Skip to main content
  • EN
Solved

Lidar feature extraction

  • 29 March 2013
  • 1 reply
  • 57 views
Hello,

 

From what I am able to tell, FME is a very powerful spatial data extraction tool.  But I have some specific use cases I am trying to decide if FME is a software that can be of use to me for extracting features out of Lidar data.

 

A perfect example of my needs would be to use FME on lidar data taken of, say, along a  street where we need to extract the edge of curb lines.  

 

Another example would be to detect changes over time, say i have a lidar dataset of a hillside taken in at two different times throughout the year.  Would one be able to use FME to detect changes over time between the two datasets?

 

Any help would be greatly appreciated on the matter!

 

Thank you all for reading my post.

 

For the first example, I doubt FME would be able to perform feature extraction from LIDAR without extensive programming.  It is possible that a good python developer could use FME as a point cloud API to write some code to handle it, but I don't think anything built into the software would work.

 

 

The second example, on the other hand, is pretty easy to set up in FME.  I'll explain the process briefly.

 

 

Each set of LIDAR data can be used to create a surface model in FME using the SurfaceModeller transformer.  You can then create a grid of 2D points (2DGrridCreator) which spans your area of interest, and drape them to each surface, saving the elevation (CoordinateExtractor). 

 

 

Now, take the points and set their elevation (3DForcer) to the difference between the two draped values.  These points now represent a DEM-like difference between the two surfaces.  You can run them into a SurfaceModeller and create contours, or manipulate them in a number of other ways to understand how the surface has changed.

Reply


Community Stats

29,690
Posts
112,851
Replies
37,345
Members
Terms & ConditionsCookie settings