Spectral Difference Segmentation - continual loop / crash
I'm segmenting and classifying 0.5m spatial resolution imagery in eCognition. In the workflow, I'm first running a multi-resolution segmentation and then a spectral difference segmentation. The MRS runs as expected but the subsequent spectral difference segmentation runs in a continual loop (~12 hours or more processing time) and thereafter it crashes with no visible segmentation output. If I manually stop the process around cycle 8 - 10 there is a good output with meaningful objects, but stopping the process manually is quite arbitary. The only thing I can think of is the size of the image files (~3.7 GB), but the hardware I'm running eCognition on should be more than capable of handling these processes.
Can you advise what is going on here?
Thanks in advance,
Was this article helpful?