (lambda) equals 1.55 micrometers is two orders of magnitude higher comparing to the sensitivity of traditional Ag-Cs-O photocathodes. The developed photocathodes may be used in time analyzing image tubes covering the spectral range from 0.9 to 1.7 micrometers .

In this paper, a universal streak camera will be introduced. This camera consists of time distortion-compensated optical lens, vacuum ultra-violet-sensitive streak image tube, multi-function sweep circuits and CCD real-time readout system. It can work in 3 modes, including single shot with temporal resolution of 1.2 ps obtained, synchroscan, dual scan and circular scan mode also. The temporal resolution of 1.2 ps can be achieved in single shot mode. The operation frequency can be 82 MHz, 164 MHz, 200 MHz and 300 MHz for synchroscan and circularscan modes. In this case, the temporal resolution is operation frequency dependent, usually in the range of 1 - 3 ps. In dual scan operation mode, one deflector is driven by sine wave circuit, another by a slow ramp circuit. The scan speed along the fast and slow axis is 5 cm/ns and 26 cm/microsecond(s) -2 cm/microsecond(s) respectively.

A set of versatile streak camera was developed in GPI Photoelectronics Department to meet various experimental requirements in laser and plasma physics, laser ranging, fiber optics communication and ecology. Among these camera there are milli-micro-nanosecond camera intended for explosion studies, nano-picosecond streak camera--for application in communication field and nano-subpicosecond streak camera--for laser and laser plasma research.

A picosecond gated MCP framing camera system is presented. The camera with four parallel microstripline cathode has longer measuring time range and better gain uniformity. The camera is equipped with a pinhole-array adjustment system and other practical accessories for laser plasma experiment. The measured spatial resolution of the camera is 15 lp/mm with 10% modulation and the exposure time ranges from 60 ps to 100 ps with different gain. The whole system was used at the 11# laser facility and got some results of several target type during laser plasma experiment.

The high-speed camera KIT-2F has been developed, intended to record spatial distribution of IR-radiation energy and power on the photographic film. The instrument has a wide dynamic range of recording (approximately 10³), high sensitivity (approximately 10^-6 J/cm²), and fast response (approximately 10^-6 s) in a wide spectral range 1 - 10.6 micrometers . The camera has found its application in investigations of fast processes in gas dynamics, ballistics, quantum electronics, and in the development of new industrial technologies.

The Corner Turning Distance (CTD) can be measured with different optical diagnostic methods or also with FXR. The CTD's are important values of the detonability of high explosive (HE) charges or indicate at what distance or radius they are able to change a forward detonation in a radial detonation direction. The less sensitive HE's have an increasing problem with this phenomenon. It will be presented an overview of the different, especially the optical diagnostic methods. Their advantages and disadvantages will be illustrated on a number of typical examples. Simple tests which give only the CTD's, and more scientific arrangements allow also to define the so-called dark or dead zones. The CTD values given also good indication for insufficient initiation strength. With this method explosive trains can be also checked very well in their initiation limitations.

It is known that emulsion explosives have non-ideally in their detonation. The behavior of the non-ideal detonation has not yet been sufficiently investigated. In order to obtain a better comprehension of the non-ideal detonation behavior, optical measurements and measurements of the velocity of the detonation wave were carried out. Streak photographs were taken by an image converter camera using an usual shadowgraph system. Five kinds of emulsion explosives which differ the aluminum contents were used in experiments. These emulsion explosives were put into a copper, an aluminum and a PMMA containers. In case of the copper pipe container, the velocity of the detonation wave is constant at any aluminum contents. But in case of the PMMA pipe container, the velocity of the detonation wave becomes lower in proportion to aluminum contents. The shapes of the reaction zone and the front of the detonation zone were also recognized from the streak photographs. In case of the copper pipe container, the detonation wave is plane, but in case of PMMA pipe container, the detonation wave curves near the pipe wall and its shape is a convex. It is considered that acoustic impedance of the container have effects on the propagation of the detonation wave.

This paper describes the use of high-speed photography, and videography, in the study of material distortion and movement when a shock wave traverses a highly deformable porous structure, such as a blob of foam or a porous bed of particles. The effects of surface porosity can be significant in determining the nature of reflection of shock waves from surfaces. Not only are wave geometries substantially modified but the resulting wall pressures are also strongly affected. It, in addition, the surface is highly deformable by being made up of an elastic matrix or a collection of discrete particles, then the reflection geometry and loading can be even more complex. It is known, for example, that shock wave impact on open-cell polyurethane foam attached to a wall can cause a significant increase in pressure on the wall compared to reflection off a plane rigid wall without covering. The motion of the interface is an essential consideration in understanding the dynamics of these interactions. These studies could have application to the effects of blast wave propagation over complex surfaces such as forests, grasslands, and snow; as well as in establishing the efficacy of safety padding and attenuation materials under shock and impact loading conditions. Studies on an assortment of materials are presented, using a variety of visualization techniques. Recording methods used range from short duration flash photography (both shadow and schlieren), through multi-frame videography; to single frame, multi-exposure video capture with a camera capable of rates up to 1 million pictures per second. In the case of shock wave impact on specimens of polyurethane foam, the results clearly show the expulsion and reingestion of shock heated gas from within the foam body as the material collapses and then recovers, coupled with longitudinal and transverse oscillations of the body of the foam material. For blast wave propagation over porous beds, occurrence of particle lift off, bed fluidization, and the generation of surface dunes are evident. The recordings allow the calculation of the velocities and accelerations of the various interfaces and particles to be made, using suitable image processing techniques. Thus, estimates may be made of the unsteady drag forces acting on the individual particles.

Results of the laboratory experimental investigation of the impact against the barriers of a number of materials by the shower (bunch) of tungsten particles (striker) having density equal to 1 g/cm³, mass equal to 0.2 g and velocity up to 24 km/s as well as their head-on collision with relative velocity equal to 48 km/s, the results of analysis and prediction are presented. By using filming with illuminating from explosive the registration has been realized. Penetration, reactive pulse and the character of energy release during the shower deceleration were studied. In the experiment the process development pictures (partial explosion during impact and subsequent penetration) which appeared to be intermediate between the complete explosion of the striker on the surface of the barrier and the explosion in the barrier after penetration has been realized. The hypervelocity collision process is also necessary to consider, to investigate and to develop as a system being possessed of its specific properties. The results presented in the report make possible to predict the explosion parameters for the wide range of strikers and barriers. The bases of new generalized approach to the problem of the impact action control are outlined. It is shown that in spite of the variety of its practical usage the process of collision can and must be considered as the system. It is stated that the basis and the tool to control the impact action is symmetry in all the diversity of its forms and violations.

The design and performance of a stable Ti:sapphire regenerative amplifier. 7 mJ (800 nm) per pulse at the repetition rate 10 Hz, has been described. The resonator consists of one concave mirror (R equals 4000 mm) and one plane mirror, spaced at a distance of 1.76 m. The two mirrors are coated for high reflection centered at 800 nm. Two Pockel cells, one broadband waveplate and polarizer are used to switch a pulse into and out of the resonator. The Ti:sapphire crystal rod is 6X6X14 mm with an absorption coefficient of 2.6 cm^-1 at 490 nm and both sides is cut at Brewster's angle. The pump source is frequency double Nd:YAG laser. For regenerative amplifier, the pump energy which we usually use is 64 mJ per pulse. The seed pulse is from a self-mode-locked Ti:sapphire laser that provides 400 mW of outpower with pulse duration about 80 - 100 fs. After stretching the pulse to about 200 - 300 ps by using the grating stretcher, we inject the pulse into the regenerative amplifier and obtain 7 mJ output energy in single pulse, a net gain of up to 10⁶. This laser system is a pre-amplifier for high energy amplifier.

