大場 信弥 (株)日立製作所中央研究所, 主任
野口 浩 (株)ナック映像技術センタ, 部長
瀬口 靖幸 大阪大学, 基礎工学部, 教授 (20031073)
MORIMOTO Yoshiharu Osaka University, Faculty of Engineering Science, Associate Professor, 基礎工学部, 助教授 (20029573)
OHBA Shinya Hitachi Ltd., Central Research Laboratory, Manager of Mobara Office
NOGUCHI Hiroshi Nac Inc., Image Engineering Center, Senior Engineer
|Budget Amount *help
¥7,900,000 (Direct Cost : ¥7,900,000)
Fiscal Year 1990 : ¥1,600,000 (Direct Cost : ¥1,600,000)
Fiscal Year 1989 : ¥6,300,000 (Direct Cost : ¥6,300,000)
In order to study high-speed deformation of materials, high speed digital video cameras are developed by using solid-state image sensors. Three types of image sensors are used. These are a MOS type image sensor with an optical black area, a MOS type image sensor without the optical black area and a CCD image sensor with an optical black area. The frame speed and frame size are controlled by a personal computer. When a small scanning-area of the image sensors is selected, the frame speed becomes faster ; however the corresponding resolution decreases. The ideal maximum frame speed is about 12 millions frames/sec. The practical maximum speed is 420 thousands frames/sec because of noises and the wide pulse-width of the control signal. The image signal is stored in IC memories by using an A/D convertor. The image signal is sent to the same IC memories by cyclically rewriting. If a trigger is ent after the phenomenon is completed, the desired images before the trigger are easily captured. T
he recording time is about one second.
In order to measure deformation of materials, the Fourier transform moire and grid method (FTMGM), which has been developed by the authors, is used. In this method, the interpolation between fringes or grating lines is performed very naturally by using the phases of the sinusoidal brightness distribution, which is obtained from the first harmonic of the Fourier spectrum of the image of the fringes or grating lines. The process of this method is simple and completely automated, without subjective process. Therefore this analysis can exclude human errors so that accurate measurements of displacement, strain, velocity, strain rate and height are performed. In spite of double differentiation with respect to distance and time, the accuracy is high because the smoothing of data is naturally performed in the FTMGM. The program of the FTMGM is developed on personal computers and workstations.
Some experiments such as stress wave propagation in a rubber tube and vibration of a rubber plate are performed. These may be the first report measuring strain rate distribution.