Skip to main content
Question

Hi! I am beginner in FME and joined the community only recently. I want to calculate the density of a point cloud (las/laz format, about 6 750 386 points, 1000kmx1000km area).


Forum|alt.badge.img

pointclouddensitySo far, I tried two approaches.

  1. Read pointcloud--> extract bounds-> tile point cloud into 1mx1m raster-->extract number of points per point cloud-->sort the number of points--> aggregate the number of points into groups-->write to csv
  2. read point cloud-->extract bounds-> calculate density by PointClouddensityCalculator--> rasterCellCoercer to extract the number of points into z--> coordinate extractor to get 3d points--> aggregate z (which corrsponds the number of points-->sorter-->write to csv

The problem with both approaches: it takes a lot of time (more than 1 h) . Does anyone have an idea how I could calculate the density quicker?

I add a screenshot of my workspace.

I appreciate any help. Thanks in advance!

Kind regards,

Ingrid

2 replies

jkr_wrk
Influencer
Forum|alt.badge.img+28
  • March 9, 2023

If you create a small sample set so the run-time does to around 2 minutes. What transformer seems to be the biggest impact on your process?

 

Why do you sort the points?

Why do you aggregate? Maybe a statistics-calculator is faster

Maybe a counter, with a group-by is faster.

 

I can't follow your second approach, the densityCalculator is a customTransformer, is it editable? What steps are done there. The first approach seems way more efficient, but maybe your result is not what you want?

 

 


Forum|alt.badge.img

Hey jkr_da,

thank you very much for your response. I think I now know why it took so much time. I had the caching on.

I want to calculate the density of the point cloud. Therefore, I divide the point cloud into a 1mx1m raster. I then calculate the number of points in each tile, hence the PointCloudPropertyExtractor. I sort the number of points with the Sorter to be able to aggregate ( as I want to group the number of points , my goal is to know how many squares have, e.g. 1 point, 2 points etc).

Thank you again for your help.

 


Reply


Cookie policy

We use cookies to enhance and personalize your experience. If you accept you agree to our full cookie policy. Learn more about our cookies.

 
Cookie settings