We report the structure and steady-state operating characteristics of the femto-second Ti:sapphire laser. The analytical expression of saturable absorption parameter and selfphase modulation are obtained based on H.A. Haus's self- focusing model and the mode-locked equation by using the self-consistent condition. The relation between the output pulse width and these parameters, the influence of the parameter on the stability of laser are discussed separatively, and to lead some useful conclusions for the optimum design of the fs-laser and how to get as minimum pulse width as possible. By using above results, we have designed the Ti:sapphire laser that it can operate in self- mode locking while the input pump beam from argon-ion-laser is TEM_oo mode about a 4 W. The pulse as short as 60 fs duration has been obtained.

A tunable multi-wavelength Ti:sapphire laser with sub-10 femtosecond jitter between the Bi-color pulses has constructed that synchronously generates femtosecond pulse trains at two independent wavelength regions. Two wavelength pulses of duration 26 fs and 29 fs were generated at central wavelength regions from 755 nm to 845 nm. Also, synchronized generation of independent three wavelength pulses has been achieved by fine metallic net slit. Three wavelength pulses of duration 93 fs, 70 fs and 55 fs were generated at central wavelength 755 nm, 808 nm and 821 nm respectively. The jitter between the two color pulses has been obtained using cross-autocorrelation profiles of the two train pulses.

We report here a femtosecond Ti:sapphire laser that generates pulses at three wavelength regions. The central wavelengths are 755 nm, 808 nm and 821 nm. The bandwidths are 4.4 nm, 6.2 nm and 11.8 nm. At 755 nm, the pulse width we measured is 146 fs. Two coupled nonlinear Schrodinger equations describing the evolution of the pulses in the laser are solved numerically. The solution should be helpful in understanding the three-wavelength characteristics of the femtosecond Ti:sapphire laser.

We present the design and preliminary experimental data for the multicolor fiducial laser to be used as a primary timing reference for UV and x-ray streak cameras on the 40-kJ UV OMEGA laser system.

In 1991, the authors developed a high-speed video camera with a frame rate of 4,500 pps, which was the world's fastest at that time and is currently marketed by a third party. An MCP-type image intensifier is attached directly to the image sensor to realize both very high light sensitivity and frame rate. Three intensified image sensors are combined with a cubic three-way beam splitter prism and three optical filter holders to form this high-speed video/ultra-high- speed triple-framing camera. The camera can capture fast- moving images under the following conditions: (1) 4,500-pps continuous color imaging with R, G & B filters and perfect synchronization of the three images for a full 256 X 256 X 3-pixel frame, or faster color imaging of up to 40,500 pps at a reduced resolution of 64 X 64 X 3 pixels. (2) Triple-speed continuous monochrome imaging without filters and delayed synchronization, i.e. 13,500 pps for a full 256 X 256-pixel frame or up to 121,500 pps for a reduced 64 X 64-pixel frame. (3) Ultra-fast three-frame capture at a speed of 1/50,000,000 s by delayed gating of the image intensifiers. The cubic prism is also convertible, enabling simultaneous image capture by three monochromatic light sources of different wavelengths with no loss of incident light energy. This is in contrast with the large energy loss caused by a cubic prism using metalized half-mirrors, which reduced light transmission to one quarter, coupled with the losses associated with the three optical filters, reducing the light energy incident on each individual sensor still further.

The framing image tube camera has some intrinsic advantages. It has got some applications in ICF and X-ray laser field. But there are two main problems with this technique: the loss of spatial resolution at short exposure time,a nd the electromagnetic disturbance in the experimental area. In order to solve these problems, we have taken several measures: to improve the design of the image tube and the electronics and to strengthen the shielding. Additionally, in intense X-ray pulse we use MgF₂ as cathode material instead of Au and CsI to reduce the space charge effects. The experiments in high power laser area have shown that 3 frame images can be provided with 75 ps (FWHM) exposure time and dynamic spatial resolution of 6 lp/mm. Some physics results of high power laser-plasma interaction have also been obtained.

Improvement in visualization of ultra fast phenomena, using high speed imaging techniques has encouraged researchers to seek more from the equipment they use. As accurate data extraction has become reality the users' expectations have risen stimulating the camera designer's imagination. High speed cameras now include features the operation of which are only possible when controlled by a dedicated computer. High spatial and temporal accuracy derived from different imaging techniques can be positively complementary particularly when recorded through a common optical axis that eliminates errors resulting from parallax. Comprehensive analysis software has considerably increased the extraction of data from electronic imaging allowing critical interrogation of the recorded sequence. Simultaneous framing and streak recording of the same event, broadens the understanding of the processes involved, and has been made possible by the novel optical design used in the Imacon 468. The data collecting potential offered by these simultaneous recording techniques extends considerably the information derived from experimental procedures.

During experimentally researching the time behavior of a intermittent film-transport force, we found a different form than those given by domestic and abroad researchers, the film-transport force is not continuous. In this paper, the dynamic model of the intermittent film-transport has been established on the basis of practical film-transport and the approximate assumptions, from which we derived the film- transport force equation, simulated the film moving process in the film road on computer, and showed out the explanations to the experimental results.

There are many different types of measurements that require a continuous time history of x-ray emission that can be provided with an x-ray streak camera. In order to properly analyze the images that are recorded with the x-ray streak cameras operated on Nova, it is important to account for the streak characterization of each camera. We have performed a number of calibrations of the streak camera both on the bench as well as with Nova disk target shots where we use a time modulated laser intensity profile (self-beating of the laser) on the target to generate an x-ray comb. We have measured the streak camera sweep direction and spatial offset, curvature of the electron optics, sweep rate, and magnification and resolution of the electron optics.

In an explosive experiment, the effects of extra covers of thin plane plate, hollow wedge or cone placed in the front of the cassette protector assembly which consisted of three layers of interspaced aluminum plates were analyzed by comparing the optical densities of the x-ray films inserted between the layers. With the plate cover, all the plates of the protector assembly were severely deformed. And all the films were severely darkened owing to the impacts of deformed plates as well as the transmitted shock waves. With wedge or cone cover, no apparent deformation of the assembly plates was observed, and only the first film showed slight darkening owing to the shock waves. When a large amount of explosive has to be used or the film plane has to be placed very close to an explosion source, the cone type of extra cover can protect the film cassette from the explosive blast without degrading the radiographic image quality.

The paper discusses potential causes for femtosecond laser pulse (FP) broadening by an optics system. Design options are suggested for optics systems with minimized distortion in the FP space-time pattern.

This paper introduces a single chip microcomputer control system for synchronization operation of several rotating mirror high-speed cameras. The system consists of four parts: the microcomputer control unit (including the synchronization part and precise measurement part and the time delay part), the shutter control unit, the motor driving unit and the high voltage pulse generator unit. The control system has been used to control the synchronization working process of the GSI cameras (driven by a motor) and FJZ-250 rotating mirror cameras (driven by a gas driven turbine). We have obtained the films of the same objective from different directions in different speed or in same speed.

Time-resolved optical computer tomographic imaging has received considerably interest recently because of its non- invasiveness and non-destructiveness to biological tissue, and several attempts have been made for the implementation. The image recovery algorithm described in this paper is based on the finite-element method solution to the diffusion equation as the photon transport model. By formulating image reconstruction as an optimization problem with the objective function expressed by the error norm between actual and modeled measurement sets, this algorithm incorporates a direct search optimization method. Rotating Coordinates Method, into the solution strategy to avoid excessive computation of the time-consuming forward problem. Several numerically simulated results for the images of the absorption and scattering coefficient distribution within a circular tissue, reconstructed from the integrated intensity, the mean time of photon flight or their weighted sum, are given. The influence of the numbers of stimulating sources and boundary detection points on the resulting images is discussed. Finally, this paper concludes that used reconstruction algorithm is feasible but needs to be improved in its accuracy and efficiency.

