Bio mechanics cinematography laboratories, by their nature require effective light levels, versatility in equipment selection and positioning, and intensity control by both mechanical and electrical means. Comfort and safety is a prime consideration in the selection and placement of equipment. Because power services have to be rated for maximum utilization of lighting loads,electrical feeds must be sized for the maximum lighting load. The calculation of power factors based on the efficiency of continuous duty tungsten halogen lighting equipment require an understanding of the efficiencies of this equipment. By establishing the wattage per square foot demand based on maximum intensity values, the appropriate electrical services may be determined and distribution equipment properly selected. Photographic lighting techniques designed to enhance the data quality are relatively simple and seldom used. Appropriate positioning and selection of lighting equipment to improve the image quality is discussed.

Several trials of many different gymnastics skills on various pieces of apparatus were filmed and the results were studied with the coach. The time to accomplish the entire skill as well as the time for each segment of the skill was important to the coach. He was also interested in angle of release or push-off and the path of the center of gravity. Lastly, graphs of velocities and accelerations of limb segments were revealing to the coach. Biomechanical analysis has helped him see why the performances were good; he is more interested in working with the investigator in all the events in gymnastics through the medium of cinematography.

Human Engineering involves a systematic approach to workspace design, and is predicated upon the concept that it is better to modify equipment in a workspace to meet the con-straints of users, rather than to modify the actions of users to meet the constraints of equipment. In accordance with Human Engineering practices, the first step in the develop-ment of a workspace is to state the purposes of the workspace. Subsequently, the equipment and personnel which will be included in the workspace should be identified, and the manner in which the equipment will function within the workspace should be examined. Finally, communication links within the system should be planned. The above sequence of steps can be applied to biomechanics cinematography laboratories whose purposes are either to perform research or to provide facilities for both teaching and research.

This paper deals with the variables encountered in lighting for high speed photography. They are film exposure index, F stops, exposure time, target size and distance and frame rates. An understanding of the inter-relationships of these elements permits calculation of required lighting intensities from continuous duty tungsten halogen lighting equipment for an effective result. Selection of the appropriate lighting equipment for the lighting is discussed based on light source efficiencies, throw distances, and intensity requirements. Effects on exposure due to extension tubes or bellows are discussed. Eastman Kodak's introduction of high speed video news film has opened a new opportunity in the enhancement of data acquisition due to its high exposure index and as a result reduced lighting intensity requirements. Types of lighting equipment and their characteristics are discussed along with the basis for the selection of lighting equipment for particular high speed lighting tasks.

The characteristics of typical timing signal generating systems are discussed. These general serial time codes, usually a succession of 0 to 5 volt DC pulsed from 10 to 10,000 HZ. Such generators produce signals for Time Pulse codes, Coded Time Pulse Codes, IRIG Time and Control Code, NASA Time and Control Code and SJ1PTE Time and Control Code. These signals are often generated together with the decoder and driving current, or the signal will be inputed from outside sources, decoded and converted into drives to pulse the light emitting diodes displays in "on camera" film data annotation systems. Typical time data annotation systems include simple camera timing markers for exposing a narrow time or data track along the edge of the continuously moving film. Systems which record 9 or more digits of time and other data on each frame of film are: BCD, i.e., 4 x 9 numeric data annotation systems; 7 segment LED characters for recording numeric time or hexidecimal information; lastly, data annotation systems using 5 x 7 LED matrices to record full alphanumeric (ASCII) characters, including numeric time. The advantages and limitations of these systems are considered.

Data Acquisition Systems interface between the real world of physical parameters, which are analog, and the artificial digital world. When the analog world is recorded digitally on film or video tape, analog parameters such as speed, azmiuth, elevation, pitch/roll and time can be converted to digital information and recorded on film or video tape along with a "picture" of the event. This paper provides an overview of some of the techniques of converting the analog data to digital information for annotation purposes.

Nac Incorporated has recently introduced a new high speed color video system which employs a standard VHS color video cassette. Playback can be accomplished on either the HSV-200 or on a standard VHS video recorder/playback unit, such as manufactured by JVC or Panasonic.

