Digital image correlation techniques in dynamic application on deformable targets

Digital Image Correlation (DIC) is a widely used optical measurement method to track the motion of an object. Even though a lot of studies have been done on this method, there is still a lot of improvements possible, especially in case of dynamic conditions. In dynamic conditions since the target is moving, images acquired during the test have motion effect (blurring). This blurring can increase the uncertainty of the motion measurement. The present study aims to investigate the uncertainty of different DIC methods in dynamic situation and find the variables of the camera and motion that can be effective in the accuracy of DIC methods. The focus of the study is only on 2D DIC, but similar problems also exist in 3D DIC. Therefore an understanding of two dimensional problems can be helpful for future studies in 3D conditions too. There were two sets of experiments done in this study. The first experiment was done in different exposure times and gains of the camera, changing the initial displacement entered to the tip of the beam. However, the second experiment is much more extensive, as each variable was changed considering the other variables constant and instead of applying a random initial displacement to the beam, a shaker was used to create the motion. Moreover, in the second experimentation, other than exposure time and gain, the effect of diaphragm aperture of the camera on the uncertainty of the motion was investigated too. The first method used in the studies was normal DIC method, which was done analyzing the acquired images in Vic-2D and MATLAB. The second technique was deconvolution method to find the motion of the beam. Finally, using deconvolution technique, the motion effect of the images acquired were estimated. Knowing the amount of motion effect, new reference images with blurring were generated for each image taken during the motion. Then DIC was applied comparing each image acquired during the test with its own generated reference image. By this method the effect of blurring in the uncertainty of the measurement is expected to decrease significantly.