Installation

The installation of lidar2dems itself is straightforward, as it only consists of a Python library and some scripts. However, there are several required dependencies that need a more manual process. These installation notes are for Ubuntu 14.04 or 16.04, but should work for most Debian-based linux systems.

Easy Install

For Ubuntu machines, save the easy-install.sh script and run it in a temporary working directory (it can be deleted afterward). If the easy install process is sucessful, you can disregard the rest of these installation notes.

The easy-install script installs all the necessary dependencies that are available from repositories. It also downloads the source for other packages, compiles, and installs them. Modify the easy-install.sh script to customize installation for your own machine.

Manual Install

Dependencies from packages

Many of the dependencies can be easily installed via the Ubuntu packaing tool from the main repositories. Note: If installing on Ubuntu 14.04 you also need the swig package in addtion to swig2.0 due to a packaging issue.

$ sudo apt-get install swig2.0 python g++ libboost-all-dev libgdal-dev \
                       libgdal-dev gdal-bin python-pip python-numpy \
                       python-scipy python-gdal cmake++ libeigen3-dev \
                       libflann-dev libopenni-dev libqhull-dev qt-sdk \
                       libvtk5-qt4-dev libpcap-dev python-vtk libvtk-java \
                       libgeotiff-dev python-setuptools libxslt1-dev \
                       python-wheel libgeos++-dev libxslt-dev

Dependencies from source

These are dependencies that must be built from source code, and are available on GitHub.

Running CMake: Several of the programs below use cmake which works in a standard fashion across platforms. The best approach is to create an out-of-source build, meaning that a build directory is created somewhere other than the source directory. Follow the steps below for any source that utilizes cmake for building.

$ cd "projdir"      # "projdir" is the project directory
$ mkdir build
$ cd build
$ cmake -G "Unix Makefiles" ../
$ make
$ sudo make install

LASzip

LASzip is used by many other LiDAR software packages to support compressed LAS files. It is not needed if you are using LAS files, however since it installs easily it is best to install now, so that PDAL can be built with LASzip support.

  1. Clone repository

     $ git clone https://github.com/LASzip/LASzip.git
 $ cd LASzip
 $ git checkout tags/2.0.2

  2. Run CMake steps as above

Points2Grid

Points2Grid is used by PDAL to create rasters from point clouds using a local gridding method. It will not be used directly, but is used by PDAL.

  1. Clone repository

    $ git clone https://github.com/CRREL/points2grid.git

  2. Run CMake steps as above

PCL:

Install PCL with this install-pcl.sh script.

PDAL

PDAL is library for conversion and filtering of LiDAR data. It is under very active development and has features particularly well suited to batch processing. It can also incorporate PCL for doing advanced algorithms on point clouds.

The notes here are for building PDAL from source, although there may be binaries available for your system, see http://www.pdal.io/download.html . See the complete PDAL Documentation at http://www.pdal.io/docs.html for more information.

  1. Clone repository

     $ git clone https://github.com/PDAL/PDAL.git pdal
 $ cd pdal
 $ git checkout tags/1.0.1

  2. Run CMake as above, adding the additional options below to the command line.

     $ cmake -G "Unix Makefiles" ../ -DBUILD_PLUGIN_PCL=ON -DBUILD_PLUGIN_P2G=ON -DBUILD_PLUGIN_PYTHON=ON -DPDAL_HAVE_GEOS=YES

Install lidar2dems

lidar2dems is a pure python library, and is easily installed with the included setup.py script.

  1. Install gippy

    $ pip install https://github.com/Applied-GeoSolutions/gippy/archive/v0.3.9.tar.gz

  2. Clone repository

    $ git clone https://github.com/Applied-GeoSolutions/lidar2dems.git
$ # as of this writing v1.1.1 is the current release
$ cd lidar2dems
$ git checkout tags/v1.1.1

  3. Run setup

    $ ./setup.py install

Run tests

Some automated tests can be run to ensure the installation has succeeded. Run the tests from the lidar2dems repository directory that was cloned from github. If you see the output below, lidar2dems is working correctly.

$ nosetests test -v

Test classification ... ok
Test creating density ... ok
Create DTM ... ok
Create DSM ... ok
Create CHM ... ok
Test getting classification filename ... ok
Test finding las files ... ok

----------------------------------------------------------------------
Ran 7 tests in 2.513s

OK