Nadeem Kolia
PROJECT PROPOSAL
Mentor: Luke Catania
Organization: U.S Army ERDC-TEC (Engineering Research and Development Center - 
Topographical Engineering Center)
Laboratory: Computer Systems


OBJECTIVE
Extract high resolution building models in three dimensions from LiDAR (Light Detection 
And Ranging) data.


JUSTIFICATION
LiDAR data is being collected by the army to aid in campaign and mission planning.
The LiDAR data represented in commercial ArcGIS software is capable of representing
the data in two dimensions, presenting only an outline of buildings in the urban
environment and masking the true external structure of the building.  Extracting
building models from LiDAR data would provide information about the urban environment
that would be essential for planning operations.  The only current software that
is capable of extracting building models requires heavy user interaction and is
thus not feasible for rapidly modeling an urban area.  Software is needed that 
will automate the process of extracting building models with the need of little
to no user interaction. 

DESCRIPTION
ArcGIS is a commercial application suite currently used to view the LiDAR data in
two dimensions.  It includes ArcScene, a geographic information system software
capable of 3d representations.  It allows for user development through Microsoft's
COM interface and ESRI ArcObjects.  I will create a DLL in Visual Basic implementing
ArcObjects which will function as a plug in to ArcScene capable of extracting 
and creating high resolution models of buildings.  

LIMITATIONS
The project does not have space, budget or safety limitations or concerns.  The
commercial software is already being deployed by TEC and TEC already has an 
extensive library of LiDAR data readily available for processing and analysis. 
There are technical limitations that accompany any algorithm intensive computer
program.  The LiDAR data is stored as a raster, and can reach sizes that are too
large to be loaded into memory all at one time.  2D Footprints of the buildings
can be used to reduce the amount of data that needs to be loaded into memory at
one time.  Along with memory limitations, there are also computational limitations
due to time.  For such large datasets, the program must utilize an efficient and
fast algorithm.  2D Footprints will also assist in this since they reduce the amount 
of data that needs to be processed.