Photographic acquisition of data often may be simplified, or the data quality improved upon by employing electronic flash sources with traditional equipment or techniques. The relatively short flash duration compared to movie camera shutters, or to the long integration time of video camera provides improved spatial resolution through blur reduction, particularly important as image movement becomes a significant fraction of film format dimension. Greater accuracy typically is achieved in velocity and acceleration determinations by using a stroboscopic light source rather than a movie camera frame-rate control as a time standard. Electrical efficiency often is an important advantage of electronic flash sources since almost any necessary light level for exposure may be produced, yet the source typically is "off" most of the time. Various synchronization techniques greatly expand the precise control of exposure. Biomechanical and sports equipment studies may involve velocities up to 200 feet-per-second, and often will have associated very rapid actions of interest. The need for brief exposures increases H.s one "ZOOMS in on the action." In golf, for example, the swing may be examined using 100 microsecond (Us) flashes at rates of 60 or 120 flashes-per-second (FPS). Accurate determination of linear and rotational velocity of the ball requires 10 Us flashes at 500-1,000 FPS, while sub-Us flashes at 20,000-50,000 FPS are required to resolve the interaction of the ball and the club, head. Some seldom. used techniques involving streak photography are described, with enhanced results obtained by combining strobe with the usual continuous light source. The combination of strobe and a fast electro-mechanical shutter is considered for Us photography under daylight conditions.

In this, the International Year of the Disabled, attention is directed among other areas toward rehabilitation and sports participation of wheelchair users. As an application of movement analysis in medicine and rehabilitation and as an application of sports research using biomechanics, this investigation was performed to compare the results of two methods of gathering data on the stress of wheelchair propelling at equivalent work loads and to account for differences in physiological responses with a mechanical analysis of wheelchair propelling. Physiological data collected were heart rate, systolic blood pressure, and rate-pressure product. A biomechanical cinematography analysis was used to determine external work in wheelchair propelling and to determine the extent to which modifications in segment actionsoccurred during increasing magnitude of work. A cycle ergometer was adjusted to replicate external work loads performed during wheelchair propelling. A t-test of equivalent external work loads indicated that heart rate was not different between the two exercise modes at the .05 level of significance. The t-test did indicate a significant difference in systolic blood pressure and rate-pressure product at the .05 level of significance. The biomechanical analysis of wheelchair propelling established that an increase in external work was accomplished by a decrease in the range of motion and an increase in the speed of movement. During cycle ergometry the range and speed of movement remained the same while resistance was increased. Results of the study established that while heart rate for equivalent external work loads was the same for wheelchair propelling and arm cranking cycle ergometry, systolic blood pressure and rate-pressure product were not the same. The suggestion was that some means of propelling a wheelchair other than that which is con-sidered "standard" might be considered which produces less stressful responses in wheelchair users.

A 16 mm film projection system was designed for biomechanical analysis of cinematographic images. The design was determined on the basis of an evaluation of several systems currently in use. The resulting system has incorporated many desirable features for digitizing motion picture films. These features include variable image size potential, front surface projection, fixed projector alignment, variable screen orientation and comfortable operation.

Biomechanics Cinematography data collection and analysis procedures were employed to compute the momentum of a pole vaulter throughout the performance. Simultaneously horizontal and vertical force components acting on the tip of the pole were measured in the vaulting box. Two phase-locked Photo Sonics cameras were used to record the vault and the analog force signal. The changes in momentum computed from the subject film were found to be within six percent ( 16 Ns ) of the concomitant impulses obtained from numerical integration of the force - time data.

The Department of Defense ranges have relied on photographic instrumentation for gathering data of firings for all types of ordnance. A large inventory of cameras are available on the market that can be used for these tasks. A new set of optical instrumentation is beginning to appear which, in many cases, can directly replace photographic cameras for a great deal of the work being performed now. These are television cameras modified so they can stop motion, see in the dark, perform under hostile environments, and provide real time information. This paper discusses techniques for modifying television cameras so they can be used for motion analysis.

High-speed videography has been an electronic analog of low-speed film cameras, but having the advantages of instant-replay and simplicity of operation. Recent advances have pushed frame-rates into the realm of the rotating prism camera. Some characteristics of videography systems are discussed in conjunction with applications in sports analysis, and with sports equipment testing.

