fit lidar data to control points

geomaticist

I see GM has a tool to check control points to Lidar data. This is very nice. There is also a tool to "fit lidar data to control points". This is supposed to rectify the entire cloud to match the control points.

My question is: How does it do it. What is the method. I cannot find any reference about this. I realize it will do some sort of data fit that warps existing cloud elevations. But how does this fit spread and join to the next control point?

I have bare earth ground classified points, so all other are removed already. I have survey control. I am happy to adjust the cloud to the control - but want to better understand how it is doing it. Not just a magic button. Once I rectify the data it all changes. Meaning some good data could be getting worse and some poorer data gets better. Exactly how far reaching I cannot tell. I don't see any documentation.

Anyone know the answer to this?

By the way - I am dealing in 0.2ft or less of miss to ground control. SO I don't really need to rectify - but if I do - what and how is it applying the adjust!!

bmg_mike

There is another thread at https://www.globalmapperforum.com/discussion/14464/how-does-the-fit-lidar-to-control-points-algorithm-work#latest that discusses the various 'fit Lidar to control point methods' and how they work.

Thanks,

Mike

Global Mapper Guru

bmg_mike

In GM 26.0 one of the adjustment options is just providing a single Z offset value. This defaults to the average delta of the control point Zs.

In an upcoming release (likely GM 26.1), you will be able to turn on and off control points in the Fit Lidar UI rather than having to unselect them and run the process again.

Thanks,

Mike

Global Mapper Guru

iaerobotics

We also need to know this information.

DerrickW

I would like to know this as well, please. 

dsmithdsmapping09

Did you ever find an answer to this? I am curious how this works too.

geomaticist

I actually just compute the shift as an average. I use GM to aid in this. I make a DEM and then load control points. Have control inherit values off DEM and then compare list 1 to list 2 in spreadsheet. And average is the shift. The standard deviation is the precision. I only use points on hard surface.

But i think since this original thread, GM might have been updated. and you can select the method of adjust?

3. Rigid, with translation - all the lidar points or shifted globally, there is are no local adjustments made around the control points.

so I accomplish the above '3' by myself. So never noticed this as an option. i prefer my method because I filter the control points by code and make decisions which to include. Also I look how many standard deviations some points are and will remove them as likely outliers. Control has error as well!

I use the t-distribution to compute the 95% certainty as a number of standard deviations. This helps when I have few control points.