I am trying to analyse a geological surface with R & Python and I would like to know if there are some ways to distinguish, divide and quantify repetitive structures such as advertised by a software company (picture below).

I don't have a precise scope but I would like to understand what shapes are repeating themselves onto the analysed surface.

Are there advised packages that could be useful to accomplish this task?

Analyze surface texture


In R, you can use library raster and the different terrain function or your custom focal functions.
- Using terrain with slope will give you information on the local slope
- with TPI will tell you if you are on a crease, a hill or a plate area.
- Using focal (potentially combined with aggregate), you can calculate some TPI with larger scope, or pass a Gaussian filter, ...

With these different indices, you can then apply some pixel classification, with PCA like proposed in your illustration or any other classification procedures like randomForest, ...

  • This is a great insight. Thanks. I am dealing with the point cloud first. Would you just consider a final mesh-reconstruction (rasterised) or xyz points could be useful as well? – user4330326 Jun 2 '17 at 23:33
  • The tools developed specifically for rasters are really interesting. They are also made to work correctly with your computer memory while loading only what is necessary. So I would find the adequate resolution, and work with rasters. – Sébastien Rochette Jun 3 '17 at 7:12

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.