I have two hyperspectral datasets. One I consider to be the 'original' dataset and the other which has been clipped to remove areas of substrate prior to classification. When I run the original data set through the classification workflow (using MLC) it works fine and gives a good output. However, when I attempt the same process on the clipped data set the workflow does not respond for a long time before resetting itself. I do use different sets of ROI's for both.

I don't know if by clipping the original hyperspectral data set I have somehow altered the file.

If anyone has any ideas on where I might be going wrong they would be much appreciated.