Features are user-defined objects that can be modeled or represented using geographic data sets. Examples include roads, buildings, and water bodies.

Use Feature Extraction to identify objects from panchromatic or multispectral imagery based on spatial, spectral, and texture characteristics. Then classify the objects into known feature types, using one of the following workflows:

• Rule Based: Define features by building rules based on object attributes such as area, elongation, spectral mean, texture, etc.
• Example Based: Select training data (samples of known identity) to assign objects of unknown identity to known features.
• Segment Only: Extract segments only without performing rule-based or example-based classification. This allows you to create a segmentation image to which you can add your own attributes or classify however you choose.

Refer to the following topics for programming routines that automate Feature Extraction workflows:

• FXSegmentation Task
• ENVI_FX_RULEBASED_DOIT
• ENVI_FX_EXAMPLEBASED_DOIT

You must have an ENVI Feature Extraction license in order to use these tools and API routines.