grow_crowns: Outline Tree Crowns In Imagery
In pieterbeck/CanHeMonR: Tools for canopy health monitoring through image analysis

Given a an image and point shapefile with points ('seeds') placed at the center of crowns, estimate the tree crown around every seed, and write the results away as a polygon shapefile.

grow_crowns(r_file, seed_shpfile, seed_extent = rep(NA, 4),
  avoid_recalculation = T, prob_cut = 99, prob_step_down = 2,
  seed_nrs = NA, outp_crown_shp_filename, process_dir = NULL,
  max_crown_RADIUS_in_m = 4, min_crown_RADIUS_in_m = 0.6, plott = F,
  RGBseq = c(1, 2, 3), parallel = F, nWorkers = 4)

`r_file`	Filename of raster brick to run crown delineation on. From tests with hyperspectral data, it appears r_file can have up to 6 bands, equally spaced in wavelength, before the colinearity crashes mahalanobis
`seed_shpfile`	Filename of point shapefile that has points at the center of trees.
`seed_extent`	A numerical vector of length four. It should contain raster::extent parameters to restrict the analysis only to the points in seed_shpfile that fall within a certain extent. To not trim the data in a certain direction, set that position in seed_extent to NA. Default is c(NA,NA,NA,NA), which means the seed_shpfile is not trimmed before seed_nrs are checked..
`avoid_recalculation`	Do you want to check if a run with these settings and same output filename was already done, and add outlines of new seeds to that output? Default is T.
`prob_cut`	The probability [in percent] cut off in the chi square distribution that determines the cut-off in Mahalanobis distance higher prob_cut values produce larger crowns. Default is 99 percent.
`prob_step_down`	As long as the region grows all the way to the max_crown_radius, prob_cut will be stepped down by prob_step_down and the region will be grown again. Default is 2 percent
`seed_nrs`	Seeds to be used for crown estimation. Starts counting at 1, for FID = 0. If set to NA, all seeds will be included. Default is NA.
`outp_crown_shp_filename`	Base for the output polygon shape filename. _probcut.shp will be added to create the final filename.
`process_dir`	An optional directory to write output to while processing. Can come in handy if network access is a bottleneck or i/o to the network causes crashes. Default is NULL in which case output is written to the final destination directory.
`max_crown_RADIUS_in_m`	The maximum radius a crown can have (in image units of r_file. Best to work in m). Default is 4.
`min_crown_RADIUS_in_m`	The minimum radius a crown should have (in image units of r_file. Best to work in m). Default is 0.6.
`plott`	Logical. Do you want to plot the results? Defaults to F.
`RGBseq`	3-element vectors given the band nummers for RGB plotting of crowns. Only considered is plott=T. Default is c(1,2,3).
`parallel`	Should the code be run in parallel using the doParallel package? Default is FALSE.
`nWorkers`	If running the ocde in parallel, how many workers should be used? Default is 4.

Writes away a .shp polygon file of crowns. Attributes include npx_t_b = the number of pixels in the crown touching the edge of the disk of maximum crown size, nstrtcl = the number of pixels taken for granted as crown at the start of the region growing, probcut = the final prob_cut value for this crown. any attributes from the seeds .shp file will be saved in the output as well.

pieterbeck/CanHeMonR documentation built on May 25, 2019, 7:11 a.m.