How are objects detected and differentiated in GIS and remote sensing?

Answer 1

When data collected everything is assigned a classifier. For example, dirt is 1, things not on ground is 2, etc.. It's up to the Analyst to determine what the range is for things to be considered an object. E.g. anything above 1 is an object (aka anything above the dirt is an object). From there the analyst sets more rules/parameters to the classifiers to differentiate it. E.g. objects classified between 2-3 are buildings while objects classified 4-5 are trees.

This is a very concise explanation meant to only explain the fundamental idea behind it.