Hi Jérôme, This mailing list is in english so my answer will be in english as well. I guess every one has its own recipes to select training areas.
Your goal is to have training areas that are pure (only 1 class
per region), and that show enough diversity to characterize your
classes all over the scene. After doing the training step, you can validate the model with a ground truth and compute the confusion matrix. See OTB applications like TrainImagesClassifier and ComputeConfusionMatrix. Also more information here :
https://www.orfeo-toolbox.org/CookBook/recipes/pbclassif.html#validating-the-classification-model Guillaume On 04/19/2017 12:55 PM, Jérôme Loukou
wrote:
--
--
-- Check the OTB FAQ at http://www.orfeo-toolbox.org/FAQ.html You received this message because you are subscribed to the Google Groups "otb-users" group. To post to this group, send email to otb-users@googlegroups.com To unsubscribe from this group, send email to otb-users+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/otb-users?hl=en --- You received this message because you are subscribed to the Google Groups "otb-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to otb-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout. |
- [otb-users] traitement d'image avec OTB 5.10 Jérôme Loukou
- Re: [otb-users] traitement d'image avec OTB 5.10 Guillaume Pasero