The systematic strain measurement error in Digital Image Correlation (DIC) induced by the environment temperature
variation around the digital camera was extensively studied. The temperature variation of different camera components
along with the changes of the environment temperature is experimentally studied and the motions of different
components are then analyzed. The strain error in DIC is then analyzed according to a physical model to express the
imaging geometry changes. Finally, the DIC measurement error caused by environment temperature variation was
Binocular stereo vision system has been widely used in 3D shape and deformation measurement. Many experimental
systems and instruments have been developed based on this method. To realize the stereo vision measurement, the
measured surface must carry some image "structures", such as points, fringes, or even random speckles, so that the
parallax of the dual camera could be obtained by matching the "structures". However, for on-line inspection of black
surface, the general projection methods are impossible to produce the appropriate "structures". In this paper a new
binocular stereo vision system based on the laser dot matrix projection is presented. The laser projector is made up of
two pieces of orthogonal fine grating and laser diode, when the laser beam was projected to the grating, a laser dot matrix
will be formed and this high contrast dot matrix could be imaged on any surfaces, including the black surfaces.
Furthermore, the image quality is greatly improved by mounting a red color filter (if a red laser) on the camera lens.
Finally, a 3D shape measurement of a black cylinder is completed to verify the validity of the system.
This paper focuses on the effect of operating temperature of uncooled CCD on the strain measurement error when the
CCD is used to measure strain in an optical extensometer. A special experiment is designed in which an optical
extensometer is used to measure the strain and a thermocouple is used to measure the temperature of CCD. The results
show that the measured temperature of CCD has a similar tendency with the change of the measured strain. For different
types of uncooled CCD, the temperature varies about 5~10 degree from the beginning of the experiment to the end, and
the error of the measured strain induced by the temperature was about 120~180με. It is considered that the error is related
to the inherent noise of the electronic components of the CCD whose operating mode is sensitive to the temperature
change. When using the optical extensometer, it is suggested to preheat the CCD 1~2 hours before the experiment
especially in doing high accuracy measurement. Otherwise, the strain induced by the operating temperature should be
eliminated from the measurement results through conducting temperature compensation method mentioned in this paper.
A simple optical system for small rotation angle measurement is constructed in this paper. A laser diode is fixed on the
rotation part of the mechanism and the laser beam is received by a screen. The position of the spot could be registered
from the image captured by CCD camera using the gray centroid algorithm. Then rotation could be obtained from the
spot movement according to geometrical relationship. In order to verify the system and evaluate its accuracy, a
deflection test of a low carbon steel cantilever beam is performed. It shows that the maximum relative error of
experiment results and theoretical value is less than 1.5%, the measuring accuracy and stability is satisfied. The system
in this paper has many advantages, such as high resolution, low cost, great convenience and could be used to measure
small rotation angles of very complex mechanisms under non-contact request.