I am able to compute the fractal dimension of a textured surface like elevation with a short script based on the blanket method described in Haidekker, Advanced Biomedical Image Analysis, 2011.

I also have a set of 100 "count surfaces" which display how many of a set of points fall into each pixel at each scale. It's tempting to subject the set of count surfaces to the same algorithm, but my instinct is that any resulting number -- ok, I gave in and computed D=2.16 -- would be meaningless.

Is it? Can you explain why, in layman's terms if possible?

Count, original (1 per cell) and twice aggregated by sum.

  • $\begingroup$ The goal of my analysis is to show that regressions between mean-aggregated surface data and sum-aggregated mass data have problems across scales. Any references on that topic would be greatly appreciated. $\endgroup$ – J Kelly Oct 14 '16 at 16:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.