voxel_metrics() results are different with all_voxels=TRUE

[zoeschindler]

Hello,

I really hope this is not some embarrassing user-side error, but I've got an issue with voxel_metrics(). Depending on whether I use all_voxels=TRUE or all_voxels=FALSE, it returns different results:

voxels <- voxel_metrics(las, length(X), res=0.1, all_voxels=FALSE)
nrow(voxels[voxels$V1 > 0 & voxels$Z > 0.65 & voxels$Z <=0.95])
#> [1] 212
voxels <- voxel_metrics(las, length(X), res=0.1, all_voxels=TRUE)
nrow(voxels[voxels$V1 > 0 & voxels$Z > 0.65 & voxels$Z <=0.95])
#> [1] 0

It also happens when I beforehand exchange the NAs with 0s:

voxels <- voxel_metrics(las, length(X), res=0.1, all_voxels=FALSE)
nrow(voxels[voxels$V1 > 0 & voxels$Z > 0.65 & voxels$Z <=0.95])
#> [1] 212
voxels <- voxel_metrics(las, length(X), res=0.1, all_voxels=TRUE)
voxels$V1[is.na(voxels$V1)] <- 0
nrow(voxels[voxels$V1 > 0 & voxels$Z > 0.65 & voxels$Z <=0.95])
#> [1] 0

The result using all_voxels=FALSE is the right one. I would attach the las-file, but I don't know how to add a 122MB las-file here.

Edit: When I don't select the data depending on height, the difference is 301134 vs 43107 rows.

[Jean-Romain]

Well, I will need the file. I'm sure you have access to a remote storage with your company/university. Or google drive with your google account if you have one. Also if it is a 122 MB las file it will become something like 25-50 MB laz file.

[zoeschindler]

Here, I hope this works. I just put it into a zip-file.

[Jean-Romain]

I found an important problem in the 3D rendering. I don't know exactly what it is but there is bug confirmed

[Jean-Romain]

I still do not understand exactly why there was a problem with the join but I fixed it anyway. The output is now correct and I also enhanced the plot function for a better rendering. Thank you for reporting this issue.