Digital-Forestry-Toolbox

A collection of tools for Matlab/Octave

View on GitHub Download .zip Download .tar.gz

The Digital Forestry Toolbox (DFT) is collection of functions and tutorials for Matlab/Octave designed to help analyze remote sensing data related to forests.

Setup

Download and uncompress the Zip or Tar archive
If you want to use the example LAZ files included in the data folder or the external links, you must first uncompress them with LASzip
Start Matlab/Octave
Add the DFT folders to the Matlab/Octave search path using addpath(genpath('path to DFT main folder'))
Open any of the functions in Matlab/Octave and check the header for syntax and example usage
Check the tutorials

Getting started

Reading a LAS file

Use LASread() to import the contents of a LAS (1.0-1.4) file into a structure:

pc = LASread('myfile.las');

You can then access any of the standard LAS format blocks, e.g.:

pc.header
pc.variable_length_records
pc.record
pc.extended_variable_length_records

Displaying a 3D point cloud

To display all classes except noise points (class 7), we start by creating a logical indexing variable:

idxl_noise = ismember(pc.record.classification, 7);

We then use scatter3() to plot non-noise points colored by return intensity:

figure
scatter3(pc.record.x(~idxl_noise), ...
    pc.record.y(~idxl_noise), ...
    pc.record.z(~idxl_noise), ...
    6, ...
    pc.record.intensity(~idxl_noise), ...
    'Marker', '.');
colorbar
axis equal tight
title('Return intensity')
xlabel('X')
ylabel('Y')
ylabel('Z')

Tutorials

License

Unless otherwise stated in the file, the code is licensed under GNU GPL V3.
The airborne laser scanning datasets are a courtesy of the States of Solothurn, Zürich and Geneva (Switzerland).

Reference

The Digital Forestry Toolbox was developed by Matthew Parkan (GIS Research Laboratory, EPFL) with support from the Swiss Forest and Wood Research Fund (project 2013.18).

If you use this code in your work, please consider including the following citation:

Matthew Parkan. (2018). Digital Forestry Toolbox for Matlab/Octave. DOI: 10.5281/zenodo.1213013. Available: http://mparkan.github.io/Digital-Forestry-Toolbox/.

Functions

The following symbols indicate GNU Octave support:

= Fully compatible

= Partially compatible

= Not compatible

Input/Output

Function	GNU Octave support	Description
`LASread()`		Read 3D point cloud from ASPRS LAS (1.0-1.4) file (.las)
`LASwrite()`		Write 3D point cloud to ASPRS LAS (1.0-1.4) file (.las)
`HDRread()`		Read ENVI header file (.hdr)
`SBETread()`		Read Applanix SBET trajectory file (.sbet)
`TRJread()`		Read Terrascan TRJ trajectory file (.trj)
`ASCread()`		Read ESRI ARC/INFO ASCII grid file (.asc)
`ASCwrite()`		Write ESRI ARC/INFO ASCII grid file (.asc)

Basic

Function	GNU Octave support	Description
`LASclip()`		Clip 3D point cloud (LAS file) with a polygon
`LASextent()`		Extract spatial extent of 3D point clouds (LAS files) and write result to ESRI shapefile (.shp)
`LASmerge()`		Merge 3D point clouds (LAS files)

Grids

Function	GNU Octave support	Description
`elevationModels()`		Compute elevation models (i.e. terrain, surface, height) from a 3D classified point cloud
`rasterize()`		Convert a 3D point cloud to a 2D raster

Laser metrics

Function	GNU Octave support	Description
`laserEchoRatio()`		Computes the echo ratio of a 3D point cloud
`laserIntensityCorrection()`		Corrects return intensity for the sensor to target range
`laserPulses()`		Determines the pulse number associated with the individual returns (sorted by acquisition GPS time)
`laserTimeFormat()`		Convert GPS week time to GPS satellite time, GPS adjusted time or UTC time

Tree metrics

Function	GNU Octave support	Description
`treeStems()`		Detect individual tree stems in a 3D point cloud
`treeWatershed()`		Extract individual tree crowns from a raster Canopy Height Model (CHM) using the (optionally marker-controlled) watershed segmentation described in Kwak et al. (2007)
`treeMetrics()`		Computes various segment metrics based on geometry, intensity, pulse return and color characteristics

Canopy metrics

Function	GNU Octave support	Description
`canopyCover()`		Computes a canopy crown cover map using either a convolution window or the Delaunay triangulation method described in Eysn et al. (2012)
`canopyPeaks()`		Find local maxima (peaks) coordinates in a raster Canopy Height Model (CHM)

Utilities

Function	GNU Octave support	Description
`alt2height()`		Converts 3D point cloud altitude to height above terrain
`topoColor()`		Topological coloring (cf. Welsh and Powell, 1967) of 2D or 3D labelled points (i.e. assign distinct colors to adjacent clusters)
`crossSection()`		Extract a 2D cross-section from a 3D point cloud
`geomedian()`		Compute the geometric median of a an array of 3D points
`gps2utc()`		Convert GPS satellite time (i.e. number of seconds since 1980-1-6) to UTC time
`subsample()`		Subsample a point cloud

Data

Airborne Laser Scanning - 3D point clouds

File	Land cover	Location	Sensor	Date
zh_2014_a.laz (external link)	Mixed forest	Zürich, Switzerland (47.58384 N, 8.76594 E, ~380 m ASL), see map	Trimble AX60	March 10-13, 2014
zh_2014_b.laz (external link)	Urban, mixed forest	Zürich, Switzerland (47.37460 N, 8.56824 E, ~550 m ASL), see map	Trimble AX60	April 2, 2014
so_2014_a.laz (external link)	Woodland Pasture	Solothurn, Switzerland (47.25952 N, 7.46979 E, ~830 m ASL), see map	Trimble AX60	April 7, 2014
ge_2017_a.laz (external link)	Mixed forest	Geneva, Switzerland (46.29240 N, 6.13380 E, ~430 m ASL), see map	Riegl LMS-Q1560	February 19-20, 2017

Airborne Laser Scanning - Rasters

File	Land cover	Location	Sensor	Date
so_2014_woodland_pasture.tif	Woodland Pasture	Solothurn, Switzerland (47.25952 N, 7.46979 E, ~830 m ASL), see map	Trimble AX60	April 7, 2014

National Forest Inventories

File	Description
swiss_nfi_flora.xlsx	Swiss National Forest Inventory (NFI), tree species Latin/vernacular (English, French, German, Italian) names and codes