3

I have used the lidR package (https://github.com/Jean-Romain/lidR) to produce a normalized point cloud. I then filter out the points tagged with the overlap bit using:

opt_filter(ctg) <- "-drop_overlap"
catalog_retile(ctg)

The resulting point cloud appears to have a directional pattern, where points are present in somewhat uniform strips. It seems the unfiltered point cloud has a similar pattern but much it is much less pronounced. Is this a normal feature of lidar data?

Here is a 3d representation of the normalized and filtered point cloud with the pattern I have described: enter image description here

Improve this question
1
  • 1
    Now I know that your dataset is sample with two beam this image makes even more sense.
    – JRR
    Mar 23, 2020 at 1:15

1 Answer 1

Reset to default
4

ALS data are usually sampled with an oscillating mirror carried on a moving platform. The point pattern is thus a zigzag. Here you only have a very tiny region of the flightline so you are seeing near parallel lines. Everything looks ok to me. I'm only questioning the relevance of using -drop_overlap because its meaning is often misunderstood

Improve this answer
4
  • Thanks for your answer. The area is indeed quite small - it is a ~1000 square meter circle. I'm using -drop_overlap to achieve a single coverage point cloud. I will be extracting statistical moments (e.g. 90th percentile height...) from the point clouds to be used as predictor variables in an Above Ground Biomass model. I don't want the distribution of returns to be influenced by double coverage in areas of overlap.
    – Lucas
    Mar 11, 2020 at 18:32
  • You are assuming that the overlap bit tags the points in overlaps. I might be true... or not. Just sayin'
    – JRR
    Mar 11, 2020 at 18:35
  • You are suggesting that the flag could have been inappropriately set by the data producer. The metadata associated with the las tiles includes the following information - "Swath overage points were identified in these files using the standard LAS overlap bit." - "the nominal swath overlap, as a Percentage: 20" I feel like I should go by what is indicated in the xml metadata...
    – Lucas
    Mar 11, 2020 at 18:41
  • 1
    If it is documented by the provider it is ok!
    – JRR
    Mar 11, 2020 at 18:48

Your Answer

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

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