What you can do with this website ...

You can contribute to and benefit from this website in many ways. Here are some examples:

  • Post your newest papers (through forums or blogs).
  • Read other users' papers and newest papers from numerous journals.
  • Ask for questions (through forums or blogs).
  • Answer other users’ questions.
  • Download source code.
  • Upload your original or improved source code.

[Hint for signed in user] For convenience, the system will keep you signed in if you do not log out.

Problems with DIC results

Dear Opticist

I have started trying to use your software to carry out DIC on some tension tests.  I am taking the images from individual frames of a high speed video capture of the test but I am having no success with the software.

Either the two result images are just black and white except for the colour scales on the right (which are scaled between 1E6 and -1E6) or there is a very small coloured zone in each of the results images.

I have attached three of the images below, I have been using Frame-300 as the reference image as this is taken from an early part of the test, therefore the applied load is small.

Are the speckle patterns that I have used too ill defined, for example is the contrast too low?  Is the image resolution too low?

I hope that you can help and I look forward to hearing from you.




DIC image from frame-300

DIC image from frame 1200

DIC image from frame 2400

Your images are OK

The speckle patterns are not very good but they are accetpable for DIC analysis. The problem seems that you did not specify the ROI (region of interest). Here is a rough testing result.

To get better results, you need to set appropriate DIC parameters.

U field

V field

Problems with DIC results / Your images are OK

Dear Opticist

Thank you for your prompt reply and results images.

I was using a ROI image but see from the results that you sent that the ROI boundary should be inside the edges of the specimen.  I have adjusted my ROI image and am now getting DIC results.

Unfortunately, as you can see from the attached, my displacement results seem to be more crude than yours and my strain results do not seem to be meaningful.

Could you help me with the following?:

  • Do you anticipate that if I use an improved speckle pattern then I would be able to obtain strain results?

  • You said that the speckle pattern is not very good.  What aspect of the patten is most important to correct?  Maybe the contrast between light and dark?  Maybe the speckle density should be more consistent?

  • I am using Microsoft Paint in order to convert the images to 256 color .BMP files, this reduces the quality of the image that I am using with your software

  • Could you let me know the settings that you used for the most important DIC parameters in order to obtain your images?

Thanks again, I look forward to hearing from you.



Example of 256 colour image

Frame 2400 x-displacement

Frame 2400 y-displacement

Frame 2400 x-strain

code of 3D DIC software


Is the software provided on this website 3d Digital Image Correlation is open source software or not?

If yes, from where can I download the code of this software. Also are there any developers forum.


Thank you. Please reply.



displacement calculation in milimeters


i was wondering how we could use the software in order to calculate
the displacement of a specific point. the displacement is small(in term
of xe-2 mm).

we applied small spots on a test material, we took 2 pictures, before
and after the traction experiment. the problem is that the displacement
is so small that it is not  obvious to choose the same point even with
the zoom option.

can the software be that accurate to give exact results?

Re: displacement calculation in milimeters

With our newest program, we can detect displacement as small as 0.00001 pixels if the image quality is very good.  Let's assume that the accuracy is 1% of that, i.e., 0.01 pixels, in order to detect 0.01mm, the actual dimension of each pixel must be no more than 1mm.  This is very easy to achieve in practice.  If your camera gives you 1024x1024 pixels, then it is fine once the object in the image is less than 1000mm wide.

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer