© 2018. All rights reserved.
© 2018. All rights reserved.
Because of the high demands of PCL library dependence， it will be better to compile the PCL source code. In the introduction, I will provide the detailed illustration of the dependence choose and compile process.
Under Windows or Mac platform, we could download the VeloView software to scan the data directly. However, the Velodyne LiDAR company doesn’t provide the official version for Linux.
To install the latest PCL on Ubuntu, +git+ is necessary. Install it by:
sudo apt-get install git
Change directory to a place where to store the source code, then proceed to clone the PCL repository and create a symbolic link. (Note:It might be a good idea to create a symbolic link referencing the latest version, this way your application does not have to change when the PCL is upgraded to the next version.)
git clone https://github.com/PointCloudLibrary/pcl.git pcl-trunk ln -s pcl-trunk pcl
Now change directory to pcl you will see the PCL source code. Before proceeding with building the PCL libraries, there’are also some prerequisites needed.
Compiling PCL especially for the VLP sensor requires a host of prerequisites, including +cmake,git+ and several other packages. I will record the function of the packages step by step to unfold their effects in case of certain needs. There’re 3 sorts of dependence, which are (Anything related to the cmake installation please refer to [[PCL based on Ubuntu 14.04]], I have illustrated the steps in detail there.)
This one is for the g++,cmake and some essential build dependence, which should be already installed or finished at the Ubuntu PCL environment build before, but still need to check them.
sudo apt-get install g++ cmake cmake-gui build-essential
In this part I will list the package, required for the compilation and usage of the PCL libraries.
Boost, version>=1.46, pcl_*
sudo apt-get install libboost-all-dev
Eigen, version>=3.0, pcl_*
sudo apt-get install libeigen3-dev
FLANN, version>=1.71, pcl_*
sudo apt-get install libflann1 libflann-dev
VTK, version>=5.6, pcl_visualization
sudo apt-get install libvtk5.8-qt4 libvtk5.8 libvtk5-dev
QT, for the GUI of PCL and can provide some extra choice for the PCL visualizer.
sudo apt-get install qt-sdk libqt4-opengl-dev openjdk-7-jdk openjdk-7-jre
+_Warning 1:_+ PLEASE DON’T install the VTK6 and QT5 although the trunk version use them or you’ll meet a real mass then. +_Warning 2:_+ The apt-get install command may generate a window like:
!http://larrylisky.files.wordpress.com/2014/03/missingphonon.png?w=720!
Answer OK to continue, but you may want to follow the recommendation and install any missing components. For the blog editor, he needed a couple of missing components to enable full operation of Phonon:
sudo apt-get install phonon-backend-gstreamer
sudo apt-get install phonon-backend-vlc
For the analyze of .pcap file, _+extremely crucial in this part!!!+_ This is the key point why I have to re-compile the PCL library for more than 4 times.
sudo apt-get install libpcap-dev
Some other dependence for the compilation:
sudo apt-get install git-core freeglut3-dev pkg-config
sudo apt-get install doxygen
sudo apt-get install mpi-default-dev openmpi-bin openmpi-common
sudo apt-get install libusb-dev libusb-1.0-0-dev
sudo apt-get install libgtest-dev
sudo apt-get install libxmu-dev libxi-dev
sudo apt-get install graphviz mono-complete
sudo apt-get install libglew-dev libsuitesparse-dev
This are the optional part in the official page, but out of the VLP numerous dependence, some of them are mandatory in fact. I will mark them.
COMPULSORY
Although the listed two dependence are marked as optional in the website but I extremely recommend installing them , so I annotated them as “COMPULSORY”.
CUDA part has been finished by @Gary. :-)
QHull >=2011.1 pcl_surface
sudo apt-get install libqhull* libqhull-dev
ELECTIVE
There’re 2 versions, OpenNI 1.x(called OpenNI) and OpenNI 2(called OpenNI2) respectively. In fact, OpenNI is the driver to support Kinect, but I have seen OpenNI includes some libraries related to pcl_io, so I installed OpenNI in case of the IO issue.
git clone https://github.com/OpenNI/OpenNI.git openni
And change the branch to unstable:
cd openni
git checkout unstable
Run these lines to install OpenNI:
cd Platform/Linux/CreateRedist/
chmod +x RedistMaker
./RedistMaker
cd ../Redist/OpenNI-Bin-Dev-Linux-x64-v1.5.8.5
Maybe not that version, just change it according to the real condition.
sudo ./install.sh
(Note: The command line ./RedistMaker will compile several source files. If there was any error, you will not get a ../Redist folder, and therefore, you won’t be able to go to the next step. Fix the build error first and then rerun ./RedistMaker.)