Development and testing of a dual microchannel plate (MCP) module to be used in the national Inertial Confinement Fusion (ICF) program has recently been completed. The MCP module is a key component of a new monochromatic x-ray imaging diagnostic which is designed around a 4 channel Kirkpatrick-Baez microscope and diffraction crystals and is located at University of Rochester's Omega laser system. The MCP module has two separate MCP regions with centers spaced 53 mm apart. Each region contains a 25 mm MCP proximity focused to a P-11 phosphor coated fiberoptic faceplate. The two L/D equals 40, MCPs have a 10.2 mm wide, 8 ohm stripline constructed of 500 nm copper over-coated with 100 nm gold. A 4 kV, 150 ps electrical pulse provides an optical gatewidth of 80 ps and spatial resolution has been measured at 20 lp/mm.

We have improved the 0 to 99% rise time voltage on our 2 frame deflection plates from 160 nS to 65 nS with the addition of a peaking circuit that works in conjunction with our primary 2 frame deflection circuitry. Our peaking technique has applications to other HV pulsers including those which must drive 51 ohm loads. Generally, rise time voltages are measured between 10 and 90%. To minimize the camera image blur resulting from the dynamic influence of deflection plate potentials acting on photocathode electrons, it was necessary to design a circuit that would rise from 0 to the 99% voltage level in under 100 nS. Once this voltage was reached, it was necessary to stay within 1% of the attained voltage level for a duration of 1 uS. This was accomplished with the use of MOSFET solid state switching.

A new high-speed photorecorder, ITMGT-device (ionization type magnifying glass of time), is presented which unlike the well-known analogs operates in a wide spectral range of 1.3 - 10.6 micrometers and is intended to record frame by frame the beam structure of laser IR radiation. ITMGT has a wide dynamic range (not less than 300) and high sensitivity (10^-4 - 10^-5 J/cm²) in a wide spectral region. It provides frame recording with an exposure length of a frame of 10 - 50 microsecond(s) , framing rate up to 100 Hz, and up 100 frames for one recording cycle. It was successfully used to measure divergence and energy density distribution at the end of pulsed and pulsed-periodic IR lasers, in facilities for laser isotope separation and also in development of new industrial technologies for processing, welding, and cutting of materials with the help of high- power laser radiation.

Arrays of small lenses have been used in many aspects of optical recording and the purpose of this paper is to describe some examples that have been realized using lenses made at the UK National Physical Laboratory. We have produced simple lenses ranging from 50 micrometers to 400 micrometers in diameter with aperture ratios of f/1 and larger and in arrays up to 120 mm X 120 mm. The lenses are formed in layers of photoresist, making use of surface tension forces to shape the refracting surface when the resist is heated to melting point. Applications range from 3D imaging to beam steering and matching of optical arrays.

This paper describes a Dropweight-High Speed Camera system designed to measure the uniaxial compressive stress-strain properties of polymers at the strain rates appropriate to ballistic and impact events over time scales of one to tens of milliseconds. The instrument compliments existing techniques and represents a novel development of existing equipment types. It operates at strain rates in the range of 10² to 10³ and over a temperature range of 20 - 200 degree(s)C. Stress is calculated directly from a measured force obtained from a calibrated accelerometer and strain is calculated from the sample displacement measured directly using the high speed camera. The camera system has the capability of providing direct visualization simultaneously with stress information of the processes occurring during the deformation and fracture of materials and provides the ability to measure dynamic Poisson's ratios.

The goal of the present work is to determinate the limiting time resolution value for streak camera versus the recorded wavelength and to detect the photoelectron energy dispersion by processing the obtained data. In the experiments the femtosecond laser system operating in spectral range from 210 nm to 1500 nm and the Imacon-500 streak cameras equipped with the PV-001/S1 were used. Based on the received data the dependence of energy photoelectrons dispersion versus the recorded wavelength was obtained which is in a good agreement with the published data.

Prebreakdown phenomena in dielectric liquids which have various molecular structure are observed in detail for point-to-plane electrode configuration under an impulse voltage condition. Experiments are carried out by using a high-speed schlieren photography and a current measurement system by LED simultaneously. As a result, the influence of a molecular structure on the streamer propagation and the polarity effect of streamer discharge is discussed.

Purple membrane suspension prepared from strain R₁M₁ of Halobacterium halobium is excited by a doubled frequency Q-switched Nd:YAG laser with 10 ns pulse duration and 10 Hz repetitive rate. A He-Ne laser and an Ar⁺ laser are respectively used as a probing beam. The sample's response signal is detected by an avalanche photodiode and measured by a 300 MHz digitizing oscilloscope. The laser photolysis kinetics of the purple membrane is theoretically analyzed and some important equations are derived. The experimental curves are fitted by the nonlinear least square method according to the theoretical model. The kinetic parameters of the purple membrane photocycle such as decay time constants, absorption cross sections are calculated out and discussed.

Thin oriented bacteriorhodopsin films isolated from the Halobacterium halobium R₁M₁ strain are made on transparent conductive ITO glass and stainless sheet respectively by the use of electrophoretic sedimentation method and Langmuir-Blodgett technique. To construct photoelectric devices, another Cu or Al electrode is applied, and electrolytes or insulant films are sandwiched between the electrodes. A Q-switched double frequency Nd:YAG laser and a pulsed dye laser are used to stimulate the bacteriorhodopsin films. The photoelectric response signals are directly measured by a 300 MHz digitizing oscilloscope. Experimental curves are fitted through the use of Marquardt's algorithm least square method, and at least five exponential decay components are found in the whole process. A theory on the bacteriorhodopsin film photoelectric dynamics is proposed and the dynamic equations are mathematically derived. Comparison between the theoretical and experimental results gives the parameters of time constants and charge displacements in the bacteriorhodopsin photocycle.

A portable flash x-ray inspector have been developed. This inspector mainly consists of three parts, they are spiral strip line high voltage generator, cold cathode x-ray tube and power supply, control and count circuits. This instrument has wide applications. It can also be used to equip cars for field in site to study ballistics, if through some improvement.

Recently, the behavior of explosive materials under intermediate loading rates (10⁰ - 10⁴ s^-1) has become of interest. Shock and quasi-static behaviors have been well characterized, but in weapons such as a deep earth penetrator, loading conditions in intermediate strain ranges will occur previous to initiation and affect the response of the explosive. Impact tests of the explosives Tritonal and PBX-N109 were conducted at the Advanced Warhead Experimentation Facility at Eglin AFB, FL. The tests show that these materials fragment under dynamic shear loading and require high hydrostatic loads after failure to generate internal friction needed for initiation. Macroscopic, adiabatic shear localization is not a feasible initiation mechanism for these materials under these loading conditions.

A powerful, compact and repetitive nanosecond pulsed source of X-ray photons named SPHINX (Source de Photons Impulsionnelle Nanoseconde X) has been developed for a large field of applications including scientific and industrial research. The flash X-ray device is based upon high voltage fast discharge, in vacuum or in low gas pressure, obtained by a cable transformer which is powered by ceramic knob capacitors disposed in a Blumlein-like configuration and switched by a triggered spark gap. The X-ray emission from a low impedance X-ray diode with a hollow cathode - pointed anode configuration was observed under a wide range of experimental conditions (charging voltage, repetition rate, electrodes material and geometry, anode-cathode separation gap, diode vacuum pressure, and nature of gas). Doses up to 1 mR* per shot, measured at 1 m from the source, of X-rays between 5 and 100 ke V can be generated. The X-ray pulse width (fwhm) is of the order of twenty nanoseconds and the repetition rate is 50 Hz. The system can operate in air or in other gases at pressures varying from 10-3 mbar for xenon to about 1 mbar for helium. The SPHINX generator is presented in a metallic box for EMI protection and easy carrying. The volume and the weight are 0.1 m3 and 45 kg respectively, including HV power supply, pumping and air cooling systems. This source is a conventionally wall-plug table top device.

We present the results obtained during work on the recording equipment to be used in studies of the space-time structure of X-ray radiation from a short-lived Z-pinch plasma on the facility SIGNAL. The report contains description of the streak camera, its main characteristics, and methods for calibration of elements of the X-ray optical scheme. Results of experiments with high-current discharges (about 200 kA, 50 ns) up to 10 mm long show that X-ray radiation is emitted from separate small `hot' spots (micropinches), light emission of which is considerably lower than the discharge time.

