Project SimpleTree - Development moved to Computree platform

Dear interested user. You will find on this homepage my phd work. I found a follow up post doctoral position and continued improving my method within the open source framework Computree. I extended my method to work on plot level and I am providing Windows executables of my plugin within Computree.

As I improved my method quite a bit already I highly recommend using the new software. You can get an impression of the capabilities from the following video tutorial:

You need to have a Computree user account (free) to be able to download. My new contact email is jan.hackenberg@inra.fr, although I still check my Allumni email.

I cannot update this homepage anymore according to my new progress but you still might find here valuable resources, especially in the statistic section.

    Software

A C++ based Implementation of a method building tree models published in Highly Accurate Tree Models Derived from Terrestrial Laser Scan Data: A Method Description and Non destructive method for biomass prediction combining TLS derived tree volume and wood density. This software tool is peer-reviewed presented in SimpleTree —An Efficient Open Source Tool to Build Tree Models from TLS Clouds The Implementation uses the Point Cloud Library as a dependency.

The source code is available on Github under BSD License. Recommended and supported OS is Linux Ubuntu. There is a possibility to install and de-install Ubuntu inside a Windows system like you install a normal software under Windows. Please refer to Ubuntu Wubi installer or Wubi FAQ. You need the PCL installed with version 1.8.0. This version is the development branch right now, so you need to compile it yourself. You will find step by step installation guide further down on this page.

If you use this software for publishing, please cite according to section Contribution.

SimpleTree screenshot. A TLS crown model of P. avium is visualized

    System requirements

I can not provide informations about the minimum system requirements. Right now the software does not use much multi threading, so the number of cores of your processor does not matter that much. Memory might be your bottleneck, I recommend at least 8GB, 16GB are better. You system should be 64 bit architecture.

  • 8-16 GB memory
  • 64 bit architecture

If you sent me information about your hardware and how my program performs with it, I will be able to make this section more precise.

    Installation on Linux Ubuntu

The following procedure was performed on three different machines with freshly installed Ubuntu Version 14.04.01 and 14.04.03. You can download 14.04.03 here. 14.04.02 contains a bug, but should not be available anyway anymore.

We will first install PCL 1.7.2. This can be done very easy, as Ubuntu is supported by PCL. You have to open a terminal and execute the following three commands as described here:

sudo add-apt-repository ppa:v-launchpad-jochen-sprickerhof-de/pcl
sudo apt-get update
sudo apt-get install libpcl-all

The last command is the actual installation command. The first two are to tell your system where to get PCL.

All dependencies of PCL are now installed, so we dont have to worry about them when installing PCL 1.8.0, which has not been released yet, so we have to compile by ourself. We install git, cmake and ccmake with one command:

sudo apt-get install git cmake cmake-curses-gui

We use git to fetch the PCL source code from Github:

git clone https://github.com/PointCloudLibrary/pcl

In our homefolder we have now a folder pcl. With the following commands we change the directory to the pcl folder, make a build folder in this directory and then change to the new created build folder:

cd pcl
mkdir build
cd build

We use from there ccmake:

ccmake ..

A commandline based editor opens and we should be able to press 'c' to configure and then 'g' to generate meta data and exit ccmake. You need to press two times 'c' to configure with no further changes made to enable the 'g' option.

The next two commands to build pcl 1.8.0

cmake ..
make

Now your C++ compiler has a lot of work for up to several hours.

If successful, you can now install the library finally:

sudo make install

Switch back to you home folder and lets get SimpleTree from GitHub. WARNING there is two versions available. One version is stable and can reproduce the Hackenberg et al. (2015b) results. The other is adapted to stems of tropical trees with a lot of undergrowth. This one is extended, can do more, but also contains an undetected bug. Contact me in case you do not know what version to install.

Here the

unstable

code/installation:

This version is moved to private repository again. Please contact me if you want access to it. The software publication release version is still available, see below.

And the
stable
code/installation:

cd ~
git clone https://github.com/SimpleTree/SimpleTreeRelease

Go to the build folder inside the new created simpleTree folder and run cmake:

cd ~/SimpleTreeRelease/build
cmake ../src/
make

In this build folder you should have now a Linux executable for my program. You can start either by clicking on it in a data manager like you are used from windows, or execute in console:

~/SimpleTreeRelease/build/simpleTree

    Windows

Currently Windows and Mac systems can not be supported. For making my program run under windows you would need to compile PCL in version 1.8.0 with its dependencies under windows. The PCL homepage gives detailed instructions about this process.

For CMake and Git you should also be able to find HowTo's with windows instructions.

 
This template downloaded form free website templates