Video systems can be configured to measure position, size, attitude, brightness, and color of objects including objects in high speed events. The measurements may be time correlated or images from several sources (perhaps widely separated) may be correlated directly by image splitting techniques. The composition and specifications of the video system will vary considerably depending on the parameters measured and the accuracy desired. The basis of making the above measurements, using video, are presented in a format to guide practitioners in applying video as a measuring tool. Topics include relative vs. absolute measurements, scales and references, data insertion and retrieval, human factors, and video digitization.

Dust, moisture, insects, rodents, explosives and cleaning solutions are among the major problems that cause equipment failure. There are others; improper use, abuse, aging and disasters, but this paper deals with those methods that will keep a camera system operating by properly maintaining it.

The Naval Biodynamics Laboratory in New Orleans, Louisiana is engaged in a series of experiments to measure the dynamic response of critical anatomical segments of the human anatomy to acceleration environments. A comprehensive three-dimensional cinematographic data acquisition system is employed to obtain the required trajectory data. This paper presents an overview of the entire Naval Biodynamics Laboratory system: described are the hardware employed, calibration requirements and methodology, data reduction and analysis, and timing systems.

A new high-speed motion analysis system provides 2000 frames/second recording and instant replay. Subframe pictures can be taken at 12,000 pictures per second. The recording medium is magnetic tape packaged in a cassette which stores 45 seconds at the highest speed. The key technical advance which made the system possible include a new solid-state 192 x 240 image sensor array which can be read at 108 pixels/second. Recording at practical tape speeds was achieved through the development of high density magnetic heads and tape specifically for this system. In addition to high speed operation, the system was designed to ease the entire range of image analysis tasks. Analog data channels may be recorded and reproduced synchronously with the image sequence. The instrument permits full remote control by a host computer, and random read/write access to the internal digital frame store for image enhancement and pattern recognition applications.

Lightweight targets mounted on the head and neck of human volunteers are photographed by high-speed cameras during impact acceleration tests. The targets must be capable of being tracked through a wide angular motion by at least two cameras to obtain three-dimens-ional displacement and orientation. Because the targets are tracked and digitized by a computerized photodigitizer, their pattern must be selected to maximize recognition and minimize crossover confusion. This pater discusses the target construction, orientation on the accelerometer mount, pattern selection, and paint scheme.

The Naval Biodynamics Laboratory (NBDL) in New Orleans is investigating the human response to different acceleration environments. In controlled experiments rigid clusters of targets are attached to specific anatomical segments of human volunteers and photographed by at least two high-speed cinematographic cameras. The x,y coordinates of each target are digitized and used in calculations describing the three-dimensional linear and angular motion of the anatomical segments. The digitized target coordinates contain significant broadband, high frequency noise which seriously compromises the usefulness of the data. An efficient recursive digital filter algorithm, developed to remove the noise from the data, is presented in this paper.

The Naval Biodynamics Laboratory is engaged in a series of experiments to evaluate the dynamic response of critical anatomical segments to acceleration environments. A three-dimensional phototarget configuration attached to a T-plate is acquired by at least two cameras and linear displacement and orientation parameters are derived by a least squares algorithm. Because of visibility problems, there are often changing target configurations as well as changes in the sensitivity of the solution to errors. Low frequency errors due to camera orientation, location, and calibration constants, as well as errors in the target locations on the T-plate coupled with changing target configurations can result in dis-continuous jumps in the derived kinematic variables. This paper describes an algorithm for modifying the least squares solution so as to minimize the effect of these perturbations without compromising the optimum long-time characteristics of the least squares solution. Several examples comparing solutions with and without the algorithm will be presented.

Biomechanics analysis frequently requires cinematographic studies as a first step toward understanding the essential mechanics of a sport or exercise. In order to understand the exertion by the athlete, cinematography is used to establish the kinematics from which the energy exchanges can be considered and the equilibrium equations can be studied. Errors in the raw digital information necessitate smoothing of the data before derivatives can be obtained. Researchers employ a variety of curve-smoothing techniques including filtering and polynomial spline methods. It is essential that the researcher understands the accuracy which can be expected in velocities and accelerations obtained from smoothed digital information. This paper considers particular types of data inherent in athletic motion and the expected accuracy of calculated velocities and accelerations using typical error distributions in the raw digital information. Included in this paper are high acceleration, impact and smooth motion types of data.

