image domain for image classification

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

image domain for image classification

mavis chisha
How do I create an image domain required in image classification?  Ilwis is only finding value domain which is rejecting.
 
Please help.  i want classify some satellite images.
 
Kind regards
 
Mavis

_______________________________________________
Ilwis mailing list
[hidden email]
http://list.52north.org/mailman/listinfo/ilwis
http://ilwis.forum.52north.org
Reply | Threaded
Open this post in threaded view
|

Re: image domain for image classification

schouwenburg

What kind of classification?

 

Martin Schouwenburg

[hidden email]

 

From: [hidden email] [mailto:[hidden email]] On Behalf Of mavis chisha
Sent: vrijdag 24 mei 2013 11:52
To: [hidden email]
Subject: [Ilwis] image domain for image classification

 

How do I create an image domain required in image classification?  Ilwis is only finding value domain which is rejecting.

 

Please help.  i want classify some satellite images.

 

Kind regards

 

Mavis

 


_______________________________________________
Ilwis mailing list
[hidden email]
http://list.52north.org/mailman/listinfo/ilwis
http://ilwis.forum.52north.org
Reply | Threaded
Open this post in threaded view
|

Re: image domain for image classification

ElenaABaldina
In reply to this post by mavis chisha
Domain in classification is just a list of objects (or classes) which you want to reveal as a result of image classification. Domain is something like a legend of a map. You should create it before classification (File -> Create -> Domain), and after a window "Create Domain" opening you may click Help, to get help on topic "what type of a domain you need. Domain "Class' is the most useful in many cases. In the next window "Domain Class" you should add some names of classes. This Domain firstly should be used for training in a process of Sample Setting/
From ILWIS Help:
The Classify operation performs a multi-spectral image classification according to training pixels in a sample set. Before classification, a sample set thus has to be prepared with Sample.