Hi there.
I'm quickly gaining confidence and knowledge in the world of ArcGIS, but I still have much to learn.
I am trying to perform an unsupervised classification on an orthorectified multispectral image. I'm aware that, in order to determine the number of classes I want, I need to know the number of different pixel values in the image. So, how can I effectively determine the number of pixel values in my image?
Thanks!
I don't know the exact answer, but did find some information that may be of help while poking around:
Number of classes: The number of classes generated during unsupervised classification should be about 10 times the number of bands in your image. For example, if you are working with multispectral red, green, blue and NIR bands, then the number here will be 40 (4 classes x 10).
Source: Supervised and Unsupervised Classification in ArcGIS - GIS Geography
Raster dataset attribute tables—Help | ArcGIS for Desktop
Also, adding Imagery and Remote Sensing for greater exposure.
Chris Donohue, GISP
Alright. Then I assume I would reclassify the image after I define my initial classes?
Also, thank you very much.
Which specific process are you using to do the Unsupervised Classification? If it is this one
Iso Cluster Unsupervised Classification—Help | ArcGIS for Desktop
the results will be a classified output. You can then Reclassify if desired.
Chris Donohue, GISP
Oops, sorry I forgot to specify. Yep - the Iso Cluster Unsupervised Classification. Gotcha. Thanks again Chris!