Do you have data as figures or pictures?

If yes, try use any software that counts for example a box dimension. I'm using for example Benoit 1.3. It is also available for Matlab.

http://www.trusoft-international.com/benoit.html

For 2D image, you can use different methods for fractal dimension estimation (box counting, differential box-counting method, spectrum method, varainances method... ), 

the most method  are available in mathworks Matlab.

Box counting is a simple method to begin with. You can use this code for matlab

 http://www.mathworks.com/matlabcentral/fileexchange/13063-boxcount/content/boxcount/html/demo.html

Thanks. You are right. but as a matter of fact these methods give different results for fractal dimension. I wonder which one is most reliable?

 @ Sajjad Foroughi  

The most acceptable one is box counting method. Actually, through different ways of picture treatment , there would be varies results. So there is no wrong on methods itself. The key is to get a reliable and high-quality binary image at the first step.

@Stephenson Yo, @ Sajjad Foroughi

Benoit 1.3, that I previously suggested, has different possibilities to calculate box counting dimension, including the possibility to rotate boxes with any angle.

As Stephenson Yo written the most important is to have high-quality binary image.

In 2D images, the box counting is the best way.

You can check the following links

https://www.mathworks.com/matlabcentral/fileexchange/13063-boxcount

https://www.mathworks.com/matlabcentral/fileexchange/30329-hausdorff-box-counting-fractal-dimension

https://www.mathworks.com/matlabcentral/fileexchange/31911-finding-the-fractal-dimension-using-box-counting-method