A twelve-channel system is described, intended for recording the laser radiation temporal parameters and dispersion on the ISKRA-5 laser facility. These measurements are needed for synchronous delivery of all twelve laser beams at the target with a minimal difference in the arrival time. The system uses a high-speed camera and fiber-optical communication lines (FOCL). Coupling of wide-aperture 720 mm beams into FOCL with a 50 micrometers aperture has been experimentally verified. The measuring channel sensitivity (10 (mu) J at the pulse width duration 0.5 ns and laser radiation wavelength 1.315 micrometers ) allows to use the recording system over the entire facility including initial amplification stages and master oscillator. It has been shown that it is possible to make complex measurements of the temporal shape and dispersion of laser beams from all channels of the facility with an error not exceeding 25 ps. As an illustration the recorded results obtained in experiments on ISKRA-5 are presented.

Three-dimensional particle tracking velocimetry (3D PTV), a technique developed for measuring 3D fluid motion within a volumetric region, entails recording of a time sequence of stereo image-pairs of hundreds of small target-particles (flow-tracer particles) that are suspended in the fluid. The fluid motion is characterized by the 3D positions of the target particles as a function of time, information that is obtained by fully-automated processing of the stereo image sequences. Therefore, by extending the concept of photogrammetry to time varying scenes, 3D PTV can be described as transient or dynamic photogrammetry. Initially developed for measuring complex fluid motion, dynamic photogrammetry is ideally suited for a wide variety of 3D motion analysis applications. Digital imaging benefits dynamic photogrammetry by providing accurate spatial and temporal registration. As digital cameras with increasingly higher frame-rates become available, the dynamic range of the velocity measurements can be expanded, i.e., both higher and lower velocities can be measured for points of interest in the time-varying scene. This is illustrated by means of an analytical parameter study that makes use of computer- generated stereo images of target particles whose motion is prescribed to follow a 3D flow field typical of that which is generated in an engine cylinder during the intake stroke.

High-speed shadow- and schlieren photography have played a decisive role in the past decades of shock wave physics. Since many of the fundamental processes seem to be sufficiently understood and numerical techniques for hydro- dynamic simulations have advanced to versatile tool in this area, the interest has shifted towards the complexities that arise in realistic scenarios where shocks and blast waves are of importance. This study presents the results of a shock tube experiment that models the flow phenomena in a chamber subjected to a shock that impinges on the entrance to the room. Despite the simple geometry the resulting flow is comparatively complex and reflects many of the effects encountered in more realistic setups. It is dominated by multifold shock reflections passing through the jet-like inflow that soon becomes turbulent. Classical visualization techniques and pressure measurements--supplemented by numerical simulations--were used to analyze the flow field. This combination of qualitative visualization, quantitative point measurements and numerically obtained results proved to be a helpful approach to rate the influence of turbulent velocity fields onto shock wave propagation.

In order to get multi-frame holograms in high speed holography, studying the framing principle is very important. The essential way is to code reference beam which is one of the three necessary conditions to make hologram. Coding reference beam by means of a grating is available and convenient. This paper describes three kinds of grating coded reference to carry out multi-frame holography: the grating spectrum framing, the grating orientation framing, and the time delaying framing combined with the grating orientation.

A gate MicroChannel-Plate Image Intensifier (MCPII) was used as an optical shutter with a 250 ps exposure time to record the transient image of the modally dispersed output of a 30 meter length of 600 micron diameter step index optical fiber. The fiber input was pulsed light from a Raman shifted, frequency-doubled, Nd:YAG laser having a wavelength of 559 nm and a width of 20 ps FWHM. The MCPII was synchronously gated on in time to the laser pulse Time Of Arrival. The laser light from the fiber was spatially resolved with a 50 mm fl Nikon camera lens and the fourier plane was imaged onto a MCPII. The output from the MCPII was then recorded with a RS-170 video camera and a PC based video digitizer. The relative timing between the input laser pulse and the shuttering of the MCPII was precisely adjusted to capture transient images at successive times. These images reveal not only meridional rays arriving in one to one correspondence of delay time versus angle (or distance in the fourier plane) but also the time and numerical aperture broadening due to the skew rays. Indeed one 250 ps snapshot reveals delayed emission at angles that exceed the critical angle in the fiber, i.e., a picture of skew rays leaving the fiber at an angle to the z-axis greater than the meridional critical angle after all the lower order meridional rays have gone. Experimental data will be presented.

This article describes the present development of a complete fast imaging microsystem, which the main aim is to propose, for biological applications and with a low cost, a high speed camera (2400 frames per second) and its data storage system associated. This microsystem uses a multiple parallel outputs CCD image sensor and will be connected to a PC computer compatible.

We are working to realize a specific sensor for a snapshot video. This sensor will be realized in CMOS technology after resulting from different works done in GERE laboratory. These works have shown the ability of this technology to achieve photosensitive sensors.

An in-situ image sensor (ISIS) capable of recording moving images at a rate of 1,000,000 pps is proposed, with the following characteristics. (1) The CCD strips for image signal storage are elongated vertically and run below several pixels, unlike existing ISIS implementations. This unidirectional charge transfer reduces the number of layers of wiring required to drive the charges. (2) The CCD strips perform the dual roles of image signal storage and readout. In the image capture phase, the electric potential of the boundary element of each CCD storage strip is kept at a constant high value, and the charge transferred to the element is drained from the image sensor. This defines the number of elements available for storage. In the readout phase, the potential of the boundary element is controlled so as to allow charge transfer downward through the remainder of the CCD strips as in usual CCD operation. The unidirectional transfer and the dual-role CCD make the proposed sensor the simplest and, therefore, the best approach to ISIS, contributing to a higher frame rate, a greater number of sequentially-stored frames, and better quality of reproduced images. In a practical design, space is required for the gate connecting each photosensor to its CCD strip. To produce an interline sensor, a photodiode can be fabricated in the knife-shaped slit between meandering CCD strips. A micro-cylindrical lens could increase the area ratio of the photodiode to around 20%.

High-definition charge-coupled device (CCD) image sensors developed as image pickup devices for television systems so far have adopted interlaced scanning. But, it is preferable for users to have the alternative of interlaced or progressive scanning in developing applications that use image input technology for purposes other than television. This paper describes a novel CCD configuration called advanced frame interline transfer (A-FIT) that can be applied to either interlaced or progressive scanning environments by merely modifying the drive pulse of the same device. What distinguishes the A-FIT CCD imager is that, while it retains the same number of stages in the imaging area register as a CCD for interlaced scanning, it features a new storage area structure enabling the imager to support both interlaced and progressive scanning. Specifically, the vertical CCD register in the storage area is vertically partitioned down the middle to form two registers. This layout permits charges of odd-numbered lines or even- numbered lines to be shifted separately. And when the charges are transferred to the horizontal readout register, interlaced scanning results if the charges are mixed and progressive scanning results if the even and odd lines are alternated or interleaved. A prototype 2/3-inch 2-million- pixel CCD pickup was fabricated based on the novel A-FIT layout, and subjected to testing to verify its capability to perform both interlaced and progressive scanning. It was found that the CCD imager's vertical resolution was superior in progressive scan mode compared to when the chip was operated in interlaced scan mode.

A novel CCD high speed videography system with brand-new concepts and techniques is developed by Zhejiang University recently. The system can send a series of short flash pulses to the moving object. All of the parameters, such as flash numbers, flash durations, flash intervals, flash intensities and flash colors, can be controlled according to needs by the computer. A series of moving object images frozen by flash pulses, carried information of moving object, are recorded by a CCD video camera, and result images are sent to a computer to be frozen, recognized and processed with special hardware and software. Obtained parameters can be displayed, output as remote controlling signals or written into CD. The highest videography frequency is 30,000 images per second. The shortest image freezing time is several microseconds. The system has been applied to wide fields of energy, chemistry, medicine, biological engineering, aero- dynamics, explosion, multi-phase flow, mechanics, vibration, athletic training, weapon development and national defense engineering. It can also be used in production streamline to carry out the online, real-time monitoring and controlling.

