6

I have two point layers with XYZ data; one set are points measured in the field and the second set is a Lidar data set.

I want to check the accuracy of the Lidar in comparison to the measured data.
Is there a way to do so without creating grids? That is, comparing the vectors and not the rasters?

Note: I found similar questions but either there was no answer or it referred to ArcGIS for Desktop.

Improve this question
5
  • 5
    This is a tough question to answer. The thing is, you say you want to assess the accuracy of the LiDAR data, but in reality you are probably assessing the accuracy of the LiDAR data's ability to represent something, like say the ground surface elevation. The raw LiDAR data is a point cloud...how can you be certain that you have a LiDAR point exactly where you have a ground-truthed point? Given positional error, you can't really. Instead, the thing to do is to assess how close the LiDAR derived ground surface (from an interpolated grid) matches your GPS point at that location.
    – WhiteboxDev
    Commented Sep 3, 2014 at 14:18
  • 5
    ...following on, interpolating to a grid is the only way that you can compare the elevation of the LiDAR at the exact location of the GPS point. Of course, this means that you're not only assessing the accuracy of the LiDAR to represent the ground surface, but also the interpolation method, grid resolution, data model, etc. to represent the surface. This is just the way it is, I'm afraid.
    – WhiteboxDev
    Commented Sep 3, 2014 at 14:20
  • 1
    Thanks WhiteboxDev. I'll try yo create two raster layers from the data and compare them.
    – H.Wiener
    Commented Oct 6, 2014 at 7:09
  • 1
    When looking for answers did you check something like this: gis.stackexchange.com/questions/11377/… which looks like it finds the nearest points from dataset A and dataset B? If that works you pick an appropriate cutoff distance and compare elevation for example.
    – cokrzys
    Commented Jan 6, 2015 at 6:14
  • I think it should be clearly specified that what is compared here is Z dimension (elevation) of two datasets - at least that is how I understand and will try to answer this question.
    – Miro
    Commented Sep 19, 2015 at 13:44

1 Answer 1

Reset to default
3

I believe at the end it is comparison of two surfaces in some way. But that doesn't necessarily mean you need to convert points and compare two grids. Depending on what you need it for there can be these options:

  1. compare points Z of first dataset to vector function interpolating Z value for the same XY from the second dataset points - I believe this is achievable only with some scripting skills, I would personally go for python.scipy.interpolate

  2. interpolate only one dataset into grid and compare points from second through point sampling tool plugin and field calculator

  3. interpolate both datasets into grids and use raster calculator - probably the only way in QGIS to get complete comparison of two surfaces sourced from two point datasets

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.