The Naval Biodynamics Laboratory in New Orleans is engaged in a series of experiments to measure the dynamic response of critical segments of the human anatomy to acceleration environments. A configuration of accelerometers and photographic targets is mounted on a T-plate which is fixed to the anatomical segment to be measured. The kinematic variables defining the linear displacement and angular orientation of the rigid body are derived independently from the accelerometer and photographic measurements. This paper illustrates an optimum procedure for combining the results from both sets of measurements into one consistent set of derived variables from acceleration to displacement. The method is applicable to non-contiguous photo-derived variables and allows for the high frequency resolution capabilities of the accelerometer system while discriminating against low frequency errors in these components. The method used and examples comparing the photo-derived variables, the accelerometer-derived variables, and the combination variables for actual data obtained at NBDL will be presented.

The history of Cinematography research can be described in levels of sophistication. The first level was the use of the single camera, primarily for qualitative research. The second level of sophistication was interpretive cinematography. The holistic method was the next level of sophistication, in which three-dimensional cinematography, two or more cameras, was used. Computerized cinematography was the next level. The last level of sophistication, used most extensively with humans, involved three-dimensional analysis and multiple cameras. One higher sophistication level that has been utilized with animals, rather than with humans, is that of cineradiography.

A personal history of biomechanical analyses from motion pictures is presented. The areas of work are of total body motions of athletes in many sports, analysis of work-related injuries in industry, measurements of synthetic turfs, diagnosis of muscular imbalance, and animal motion. A film presentation illustrates each of these areas.The filming and analysing of any motion should produce data that has practical applications. This means taking measured data, interpreting it properly, and then using the information to actually obtain useful results. The practical application of data obtained from the high speed filming of humans and animals is presented here.

The increased capabilities of minicomputers today allows a biomechanics laboratory to establish a self-contained computer system for a reasonable price. The system includes a microprocessor, a printer and a CRT. Analog to digital conversion is an important feature to consider as well as the ability to interface with a mainframe computer. A minicomputer adapted for film analysis should be a consideration for data analysis when developing a cinematography laboratory. For the past 10-15 years the area of biomechanics has enjoyed the advances in technology. Equipment and instrumentation once used exclusively by engineers and physicists have become readily available to those involved with snorts analyses. Among the various pieces of equipment accessible to biomechanists today, probably the most important one is the computer. At this time several biomechanics laboratories are using the computer to analyze kinematic and kinetic data obtained from film. The computer in use at each school is generally the main University or College computer with a remote terminal set-up in the biomechanics laboratory. This system functions well if there is adequate response from the time-sharing system of the main computer, and if there is at least one knowledgeable technician available. With the trend toward minicomputers today, their increased capabilities, and their ease of use, a self-contained minicomputer system in the biomechanics laboratory appears to be a viable alternative. The computer system in use in the ,Biomechanics Laboratory at the University of Oklahoma is based around the Cromemco Z2D computer connected to a PCD motion analyzer (Figure 1). The data acquisition system consists of the eight-bit microprocessor-based minicomputer connected to an analog to digital converter (ADC). As a terminal for the computer, we have either a video display unit or a Model 43 Teletype. The Model 43 provides a hard copy out-put while the video terminal provides much faster I/O, useful for debugging and program development. The computer itself consists of the high current power supply mounted behind a 22 slot card cage. The CPU, 48K-byte memory, and I/O cards plug into the S100 card cage slots. The size of the power supply, in addition to the large number of card slots, give the Cromemco Z2D considerably more flexibility and expandability than more common "home computer" systems. The basic computer also includes two 51/4 inch flexible disk drives with a disk controller card capable of running four disk drives. As mentioned, one of the slots contains a card for analog to digital conversion. This particular card has seven analog input channels and seven analog output channels. Two of the analog inputs are allocated by the biomechanics program to the x and y data channels of the PCD film analyzer. In between the PCD machine and the ADC inputs, it is necessary to use a few circuits for analog signal conditioning. These circuits are used to match the 0 to 5 volt output of the film analyzer to the -2.56 to +2.54 voltage range of the ADC. In additon to the analog conversion card mentioned above, other cards available include parallel I/O, serial I/O, and TV video display drivers. The serial I/O card supports two channels of serial data which are useful for communication with other computers via a modem and output to printers. Figure 2 illustrates the configuration of this set-up. Although the cost of the Cromemco Z2D ($10,000) is somewhat higher than other computers available, the A-D conversion and extensive I/O canabilities are important features that must be considered. The system can either stand alone or be interfaced with a mainframe computer via a serial I/O port, another important asoect when time-sharing is not only expensive but difficult to obtain. A third reason for choosing this computer is its compactness; it is small enough to be placed on a moveable rack and can be rolled around to any location without the need for exoansion interfaces or additional power supplies.

