Automated Accuracy Assessment
I may have missed something so do correct me if I'm wrong, but there doesn't seem to be a way to perform an accuracy assessment automatically within a ruleset. I think this could be quite a useful ruleset algorithm when analysing large amounts of data over several projects using eCognition server, especially if CSV files could be exported automatically. I have found myself manually going back to each project to extract my confusion matrices and this would be a real time saver.
Was this article helpful?