The cmake sometimes will warn that the OpenNI2 library couldn’t find. But I searched that the OpenNI2’s function is to support the Kinect2, as I have installed the OpenNI before and that one will act for some IO function, and I didn’t find any specific problem until now. So, the installation depends on your need.
+However, still some problem now.The cmake is always reporting “missing the OpenNI2 library files”, will be fixed later.+ +Recommended not use it now.+
Download the install package from the official website:
wget http://com.occipital.openni.s3.amazonaws.com/OpenNI-Linux-x64-2.2.0.33.tar.zip
For me, the latest version will be 2.2.0.33 but you can check it on the “OpenNI2”:http://structure.io/openni of the newest one.
Decompress it:
unzip OpenNI-Linux-x64-2.2.0.33.tar.zip
tar -xjf OpenNI-Linux-x64-2.2.tar.bz2
cd OpenNI-Linux-x64-2.2
It’s same to change the version number so that the command will match the actual one.
Install it:
sudo ./install.sh
After the generation of ??OpenNIDevEnvironment?? file then add the environment variables of the OpenNI2:
cat OpenNIDevEnvironment » ~/.bashrc
So that the path of the Include and Redist folder will be imported to the environment variables: @OPENNI2_INCLUDE@ and @OPENNI2_REDIST@. (Note: Reboot needed and check it by @echo@ command.)
To summarize the compile process, I install the below requests:
sudo apt-get install libboost-all-dev
sudo apt-get install libeigen3-dev
sudo apt-get install libflann1 libflann-dev
sudo apt-get install libvtk5.8-qt4 libvtk5.8 libvtk5-dev
sudo apt-get install qt-sdk libqt4-opengl-dev
sudo apt-get install libqhull* libqhull-dev
sudo apt-get install libpcap-dev
sudo apt-get install git-core freeglut3-dev pkg-config
sudo apt-get install doxygen
sudo apt-get install mpi-default-dev openmpi-bin openmpi-common
sudo apt-get install libusb-dev libusb-1.0-0-dev
sudo apt-get install libgtest-dev
sudo apt-get install libxmu-dev libxi-dev
sudo apt-get install graphviz mono-complete
sudo apt-get install libglew-dev libsuitesparse-dev
sudo apt-get install openjdk-7-jdk openjdk-7-jre
plus the procedure of OpenNI.
In addition, I have seen such words at the PCL official website: pcl_* denotes all PCL libraries, meaning that the particular dependency is a strict requirement for the usage of anything in PCL. So, it’s possible to use:
sudo apt-get install pcl_*
as a complementary of PCL basic dependence.
PCL is compiled using the cmake or cmake-gui utility. Change directory to the PCL source directory and enter the following commands:
mkdir release
cd release
cmake -DCMAKE_BUILD_TYPE=None -DBUILD_GPU=ON -DBUILD_apps=ON -DBUILD_examples=ON ..
make
(Note: For some reason the use of -DCMAKE_BUILD_TYPE=Release will lead to assembler error, but -DCMAKE_BUILD_TYPE=None will build just fine.)
Finally, to install PCL, enter the command:
sudo make install
The library files are installed at /usr/local/lib, and the header files are installed at /usr/local/include/pcl-1.8 (or whatever the version is).
[[Velodyne 16 Mapping(VLP-16) Grabber Demo]]
“INSTALLING AND RUNNING POINT CLOUD LIBRARY ON UBUNTU”:http://larrylisky.com/2014/03/03/installing-pcl-on-ubuntu/ “Compiling Point Cloud Library from Source”:http://www.pointclouds.org/downloads/source.html “PCL/OpenNI tutorial 1: Installing and testing”:http://robotica.unileon.es/mediawiki/index.php/PCL/OpenNI_tutorial_1:_Installing_and_testing “PCL/OpenNI troubleshooting”:http://robotica.unileon.es/mediawiki/index.php/PCL/OpenNI_troubleshooting#OpenNI “Sina Blog_Openni2+libFreenect under Ubuntu”:http://blog.sina.com.cn/s/blog_4a628f030102vb9a.html