The Naval Biodynamics Laboratory in New Orleans is engaged in a series of experiments to evaluate the dynamic response of critical anatomical segments. A three-dimensional array of photo-targets attached to a T-plate affixed to the anatomical segment is acquired by at least two cameras, and the kinematic variables defining the linear displacement and orientation of the rigid body are developed from a least-squares algorithm. This paper presents a method for plotting the film coordinates of each camera such that macroscopic errors in the locations of the automatically tracked targets can be easily detected and corrected. Examples of the types of errors found when the method was applied to actual data at the Naval Biodynamics Laboratory will be presented.

Three-dimensional target configurations attached to the head, neck, and pelvis of human volunteer subjects undergoing impact acceleration and vibration experiments are acquired by at least two cameras. Linear displacement and orientation parameters are derived from the digitized data film by a least squares algorithm. In order to minimize low frequency errors, it is necessary to calibrate individual high-speed cameras to determine points in the film plane. In addition, it is required to know how these cameras are located and oriented in the laboratory reference system. This paper will describe the procedure and the dedicated apparatus for camera calibration, discuss problems, and compare three surveying methods used to locate and determine camera orientation.

The Naval Biodynamics Laboratory is engaged in a series of experiments to evaluate the dynamic response of critical anatomical segments to acceleration environments. A target configuration attached to a T-plate and affixed to an anatomical segment is acquired by at least two cameras, and linear displacement and orientation components are derived from this data. The quality of these cinematographic solutions depends on the number of targets and on the geometry of the target configuration relative to the camera axis orientation, as well as on errors in camera location and orientation, errors in photo-target location on the T-plate, timing errors between cameras, and high frequency noise errors. This paper presents methods for statistically assessing these errors while obtaining the solution for the kinematic variables and presents plots of these errors for representative runs in the NBDL data base. The possibility of using the method for dynamically eliminating poor quality solutions is also discussed.

Recent developments in the modeling of multi-body system dynamics are incorporated into an integrated, computer-oriented method for analyzing human body motion. The formulation, which represents the human body as a set of 17 finite, rigid-body segments including hands, feet, arms, legs, head, neck, and upper and lower torso, also accounts for the effects of connective tissues and muscles with non-linear springs and dampers at the connections of the linked rigid-bodies. Specific application of this biomathematical modeling of the body segments includes the estimation of musculoskeletal injury potential during aircraft and land vehicular crashes. With the integration of the output dynamics of the model, the injury profiles of the occupants, and human tissue tolerance limits, a more complete analysis and reconstruction of the details of the human occupant trajectory responses and injury incurrence can be made.

The key to kinetic analysis procedures via cinematography is the determination of the kinematics of the situation under analysis. Calculation of kinematic parameters of a body, namely, its displacement, velocity, and acceleration, is dependent on the precise and accurate location of points in two- or three-dimensional space. To insure optimal precision (reliability) and accuracy (validity) of the resulting kinematic data, consideration must be given to the selection and implementation of appropriate photographic procedures, as well as to data reduction procedures that will minimize errors.

Analytical techniques for reconstructing three-dimensional trajectories from motion-picture data have been described in the literature, but a number of practical problems limit the application of these techniques. The accuracy of coordinate reconstructions is influenced by the selection of the photographic targets, the quality of the images, the time-base, and the data reduction device. State-of-the-art data recording and data reduction devices are discussed.

