As a means of checking spatial autocorrelation in the selection of training samples for image classification, I am trying out semivariance analysis. I have managed to generate variograms from image pixels using gstat package in R. And I understand from this question (https://stackoverflow.com/questions/10769884/what-are-the-units-of-distance-in-gstat-variogram) that in my case the computed distance for the variogram is Euclidean distance. Now I wish to implement the range from the fitted variogram as spacing for regular points in QGIS, but I’m not sure how to translate Euclidean distance into point spacing.

Any Ideas, suggestions or experiences with what I want to do? I have seen some publications that used this approach, with the distance in meters or pixel count but don't know how. I would also be open to try this in R or any other feasible software if pointed in the right direction..

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.