The digital image correlation (DIC) is a smart and easy optic method able to compute the displacement field of material by comparing two pictures. Here, the DIC is applied for mechanical problems and the proposed application is able to compute 2D strain fields from images. The figure below gives a general overview of the method.

The main problem with the local digital image correlation is the induced noise produced by the data acquisition. This noise impacts dramatically on the results. The proposed tool embeds a set of numerical algorithms in order to reduce the noise and to compute smoothed strain fields. Another interesting feature is the capability to compute displacement and strain fields from an unstructured set of points. You can enable this option by choosing the Non aligned markers option. Note that the proposed tool is based on pydic, a free Python code for mechanical DIC.

Usage

First, you need to set the grid size and the window size. The grid size are the horizontal and vertical spaces (in pixel) between the correlation windows. The window size are the horizontal and vertical lengths (in pixel) of the correlation windows.

In the mechamatic terminology, a correlation window is a small rectangular area used to compute the DIC and the markers are simply the points located at the center of these correlation windows.

If you want to promote local results, you have to choose a lot of correlation windows with small sizes. In this case, the results may be perturbed by a high level of noise. You can smooth the data by choosing the smooth data option. If you want to promote global results, you have to choose few correlation windows with big sizes.

Secondly, you have to choose your reference image in the Image of reference section. Note that you can choose an image stored on your computer or you can take a live snapshot with your camera. After that, you have to choose one or more other images in the Images to process section for DIC comparison with the reference image.

Finally, you can choose if you want to smooth the data or not. In most of cases, a good choice is the Aligned markers option associated with the Smooth data option.

Results

Many kind of results are available. You can plot the xx, yy and xy strain fields, download a comma separated (*.csv) result file or get some pedagogical gif animations. Many tools, such as spreadsheet software can read *.csv files. It means that you can easily post-treat by yourself the data contained in this file. These data are : the coordinates of the correlation points (markers), their related displacements and their xx, yy, xz strains. Note that the displayed strain maps correspond to the last image from the list that you choose for the DIC process.

Test it now !

If you want to test it now, you can download this archive that contains two images captured during a four point bending test. So, you are able to use these images for testing immediately this me-cha-ma-tic application !