range_correction function

Intensity normalization algorithm

This function is made to be used in normalize_intensity . It corrects intensity with a range correction according to the formula (see references): [REMOVE_ME]$I_{norm} = I_{obs} \left(\frac{R}{Rs}\right)^fInorm = Iobs * (R/Rs)^f [REMOVE_ME_2]$

To achieve the range correction the position of the sensor must be known at different discrete times. Using the 'gpstime' of each point, the position of the sensor is interpolated from the reference and a range correction is applied.

range_correction(sensor, Rs, f = 2.3, gpstime = "gpstime", elevation = "Z")

get_range(las, sensor, gpstime = "gpstime", elevation = "Z")

Arguments

• sensor: SpatialPointsDataDrame or sf object containing the coordinates of the sensor at different timepoints t. The time and elevation are stored as attributes (default names are 'gpstime' and 'Z'). Z can also come from the geometry if the input records XYZ coordinates. It can be computed with track_sensor .
• Rs: numeric. Range of reference.
• f: numeric. Exponent. Usually between 2 and 3 in vegetation contexts.
• gpstime, elevation: character. The name of the attributes that store the gpstime of the position and the elevation of the sensor respectively. If the input contains 3 coordinates points, elevation is not considered.
• las: an object of class LAS. get_range() is a regular function documented here for convenience.

Description

This function is made to be used in normalize_intensity . It corrects intensity with a range correction according to the formula (see references):

To achieve the range correction the position of the sensor must be known at different discrete times. Using the 'gpstime' of each point, the position of the sensor is interpolated from the reference and a range correction is applied.

Examples

# A valid file properly populated
LASfile <- system.file("extdata", "Topography.laz", package="lidR")
las <- readLAS(LASfile)

# pmin = 15 because it is an extremely tiny file
# strongly decimated to reduce its size. There are
# actually few multiple returns
sensor <- track_sensor(las, Roussel2020(pmin = 15))

# Here the effect is virtually null because the size of
# the sample is too small to notice any effect of range
las <- normalize_intensity(las, range_correction(sensor, Rs = 2000))

# This might be useful for some applications
R = get_range(las, sensor)

References

Gatziolis, D. (2011). Dynamic Range-based Intensity Normalization for Airborne, Discrete Return Lidar Data of Forest Canopies. Photogrammetric Engineering & Remote Sensing, 77(3), 251–259. https://doi.org/10.14358/pers.77.3.251