"Model 3H-2100 CCD High Speed Videography System" is developed by Zhejiang University recently. This system has high videographic frequency, strong "freezing" ability, large range of videographic parameters that can be controlled according to needs. The problems of measuring moving object have been solved with quick, online, real-time and automatic method by this system. Principles of the system are introduced briefly; Successful experiments of the system such as flying bullet test, target hitting test, rotation test are illustrated; Problems of choosing videographic parameters, searching factors influencing image resolution and measuring presicion are discussed. Potential application fields are also introduced in this paper. Keywords: High speed videography, video graphic frequency, freezing ability, bullet testing, rotation testing, image processing

A high-power, high-energy laser is used to accelerate miniature metal plates (3 mm diameter X 50 micron thick, typ.) to high velocities of perform 1D impacts. Several plate conditions are critical to their performance. The plates must be flat and intact, and traveling at a known velocity and acceleration profile. Because the plates are small only optical diagnostic techniques can be used without perturbing the plate are small only optical diagnostic techniques can be used without perturbing the plate performance. We use pulsed laser stereo-photography to determine plate integrity and flatness, and velocity interferometry (VISAR) to evaluate plate acceleration and terminal velocity. Laser-launched plates can accelerate to over 10⁹ m²/s² and reach terminal velocities over 5 km/s. The high accelerations require temporal resolution of < 0.200-ns. The temporal resolution is accomplished by transferring the VISAR optical signals by optical fiber to an electronic streak camera for recording. Near 1D plate impacts are accomplished by converting the spatial Gaussian laser beam to a Tophat spatial profile. The pulsed laser stereo-photography system consists of a 10X stereo camera and a 5-ns pulsed doubled Nd:YAG laser for the light source. By incorporating these different optical techniques a complete understanding of the dynamic performance of miniature plates in flight is accomplished.

Miniature plates are laser-launched with a 10-Joule Nd:YAG for 1D impacts on to target materials much like gas gun experiments and explosive plane wave plate launch. By making the experiments small, flyer plates (3 mm diameter X 50 micron thick) and targets (10 mm diameter X 200 micron thick), 1D impact experiments can be performed in a standard laser-optical laboratory with minimum confinement and collateral damage. The laser-launched plates do not require the traditional sabot on gas guns nor the explosives needed for explosive planewave lenses, and as a result are much more amenable to a wide variety of materials and applications. Because of the small size very high pressure gradients can be generated with relative ease. The high pressure gradients result in very high strains and strain rates that are not easily generated by other experimental methods. The small size and short shock duration (1 - 20 ns) are ideal for dynamically measuring both strengths of micron-thick coatings. Experimental techniques, equipment, and dynamic material results are reported.

We have previously reported our observations of the dynamic behavior of laser driven plates. Recent improvements and modification of the imaging techniques have identified and provided measurements of Raleigh-Taylor (R-T) instabilities that occur in these events. The microscope system in the LLNL Micro Detonics Facility, was converted to an epi- illuminated polarization configuration. A double pulse nanosecond illuminator and a second independently focusable frame camera were also added to the system. A laser driven plate, that is a dense solid driven by a laser heated, lower density plasma, is inherently R-T unstable. The characteristics and growth of the instability determine whether or not the plate remains intact. In earlier reports we correlated the surface patterning of thin plates with the fiber-optical transmission modes. In subsequent experiments we noted that the plasma burn through patterning in thin plates and the surface patterning of thicker plates did not correspond to the thin plate early time patterning. These observations led to the suspicion of R-T instability. A series of experiments correlating plate thickness and pattern spatial frequency has verified the instability. The plates are aluminum, deposited on the ends of optical fibers. They are launched by a YAG laser pulse traveling down the fiber. Plate velocities are several kilometers per second and characteristic dimensions of the instabilities are a few to tens of microns. Several techniques were used to examine the plates, the most successful being specularly reflecting polarization microscopy looking directly at the plate as it flies toward the camera. These images gave data on the spatial frequencies of the instabilities but could not give the amplitudes. To measure the amplitude of the instability a semi-transparent witness plate was placed a known distance from the plate. As above, the plate was observed using the polarization microscope but using the streak camera as the detector. Both the launch of the plate and its impact into the witness plate are observed on the streak record. Knowing the plate velocity function from earlier velocimetry measurements and observing the variations in the arrival time across the plate, the amplitude of the instability can be calculated.

The shock compression behavior of LiF has not been well studied in the pressure region between several and several 10s of GPa. In this study, the Hugoniot-measurements experiments were performed on a LiF single crystal for the <100> and <110> axis orientations in such pressure region by means of the inclined-mirror method combined with a powder gas. As a result, anisotropic Hugoniot-compression behaviors were observed. The shock front profiles showed an anomalous multi-wave structure, which were consistent with the VISAR measurements. The final Hugoniot points along the <100> axis roughly coincided with the static compression curve, but, those along the <110> axis were situated in the higher stress region than those of the <100> axis and converged to them at up to 15 GPa. The Hugoniot data in the higher pressure region coincide with those obtained by the flash-gap method.

The optical measurement system using long-pulsed lasers were constructed for shock-wave measurements and spectroscopy under shock compression of solids. We have produced two types of long-pulsed lasers with no Q-switch for such purposes: Nd:YAG frequency-double laser using an intracavity KTP crystal, and dye laser using a rhodamine 6G. They consisted of doubled-elliptical pump cavity, two zenon flash lamps, and a high-voltage electrical-pulse source. The former one can be used as a light source for a Fabry-Perot type Interferometer (FPI), and the latter one can be used as a constant light source for a luminescence or an absorption spectroscopy and for the inclined-mirror method. The inclined-mirror Hugoniot measurements of some materials were performed by using the long-pulsed dye laser and the mirror- rotating type streak camera. The time resolution was increased by using a narrow width slit and the laser. The velocity-interferometer system for shock-wave measurements using a FPI and the time-resolved optical spectroscopy system using a spectrometer were constructed combined with the Nd:YAG frequency-doubled laser and the dye laser, respectively, and with an image-converter streak camera.

It has recently been established that the porosity of surfaces can have a significant effect on the reflection geometry of shock waves, and thus on the loads that are generated. This paper describes a comprehensive series of tests on the patterns of shock wave reflection from a surface covered with a series of slits, over the full range of angles of incidence from glancing to normal. The use of double-pulse interferometry is shown to be ideally suited to the study of complex compressible flow fields of this type, not only because of the high resolution but also because the tracking of fringes gives a very clear indication of both the general flow field as well as the fine structure, and thus helps clarify the mechanisms whereby the interaction process is modified from the case of reflection off a plane impervious wall. These features of the method allow a number of effects to be established which have not previously been evident, or are in conflict with the assumptions of previous studies. Specifically it is shown that the inflow angle relative to the plate is almost constant (at about 17 degrees) for shock incidence angles from zero to about 50 degrees and that the flow leaves the plate almost normal to the surface; although there is a slight drift in a direction opposite to that of the shock induced flow. Furthermore it is shown that many of the flow variables, and specifically the inflow velocity, exhibit a maximum in the vicinity of transition from regular to Mach reflection. An analysis of the motion of the acoustic waves generated from the lips of the perforations allows estimates to be made of the constancy of the inflow along the surface. These waves are convected with the flow and as they do not meet the wall at a right angle show that the convection is towards the wall. Using this method it is found that the inflow is constant behind the reflected wave in regular reflection but variable for the case of Mach reflection. Photographs of the region under the plate clearly show how the wavelets emerging from the perforations coalesce to form a transmitted wave. Measurements of this wave enable the pressure below the plate to be assessed and thus to obtain an estimate of the pressure drop across the plate. A striking feature on the underside of the plate is that the contact surface separating flow that was initially above the plate from that engulfed by the transmitted shock is made up of a string of vortices arising from the wave diffraction into the perforations.