Cinematography is an integral element in the interdisciplinary biomechanics research conducted in the Department of Kinesiology at the University of California, Los Angeles. For either an isolated recording of a movement phenomenon or as a recording component which is synchronized with additional transducers and recording equipment, high speed motion picture film has been effectively incorporated into resr'arch projects ranging from two and three dimensional analyses of human movements, locomotor mechanics of cursorial mammals and primates, to the structural responses and dynamic geometries of skeletal muscles, tendons, and ligaments. The basic equipment used in these studies includes three, 16 mm high speed, pin-registered cameras which have the capacity for electronic phase-locking. Crystal oscillators provide the generator pulses to synchronize the timing lights of the cameras and the analog-to-digital recording equipment. A rear-projection system with a sonic digitizer permits quantification of film coordinates which are stored on computer disks. The capacity for synchronizing the high speed films with additional recording equipment provides an effective means of obtaining not only position-time data from film, but also electromyographic, force platform, tendon force transducer, and strain gauge recordings from tissues or moving organisms. During the past few years, biomechanics research which comprised human studies has used both planar and three-dimensional cinematographic techniques. The studies included planar analyses which range from the gait characteristics of lower extremity child amputees to the running kinematics and kinetics of highly skilled sprinters and long-distance runners. The dynamics of race cycling and kinetics of gymnastic maneuvers were studied with cinematography and either a multi-dimensional force platform or a bicycle pedal with strain gauges to determine the time histories of the applied forces. The three-dimensional technique implemented at UCLA is the Direct Linear Transformation (DLT) method. DLT was developed from a close-range stereo-photogrammetry method to a technique flexible and accurate for 16 mm film applications in biomechanics. The DLT method has been used to document the three-dimensional kinematics of the ball, hand, forearm, and upper arm segments of pitchers during high velocity baseball throwing. The animal research which has incorporated cinematography has focused on both normal locomotor kinematics and kinetics, as well as spinalized locomotion, to assess neural control mechanisms which regulate gait. In addition, a new technique has been developed which allows the recording of in vivo tendon forces in an animal during unrestrained locomotion; via cinematography, movements of the limbs can be correlated with both myoelectric activity and tendon forces to analyze dynamics of muscle contractions during walking, running, and jumping. An additional area in which cinematography has proven useful is in the measurement of the architectural and structural deformations and strains which occur in skeletal muscles, tendons, and ligaments. These experiments have been done both in situ and in vitro, and have included both normal functional ranges of the tissues and incidences of mechanical failure or ruptures. The use of photographic techniques in these experiments is advantageous because the tissue changes can be documented without attaching mechanical apparatus to the tissue which can introduce artifacts. Although high speed cinematography does not solve all the data collection and recording needs in an integrated approach to biomechanics, it nevertheless forms an important constituent in a comprehensive research program. The positive attributes of high speed film records outweigh the laborious and tedious data reduction techniques which are frequently necessary to achieve high quality data.

Since the formalization of the myoelastic-aerodynamic theory of vocal fold vibration, it has been generally accepted that biomechanical and aerodynamic forces determine the nature of vocal fold vibration patterns, speaking fundamental frequency and vocal intensity. The speech of the deaf is frequently characterized by abnormal voice qualities and aberrant frequency and intensity variations suggesting mismanagement of the biomechanical and aerodynamic forces acting on the larynx. Unfortunately, efforts to remediate these abnormal laryngeal activities are frequently ineffective. It is reasonable to suggest that more effective remedial strategies could be developed if we had a better understanding of the underlying nature of the problems deaf persons experience when trying to control laryngeal functioning for speech purposes. Toward this end, we are employing high speed laryngeal filming procedures in conjunction with glottal impedance, respiratory kinematic and acous-tical measurement procedures to assess abnormal laryngeal functioning of deaf speakers. All data are collected simultaneously and are time-locked to facilitate analysis of specific laryngeal events. This unique combination of instrumentation has provided important insights regarding laryngeal functioning of the deaf. For example, we have observed that deaf speakers may assume abnormal glottal configurations during phonation that pro-hibit normal laryngeal functioning and disturb upper airway dynamics. Also, normal vibratory patterns are frequently disturbed. Instrumentation, data collection protocols, analysis procedures and selected findings will be discussed.