Uncertainty of digital image correlation with vibrating deformable targets

Abstract

Digital Image Correlation (DIC) is a full-field, non-contact optical technique to measure the contour, deformation, vibration and strain on the surface of an object. Studies on this method and its uncertainty are mostly focused on motion in static conditions. In dynamic conditions, as the target is moving with respect to the camera, the images acquired during the experiments have a considerable amount of motion effect (blurring). This motion effect is the main cause of uncertainty in measurement with DIC method. This work experimentally investigates the effect of the shutter time and the motion of the target on the uncertainty of DIC method. Tests were done on an aluminum cantilever beam, with first natural frequency of 7.6 Hz and a random speckled pattern on its surface. An 8 MP camera acquired the decay free vibration of the full length of the beam (0.93 m) with 50 frames per second while a laser triangulation sensor was used as a reference transducer in a section of the beam. Effect of exposure time, sensor gain and vibration amplitudes on measuring the uncertainty is explored. A software was developed to synchronize the laser and camera, while DIC analysis was done with Vic-2D software. The results show that the uncertainty of DIC is proportional to the product of the exposure time and the initial displacement of the beam (i.e. the stripe length). Afterwards, deconvolution method, which is a processing technique to estimate the displacement of an object and the motion effect that exist in an acquired image, was used to improve the results in case of having a deformable target in dynamics conditions.