For studying the diffuse reflectivity of tungsten target surface, a series of experiments of symmetric flat impact between copper flyer and tungsten target has been done on a two stage light gas gun. Argon ion laser with wavelength 514.5 nm was used as light source to illuminate testing point. The laser light scattered from target was collected by a lens and the light power fore-and-after shock compression was measured. The reflectivity of shock- compressed tungsten target surface could be derived through the power of scattered laser light. In these experiments, the shock pressure in tungsten target was limited from 85 GPa to 250 GPa, five pressure points being chosen. The experimental results show: shock compression makes the reflecting ability of target surface lower, and when the shock pressure is higher than 165 GPa, the reducing speed apparently becomes lower. In addition, the experimental conditions are described in this paper, and the results also are discussed.

Double exposure holographic interferometry and high speed laser shadowgraph photography and videography are used to investigate the mutual reflection of two plane shock waves. Normally research on the transition from regular to Mach reflection is undertaken by allowing a plane shock wave to impinge on a wedge. However due to the boundary layer growth on the wedge, regular reflection persists at wedge angles higher than that allowed for by inviscid shock wave theory. Several bifurcated shock tubes have been constructed, wherein an initially planar shock wave is split symmetrically into two and then recombined at the trailing edge of a wedge. The plane of symmetry acts as an ideal rigid wall eliminating thermal and viscous boundary layer effects. The flow visualization system used needs to provide high resolution information on the shockwave, slipstream, triple point and vortex positions and angles. Initially shadowgraph and schlieren methods, with a Xenon light source, were used. These results, while proving useful, are not of a sufficient resolution to measure the Mach stem and slipstream lengths accurately enough in order to determine the transition point between regular and Mach reflection. To obtain the required image resolution a 2 joule double pulse ruby laser, with a 30 ns pulse duration, was used to make holographic interferograms. The combined advantages of holographic interferometry and the 30 ns pulse laser allows one to obtain much sharper definition, and more qualitative as well as quantitative information on the flow field. The disadvantages of this system are: the long time taken to develop holograms, the difficulty of aligning the pulse laser and the fact that only one image per test is obtained. Direct contact shadowgraphs were also obtained using the pulse ruby laser to help determine triple point trajectory angles. In order to provide further information a one million frames per second CCD camera, which can take up to 10 superimposed images, was used to obtain multiple focussed shadowgraphs. Although limited resolution is obtained, due to the low resolution of the camera, information is obtained about the time evolution, and validity of the self similar assumption, of the shock wave structure. This paper highlights the practical implementation of, and the results obtained, using the above mentioned techniques in order to further explain the transition from regular to Mach reflection, as well as to describe the interaction of unsynchronized shock waves at the apex of a wedge. The advantages and disadvantages of each system are discussed as well as the benefits of using these different optical systems in conjunction with each other, to obtain a more complete description of the shock wave interaction.

The use of motion picture cameras and film in analyzing moving objects, particular] high speed motion, has played an important role in the access of information. Not only does film furnish a picture with high resolution, it is relatively inexpensive and easy to use. Various situations where film has been used successfu include missile firings, sled track runs, rocket lift-offs, automobile crash studie radiology studies, sports analysis, with emplasis on football, and various industri motion applications. More specifically, in research applications, the study of Met Attitude, Failure, and Position. This presentation is directed primarily to the history and development of analytica projectors used in viewing motion picture film with emphasis on high speed moving pictures in the 100 to 10,000 frames per second mode. Various types of film have been used. Cameras such as Fairchid, Fastax, PhotoSonics, Askania, Red Lake, Milliken, Bell Howell, Eastman Kodak, etc.

This paper deals with what is known as photo/optical instrumentation technology and/or technical photography. In 1940 this was called photography, in the late 40's the Civil Service Commission introduced a new classification called photography/technical to differentiate between still photographers and those engaging in recording engineering data. In October of 1958 a historic event took place, Congress transferred all of the duties of NACA to a newly formed agency called NASA, and with it came a call for systems that would keep up with new requirements. There was a need to change the type and style of equipment to keep up with the demands for more accurate information. Existing hardware was modified and new hardware was developed and designed to meet the new requirements of space travel of manned and unmanned orbital vehicles. This family of equipment had to withstand the rigors of space travel such as extremely high `G' forces, temperature changes and `O' gravity, while on earth we needed equipment to document launch of space vehicles as well as wind tunnel testing, rocket sled stands etc.. Some requirements were similar to those of launch vehicles, some were totally different and had other requirements, eventually they were all resolved. As electronic data systems became available NASA experimented with their use in data acquisition. This portion of this session will discuss the changes over the years and their effect on the acquisition of data, those that worked, as well as those that were a disappointment.

Field deployable, high frame rate visible CCD camera systems have been developed to support the Test and Evaluation activities at the White Sands Missile Range. These visible cameras are designed around a Sarnoff 1024 X 1024 pixel, backside illuminated CCD with a 32-port, split-frame transfer architecture. The cameras exploit this architecture to provide selectable modes from a 30 Hz frame rate at 1024 X 1024 pixels to a 300 Hz frame rate with 1024 X 512 pixels (2:1 vertical binning). The cameras are configured with a 500 mm, f/4 lens, and a Ferro-electric liquid crystal electro-optic shutter, to provide variable integration times from 0.5 to 32 msec. Video outputs provided are RS170 analog video in a reduced 512 X 480 pixel format, and 12-bit full resolution digital video data stream provided through a high speed serial/parallel digital coaxial interface. At a frame rate of 300 frames per second, these cameras deliver video data at an average rate of 1.9 Gbits/sec, and a burst rate of 2.8 Gbits/sec, with the capability of reaching an average 12 bit digital data rate of 3.8 Gbits/sec when higher frame rate imagers become available.

Sandia brought its first laser tracking system on-line in 1968 to replace the fixed camera technique for producing trajectory data on test vehicles. Today, test operations are supported by two mobile self-contained tracking systems. These standalone systems provide accurate time-space- position information and high speed photometric coverage.

There has been much study of the penetration of semi- infinite and finite thickness targets by long rods at normal incidence. The effects of oblique impact have received relatively little attention and techniques of modeling are thus less developed. It was decided to conduct an experimental investigation of the effects of rod penetration at various angles of impact at zero yaw. The rods were mounted in a reverse ballistic configuration so that their response could be quantified through the impact. Scale copper, mild steel and tungsten alloy rods with hemispherical ends were suspended at the end of the barrel of a 50 mm gas gun at the University of Cambridge. The rods were instrumented with embedded manganin piezoresistive stress gauges. Annealed aluminum, duraluminum and rolled homogeneous armor plates of varying thickness and obliquity were fired at the rods at one of two velocities. The impacts were backlit and photographed with an Ultranac FS501 programmable high-speed camera operated in framing mode. The gauges were monitored using a 2 GH s^-1 storage oscilloscope. Rods and plates were recovered after the impact for microstructural examination. Additionally, penetration of borosilicate glass targets was investigated using high-speed photography and a localized Xe flash source and schlieren optics. Additional data was obtained by the use of flash X-ray. Waves and damage were visualized in the glass. High-speed sequences and gauge records are presented showing the mechanisms of penetration and exit seen during impact.

An experimental investigation of shear localization at the tip of a notch is reported. The initiation and propagation of shear localization in two aging conditions of 300 maraging steel is recorded using ultra-high-speed photography with the goal of discovering more about; (1) the shear fatigue susceptibility of the materials, and (2) the transition from shear failure to tensile failure. These two areas are identified as important because shear localization as a failure mechanism requires, first, that the material be susceptible to such a localization, second, that the localization be dominant over other modes of failure and, last--due to boundary conditions in specific problems--that the shear localization propagate from one point into another. In reference to the first topic, the fundamental issue is whether shear localization susceptibility can be measured at all. In this work shear localization susceptibility is qualitatively investigated indicating that peak-aged 300 maraging steel is more susceptible than under- aged. The final failure within the shear band is characterized through ultra-high-speed observation and post- mortem examination. Propagation is observed and characterized by impacting side notched plates while observations are made using high speed photography at framing rates of 480,000 fps. Shear failure is seen to propagate at 1000 m/s in peak-aged material and 200 m/s in under-aged material. The peak aged material fails fully by shear while the shear failure in under-aged material arrests and is followed by tensile failure. Finite element modeling is used to determine the nature of elastic wave propagation in the specimen.

