I’ve been tasked with a habitat classification project which is requiring me to assign thousands of empty polygons different classification types and I do not know how to automate this process.
Examples of assignment are things like “upland forest”, “tidal flat”, “unconsolidated mud”, etc. Said features also have other attributes which need to be included, such as sub-types, unique identification codes and other required records as per. NOAA guidelines.
Background: polygons derived from eCognitition software, polygons delineated off of variability in from satellite imagery. The polygons have automatically been parsed out to these “upland forest”, “tidal flat”, “unconsolidated mud”, etc. categories based on variability in the imagery.
What I need to do now is assign the blank polygons their respective categories into the attribute table – as right now are just empty polygons. Below is an example of what the finished product should look like.
My idea is to train pixels an order to define the numerous habitat classification types and then create some kind of temple which has the attributes already populated – then join them somehow.
I know there are better, more consistent and more accurate ways to accomplish this though they are unknown to me.
Can anyone advise?