2

I am using dense time stacks of remotely sensed imagery to analyse vegetation change in the Arctic from 1984-2010 for one scene. I will be needing to classify each image, likely using the Maximum Likelihood Classifier tool in ArcMap 10.2. In the Batch mode of the tool, it just seems to allow importing multiple raster bands, then exporting to a raster output. Is there a way to feed multiple raster .tiffs in to the same classification schema? I read a paper (Castrence et al 2014) where they performed "a single classification over the entire time span, thereby [avoiding] problems associated with multiple independent classifications of images from different dates." Could you build one signature file and then use it for all the rasters? I am not experienced with Pyton scripting, but I'm getting the feeling this will be needed...I know that the images must be relatively calibrated against a master. Basically I'm lazy and I want to avoid doing things manually.

I also have PCI Geomatica, if anyone knows of a batch classification tool/model with EASI.

  • I want to be able to use the same training data for multiple years of the same Landsat scene, covering the same AOI. First I think I need to calibrate my images to a master image to make the signature file. This paper claims its possible, but they didn't explain how they did it. – vputinski Nov 7 '14 at 21:04
  • In the paper that you refer to they had one classification for the stack. A class might, for example, be 'agriculture until the 1990s', i.e. they are temporally dependent. If I'm interpreting your question correctly, you're looking at creating a classification for each year in the stack. – WhiteboxDev Nov 7 '14 at 22:21

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.