A method of angle-multiplexing holography is applied to develop high-speed holographic microscopy, which can take three successive photographs of a fast propagating crack. The crack speed is at several hundred m/sec. The optical system for holographic recording has three pulsed ruby lasers as light sources. When the crack is propagating in the observation area in a plate specimen, the three ruby lasers oscillate successively. The time interval of the laser oscillation is about 1 microsecond or longer. Varying the angle of incidence of reference beams, the optical system records the crack as three holograms which are superimposed on a holographic plate. Illuminated with a He- Ne laser beam, the three holograms reconstruct for three crack images. Lens assisted holography with spatial filters are utilized to reconstruct the three crack images separately. The spatial resolution of the reconstructed images is more than 120 lines/mm. Along a crack, one can measure the crack opening displacement (COD), that is of the order of ten microns. The dynamic stress intensity factor can be obtained from the COD measurement. The high-speed holographic microscopy is useful to study the rapid fracture phenomena in microseconds.

Model 3H-2100 High Speed Videography System developed by Zhejiang University has been applied successfully in the field for capturing flying bullet and obtaining flying bullet moving parameters with real-time image processing. While a flying bullet is detected by synchronous device, the system is in the condition of delaying time which can be controlled according to needs. As soon as the flying bullet enters visual field, a series of color flashes for `freezing' bullet images will be given out and the images captured by CCD videography camera will be sent to the computer. Obtaining moving parameters of a flying bullet with image processing and image recognition technique is illustrated chiefly in this paper. After the digital subtraction, binarization and morphological filtering, chain code tracing technique will be adopted to extract characteristic value for image recognition. For chain code tracing, image region will been segmented and the top, the end and the center coordinate of every flying bullet image will be obtained. When these data have been processed by a special software, some moving parameters such as velocities, acceleration, angles of elevation etc. can be obtained in real time.

The power of laser emission is very important for Doppler velocimeters. Doppler velocimeters with high power laser allow us to investigate structure of shock and detonation waves in a nanosecond region and to perform multichannel diagnostics. Two types of single mode lasers are described in this work. High power iodine laser (equals 1.315 micrometers ) of about 2 kW power and impulse duration of about 5 - 10 microsecond(s) . Ruby laser of 150 W power and of 200 microsecond(s) impulse duration. The width of spectral emission is 3.10^-3 cm^-1 for both lasers. Two interferometer systems, based on these lasers were made: Fabry-Perot interferometer and ORVIS. Also this paper describes the examples of application of interferometer systems with powerful lasers.

In this work we employed two-wave optical pyrometer to record characteristics of light emission and brightness temperature in shock and detonation waves. Light is transmitted to photoelectronic multiplier over a fiber light conductor with its end located in the immediate neighborhood of a light source. Error in brightness temperature determination does not exceed value (Delta) T equals +/- 5%. Limit space resolution of the technique is X equals 50 micrometers at brightness temperature T equals 2500 K. Time resolution reaches value (Delta) t equals 1 ns. The paper presents data on character of luminescence and brightness temperature of detonation front in plastic high-explosive, nitromethane, mixture of trinitrotoluol with RDX, and HE basing on TATB. This has been demonstrated that existence of micropores with air in solid HE because of intensive light flashes can exceed integral brightness of detonation front luminescence significantly.

This paper describes the unconventional method of pulse radiography, which allows to receive some (up to 10) images of different phases of high-speed process using one X-ray source. The images are recorded on a single X-ray film by some equivalent channels. Recording of different phases of high-speed process is performed by increase of X-ray film sensitivity under the effect of pulse electric field. The images, which were recorded in one experiment concerning high-speed deformation of metal body, are presented in the paper to illustrate this unconventional method of pulse radiography.

At research of high speed gas-dynamic processes x-ray image recording method is widely applied. The informativity of this method is defined by characteristics of the x-ray image formation system, which consists from x-ray or gamma- radiation source and a system of images recording. Application of developed in RFNC VNIIEF electron-optical recorders of the x-ray images has allowed to realize multi- frame recording of gas-dynamic processes and to expand methodical opportunities of x-ray imaging technique. The ways of technical realization of the high speed electron- optical recorders for single and multi-frame recorders for single and multi-frame recording are stated. Experimental results of check and the main characteristics of the electron-optical recorders are discussed. It is shown, that the sensitivity of such recorders can be changed in very wide range up to limiting, when each absorbed gamma-quantum is registered. At present in RFNC VNIIEF the electron- optical recorder `FIAR' is developed and widely applied. The result of four frames recording of plasma jet development on an output of plasma covering device also presented as an example of the recorder work. Prospects of development and perfection of characteristics of the electron-optical recorders and expansion of their application field are considered. An opportunity of creation of the x-ray image recorder with sensitive field above 4 square meters, enabling to receive the information on researched object in real time mode is shown. The design of such recorder permits fast adaptation for the requirements of gas-dynamic experiment.

The primary mission of the Holography Laboratory organized at the Opto-Physical Research Institute (YNIIOFI, Moscow, Russia) in 1966 was application of holography techniques in metrology [1]. The following basic tasks have been successfully accomplished [2-4]: (i) theoretical analysis of sources of measurement errors; (ii) development and batch production of optical holographic devices and reference instruments for their legal certification; (iii) development of methods and means for processing measurement data; specifically, the Radon transformation is applied for calculation of spatial distribution of refractive index in transparent objects; (iv) a method and device for stroboscopic measuring of 3D reconstructed images by a "periscope" technique; (v) utilization of holographic tomography and laser interferometry; (vi) instrumentation for microwave and ultrasonic holographometry, correlation-based techniques for rejection of imperfect products and for analysis ofpulsed signals. The developed holographic instrumentation has found widespread application at various scientific and industrial enterprises in the former Soviet Union and Eastern European countries. This report represents only a small selection of the total body of work accomplished in the laboratory.

The distinction between exposure time and sample rate is often the first point raised in any discussion of high speed imaging. Many high speed events require exposure times considerably shorter than those that can be achieved solely by the sample rate of the camera, where exposure time equals 1/sample rate. Gating, a method of achieving short exposure times in digital cameras, is often difficult to achieve for exposure time requirements shorter than 100 microseconds. This paper discusses the advantages and limitations of using the short duration light pulse of a near infrared laser with high speed digital imaging systems. By closely matching the output wavelength of the pulsed laser to the peak near infrared response of current sensors, high speed image capture can be accomplished at very low (visible) light levels of illumination. By virtue of the short duration light pulse, adjustable to as short as two microseconds, image capture of very high speed events can be achieved at relatively low sample rates of less than 100 pictures per second, without image blur. For our initial investigations, we chose a ballistic subject. The results of early experimentation revealed the limitations of applying traditional ballistic imaging methods when using a pulsed infrared lightsource with a digital imaging system. These early disappointing results clarified the need to further identify the unique system characteristics of the digital imager and pulsed infrared combination. It was also necessary to investigate how the infrared reflectance and transmittance of common materials affects the imaging process. This experimental work yielded a surprising, successful methodology which will prove useful in imaging ballistic and weapons tests, as well as forensics, flow visualizations, spray pattern analyses, and nocturnal animal behavioral studies.

