Error in normalize_intensity: only point geometries are supported

[cflynn8]

I am trying to use normalize_intensity on a small set of las files. I am linking to the data here. Here is my r code:

# Normalize intensity of a folder of las files
library(lidR)
library(future)
plan(multisession)

# Read in the folder of las files as and LASCatalog
marked_cat <- readLAScatalog("S:\\IRAD\\LidarProcessing\\IntensityNormalizationTesting\\SaltonSeaTestData\\LAS_Copy\\LAS_SmallTiles\\marked",
                             filter = "-drop_withheld -drop_class 7 18")

las_check(marked_cat)

# Visualize the las tiles of the files being processed
plot(marked_cat)

# Estimate the positioning of the lidar sensor. Uses first and last returns with gpstime.
sensor <- track_sensor(marked_cat, Roussel2020(pmin = 50))

# Set a location to store the output files with normalized intensity values
opt_output_files(marked_cat) <- "S:\\IRAD\\LidarProcessing\\IntensityNormalizationTesting\\SaltonSeaTestData\\LAS_Copy\\LAS_SmallTiles\\marked\\norm"

# Normalize the intensity using the sensor positioning from above.
las_norm <- normalize_intensity(marked_cat, range_correction(sensor, Rs = 2000))

From the last line of code, I get the following error:

#> Error: Only point geometry types are supported

The files I am reading in are las files (v 1.4).

Thank you!

[Jean-Romain]

Please show the output of marked_cat, las_check() and sensor

[cflynn8]

marked_cat does not output anything in the console but reads in to the environment as a "Formal class LAScatalog".

las_check(marked_cat) output:

 Checking headers consistency
  - Checking file version consistency... ✓
  - Checking scale consistency... ✓
  - Checking offset consistency... ✓
  - Checking point type consistency... ✓
  - Checking VLR consistency... ✓
  - Checking CRS consistency... ✓
 Checking the headers
  - Checking scale factor validity... ✓
  - Checking Point Data Format ID validity... ✓
 Checking preprocessing already done 
  - Checking negative outliers...
    ⚠ 4 file(s) with points below 0
  - Checking normalization... no
 Checking the geometry
  - Checking overlapping tiles... ✓
  - Checking point indexation... no

sensor output:

Warning messages:
1: In min(bb[, 1L], na.rm = TRUE) :
  no non-missing arguments to min; returning Inf
2: In min(bb[, 2L], na.rm = TRUE) :
  no non-missing arguments to min; returning Inf
3: In max(bb[, 3L], na.rm = TRUE) :
  no non-missing arguments to max; returning -Inf
4: In max(bb[, 4L], na.rm = TRUE) :
  no non-missing arguments to max; returning -Inf

[Jean-Romain]

I can probably enhance lidR by triggering an informative error but your problem is pretty simple: you have only 6 pulses with more than one returns over the entire coverage. And two files have literally 0 multiples returns.

las = readLAS(marked_cat, filter = "-drop_withheld -drop_class 7 18")
filter_poi(las, ReturnNumber > 1)
#> class        : LAS (v1.4 format 6)
#> memory       : 4.2 Kb 
#> extent       : 620100.2, 620400, 3661536, 3661642 (xmin, xmax, ymin, ymax)
#> coord. ref.  : NA 
#> area         : 4873.4 units²
#> points       : 6  points
#> density      : 0 points/units²

[Jean-Romain]

Your example no longer fails and returns an empty sf data.frame and triggers a warning.