Walid Bouceffa wrote: > > Feed: gmane.comp.graphics.otb.user > Item: Object-oriented classification steps > > Hi, > > Could anyone please help with the steps for an object oriented > classification using otb app (without using Monteverdi). > I want to use a shapefile instead of the Object Labeling module in > Monteverdi. > >
Hi Walid, If I am not mistaken, there is no equivalent application to do what the Object Labeling module does in Monteverdi 1. What you could do is: 1. generate a raster image where all pixels belonging to the same "object" (segment) have the same radiometric values : use the segmentation application (in raster mode, if you do it in vector mode, you will have to rasterize with gdal afterwards) to generate the segments 2. extract radiometric features on this image (radiometric indices, SFS) 3. perform a supervised classification (create training sets using QGis) on the concatenation of your raster segmentation + features Of course this approach does not exploit shape attributes, but SFS textures can help here. You don't have access either to the active learning approach proposed by the Object Labeling application. But, other than that, you could achieve similar results, and using the command line applications (or the Python bindings) you can script things. I hope this helps a little bit. Jordi > Best Regards, > Walid > > > ---------------------------------------------------------------------- > > Date: Sun Oct 26 10:32:28 +0100 2014 > Author: Walid Bouceffa -- -- 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 [email protected] To unsubscribe from this group, send email to [email protected] 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 [email protected]. For more options, visit https://groups.google.com/d/optout.