The purpose of the development of the Airborne Separation Video System (ASVS) digital imaging system for the tri- services is the replacement of high-frame rate 16 mm film cameras for airborne weapons separation testing. Until recently, this substitution was hindered by the lack of an available compact and rugged digital camera system that could offer an adequate combination of resolution and frame rate performance. The ASVS, however, encompasses all of the requirements for high-frame rate digital imaging. Concurrently, it provides a complete solution that effectively manages the large quantities of image data created in a high-speed, high-resolution digital imaging system, while offering cost and schedule benefits to airborne and related test applications. This paper discusses the class of applications similar to that supported by the ASVS that uses synchronized, multiple, high frame rate cameras for image analysis. A set of characteristics for these applications is identified to provide an effective digital multi-camera imaging system infrastructure. The components and operation of the ASVS are detailed, along with their relation to an imaging system infrastructure solution. the ASVS's adaptation to range applications, and additional capabilities that digital imaging provides over film cameras for range applications, are discussed.

We describe a new family of ultra-high frame rate CCD image sensors. The area array devices employ an interline transfer architecture with two phase horizontal and vertical clocking, vertical antiblooming, and high-speed output buffers. Two formats have been designed to accommodate the needs of high-speed imaging applications; a 1024 X 1024, 1000 frame-per-second and a 256 X 256, 12,000 frame-per- second version. Design details and preliminary performance data are included.

Dynamic observation of impact phenomena is extremely important for designing automotive and armored structures. A dynamic moire interferometer has been developed at the INEL to image surface displacement fields generated by dynamic loading events. A pulsed ruby laser is used to provide sufficient intensity to allow the imaging of displacement fields in periods of 20 ns. An electro-optic Q-switching system has been designed using standard electronic and optical components to pulse the ruby laser at rates of up to 1 Mhz. In order to capture images, a Cordin high speed framing camera was integrated into the system. The dynamic moire interferometer produces fringe fields of in-plane displacement data, similar to the fringe data captured in dynamic holography or dynamic photoelasticity. Dynamic loading events generate moving fringe fields which reduce the fidelity of acquired fringe fields. A dynamic fringe analysis similar to Neuman's analysis for dynamic holography and Dally's analysis for dynamic photoelasticity has been performed to quantify this effect for a moire interferometer. It has been determined that the contrast of the fringes depends not only on the pulse duration of the laser and transience of the fringe field, but also on the frequency of the fringe field. This analysis has been used to interpret fringe data obtained from short duration stress pulses traveling through laminated carbon-fiber epoxy composites.

Advanced in solid state detectors and dense scintillators make radiographic cameras competitive with film in applications where the SNR is dominated by quantum statistics. In addition, these cameras offer the advantage of sub-microsecond time-gating required by present and future radiographic facilities (e.g. PHERMEX, and DARHT). We compare theoretical considerations in camera systems incorporating lens, and fiber coupling to MCPs, and CCDs. We also present Monte-Carlo blur calculations for various candidate scintillators and discuss the relative merits of the PHERMEX camera design.

Bechtel Nevada, in collaboration with Los Alamos National Laboratory, has designed a radiographic imaging system that takes advantage of large format electron optical elements to produce a highly sensitive system for large diameter radiographic fluxes. Using specially designed fast lenses, the system is able to observe scintillator screens as large as 300 mm in diameter.A gated microchannel plate intensifier allows the system to be synchronized to pulsed gamma, proton and neutron sources of radiation to help reduce background noise levels. The entire system is deployed in a transportable housing with sealed heat exchanger and electrical patch panel that is designed to be lighttight so that the electron optics can be operated at extremely high gain. External controls allow manipulation of system gain, gate width and focus. The resolution is about 1 to 2 line pairs per millimeter at the radiation-to-light converter, and the f-number of the optical system is f/1. The image is digitized from a fiber-optically coupled 1024 X 1024 cooled charge-coupled device array. The system will have interchangeable components so that system performance can be optimized to meet specific recording requirements. The major trade-off is between field of view and resolution.

The construction and the characteristics of recent high- speed soft x-ray generators designed by the authors are described. The flash x-ray generators having cold-cathode radiation tubes are three types as follows: (1) soft generator utilizing an ignitron, (2) plasma generator for producing high-intensity characteristic x rays, and (3) water-window generator having a high-durability fermite capillary. In general, when we employed the flash x-ray generators with diodes, the pulse widths had values of less than 200 ns. Next, the x-ray duration was almost equivalent to the durations of the tube voltage and current during their damped oscillations when the water-window generator was employed. The maximum tube voltage was increased up to 100 kV, and the tube currents achieved with high-intensity generators were more than 10 kA. In order to obtain kilohertz-range repetition rates, we have developed two types of stroboscopic x-ray generators having hot-cathode tubes as follows: (4) low-photon-energy generator utilizing and triode and (5) high-photon-energy generator with a diode. As the duration was controlled in a microsecond range by using the low-photon-energy generator, sufficient x-ray intensifier for the normal radiography were obtained. The maximum photon energy could be increased up to about 200 keV by the high-photon-energy generator having a double transformer. Using these generation, we performed high-speed soft radiography.

A novel subpicosecond x-ray streak camera (called PX1) was developed by the INRS group for subpicosecond time resolved spectroscopy in x rays and X-UV range. Using the PX1 camera, we have measured keV x-ray pulses with a 950 fs FWHM and a 850 fs rise time. The camera has also been coupled to ultrafast photoconductive switches and tested in jitter-free mode as a signal averaging detector. This instrument allows to analyze ultrafast changes in short wavelength signals with an unlimited dynamic range.

Motion camera Focusing Schlieren photographs were obtained for test vehicles at the high speed test track at Holloman AFB. These photographs were used to examine the shock waves and turbulent flow generated by a variety of rocket driven sled vehicles at speeds from Mach 1.3 to Mach 5.5. Some of the images show non-uniform exposure, but the shock wave structures are still clearly seen. An improved light source setup is described which should improve the uniformity of the exposures.

This paper describes the photoinstrumentation employed to record the blast and fragmentation effect on a warship resulting from the on-board detonation of various explosive warheads. Details are presented of cameras, camera protection, lighting and associated instrumentation that were used in this harsh environment. Good quality results were obtained which provided scientists with a much better understanding of the damage processes involved. In particular the flashbulb lighting and camera protection equipment developed for the trial proved very successful.

The ULTRANAC Ultra-High Speed Framing and Streak Camera System, from Imco Electro-Optics Limited of England was first presented to the market at the 19th ICHSPP held in Cambridge, England, in 1990. It was the world's first fully computerized image converter camera and is capable of remote programming at framing speeds up to 20 million fps and streak speeds up to 1 nS/mm. The delay, exposure, interframe and output trigger times can be independently programmed within any one sequence. Increased spatial resolution is obtained by generating a series of static frames during the exposure period as opposed to the previously utilized sine wave shuttering technique. The first ULTRANAC was supplied to Japan, through the parent company, NAC, in 1991. Since then, more than 40 cameras have been installed world-wide. The range of applications is many and varied covering impact studies, shock wave research, high voltage discharge, ballistics, detonics, laser and plasma effects, combustion and injection research, nuclear and particle studies, crack propagation and ink jet printer development among many others. This paper attempts to present the results obtained from such tests. It will describe the methods of recording the images, both film and electronically, and recent advances in cooled CCD image technology and associated software analysis programs.

The laser cladding process consists in adding melting powder to a metallic substrate. The process usually involves the delivery of fine grained alloy powder blown into the melt pool during the laser irradiation. The temperature of the grains when they reach the surface and their speed within the laser beam are extremely correlated and influence on the quality of the coating. We present three methods which can be used to measure the speed of the particles in the powder stream; they all make use of digital images obtained with a standard matrix CCD camera and a strobed lighting. The first method, which we call the basic one, allows speed measurements in a slim cross section of the powder stream. Thanks to its low requirement in calculation amount we were able to implement it in a strictly software design for real time application (1 image/20 ms). In the second method we work on the Fourier transform plan of a strobed image, between two flashes the particles have moved from the same distance this periodic distance represents a specific frequency which is estimated in the Fourier space. In the third method, we strobe the image using three color filters and obtain a RGB image. The autocorrelation function of the image provides us with information related to the speed of the particle. The particularity of the two last methods is their complementarity. Indeed, the FFT gives good and averaged results for wide area, it is a large scale method. At the opposite, the three colors imaging method is powerful for small scale study.