tree_detection: Tree detection

Description

Performs tree detection by applying the functions tree_segmentation and tree_extraction to objects of class SpatRaster-class, LAS-class or LAScatalog-class

Usage

tree_detection(las, res = 1, ROI = NULL, normalize = FALSE, crown = FALSE, ...)

Value

A sf collection of POINTs with 7 fields: tree id, local maximum stats (height, dominance radius), segment stats (surface and volume), coordinates (x and y). In case argument crown is TRUE, a crown field containing the WKT geometry of the 2D crown is also present.

Arguments

las: An object of class SpatRaster-class, LAS-class or LAScatalog-class
res: numeric. The size of a grid cell in point cloud coordinates units, used to rasterize the point cloud. In case the las argument is a SpatRaster res is not used.
ROI: spatial polygons in sf/sfc format, in the same CRS as argument las. geometric object that defines the region where tree detection has to be performed. In case the input is of class LAScatalog-class, the chunk buffer set with engine_options is applied to the point cloud to prevent border effects, but only treetops lying within the ROI are returned.
normalize: boolean. Should the point cloud be normalized before detection (not applicable if las argument is a SpatRaster) ?
crown: Parameter passed to tree_extraction
...: Parameters passed to tree_segmentation

References

Monnet, J.-M. 2011. Using airborne laser scanning for mountain forests mapping: Support vector regression for stand parameters estimation and unsupervised training for treetop detection. Ph.D. thesis. University of Grenoble, France. Section 6.2 https://theses.hal.science/tel-00652698/document

Monnet, J.-M., Mermin, E., Chanussot, J., Berger, F. 2010. Tree top detection using local maxima filtering: a parameter sensitivity analysis. Silvilaser 2010, the 10th International Conference on LiDAR Applications for Assessing Forest Ecosystems, September 14-17, Freiburg, Germany, 9 p. https://hal.science/hal-00523245/document

Examples

Run this code

# load canopy height model
data(chm_chablais3)
chm_chablais3 <- terra::rast(chm_chablais3)
# create polygon of region of interest
ROI <- sf::st_polygon(list(cbind(
 c(974360, 974360, 974380, 974380, 974360),
 c(6581640, 6581680, 6581680, 6581640, 6581640)
)))
# convert to sfc and set projection
ROI = sf::st_sfc(ROI)
sf::st_crs(ROI) <- terra::crs(chm_chablais3)
#
# tree detection
trees <- tree_detection(chm_chablais3)
# plot results
# canopy height model background
terra::plot(chm_chablais3)
# detected trees
plot(trees["h"], add = TRUE, cex = trees$h/20, col = "black")
#
# tree detection in ROI and minimum tree height set to 10
trees_ROI <- tree_detection(chm_chablais3, ROI = ROI, hmin = 10, crown = TRUE)
# create polygons from WKT field
trees_ROI_crowns <- sf::st_as_sf(sf::st_drop_geometry(trees_ROI), wkt = "crown")
# plot results
# canopy height model background 
terra::plot(chm_chablais3)
# detected trees
plot(trees_ROI["h"], add = TRUE, cex = trees_ROI$h/20, col = "black")
# corresponding crowns
plot(sf::st_geometry(trees_ROI_crowns), add = TRUE, border = "black", col = NA)
# add ROI
plot(ROI, add = TRUE, border = "red", col = NA)