Extracting buildings with IMAGINE Objective: Indepth

This movie goes into greater detail on the use of IMAGINE Objective to extract residential building footprints from a high resolution color air photo into a vector GIS shapefile format. Explanation is given on the steps that are applied within the Feature Model to analyze, extract and clean up the building footprints.




IMAGINE Objective is an add-on module for the ERDAS IMAGINE suite providing object based multi-scale image classification and feature extraction capabilities to reliably build and maintain accurate geospatial content.




This is the third in a series of introductory videos to IMAGINE Objective.

Category:

Science & Technology

Tags:

• IMAGINE
• Objective
• Feature
• extraction
• ERDAS
• Leica
• image
• classification
• remote
• sensing

[1ananderson]

Dense urban is always going to be an issue. More often than not you'll be better served using a stereo feature extraction package (such as Stereo Analyst for ArcGIS) to manually capture dense urban features than tryign to use an automated feature extraction package.

[1ananderson]

ERDAS will continue to resell the Feature Analyst products as well as offering IMAGINE Objective because we see two very different approaches being used. As with most image classification techiques you'll find that one package works better in some circumstances and another in others - it's always going to be down to the exact problem you are trying to solve and the data layers available to you.

[uncgrl84]

I've worked with this kind of feature extraction with Overwatch Geospatial's Feature Analyst software. Do you know how this compares to that program?

[ninefourMC]

That was very interesting, thanks. How well does this building footprint extraction process work in densely built up areas with varying building types? Is it just a matter of using more "rules"?

(BTW I came to this site through the erdas imagine newsgroup.)