SeptaCam

Code for controlling the seven-camera system

Installation

• Create environment

$conda create -n septacam python=3.6 numpy
$conda activate septacam

• Install Pylon:

  • We've successfully used multiple versions including 5.0.12 and 7.3.0, so probably you can just use the newest version and you'll be fine. If weird stuff happens, try 7.3.0 which is known to work.
  • Note that there are two sets of installation instructions that you need to follow: the "basic" install instructions come in a text file named INSTALL that is directly in the .tar.gz file that you download from Basler. Then there are some more "advanced" configuration instructions in /opt/pylon/share/pylon/README (which is a file that gets unpacked from the .tar.gz file inside the .tar.gz file you downloaded from Basler) that you also want to go through -- specifically as of version 7.3.0, the sections Performance Optimization > USB Vision Cameras and Permissions for Real-Time Thread Priorities are important.

• In the SeptaCam Makefile, set PYLON_ROOT to point to your pylon installation.

• Build and install opencv:

$sudo apt install build-essential cmake git pkg-config libgtk-3-dev \
    libavcodec-dev libavformat-dev libswscale-dev libv4l-dev \
    libxvidcore-dev libx264-dev libjpeg-dev libpng-dev libtiff-dev \
    gfortran openexr libatlas-base-dev python3-dev python3-numpy \
    libtbb2 libtbb-dev libdc1394-22-dev libopenexr-dev \
    libgstreamer-plugins-base1.0-dev libgstreamer1.0-dev
$git clone https://github.com/opencv/opencv.git
$git clone https://github.com/opencv/opencv_contrib.git
$git checkout 4.0.0 (both repos)
$cd ~/opencv
$mkdir build
$cd build
$cmake -D CMAKE_BUILD_TYPE=RELEASE \
       -D CMAKE_INSTALL_PREFIX=/usr/local \
       -D INSTALL_C_EXAMPLES=ON \
       -D INSTALL_PYTHON_EXAMPLES=ON \
       -D OPENCV_GENERATE_PKGCONFIG=ON \
       -D OPENCV_EXTRA_MODULES_PATH=../../opencv_contrib/modules \
       -D BUILD_EXAMPLES=ON ..
$make -j8
$sudo make install

• Install jsoncpp $sudo apt-get install libjsoncpp-dev

• Install boost $sudo apt-get install libboost-all-dev

• Install tcl $sudo apt-get install tcl-dev

• Install tk $sudo apt-get install tk8.6-dev

• Create .obj directory in SeptaCam folder $mkdir .obj

• Compile software $make GUI_BASLER_m

• Test that you can launch the GUI with no cameras plugged in $./GUI_BASLER_m

Configuration

Set paths

In C++_Source/constants.h:

• Update INSTALL_DIR, SEQ_FILE, VID_BATCH_FILE, MAKE_VIDEOS_FILE to have the correct path to the SeptaCam software.
• Update OUTPUT_DIR to have the path where you want to save your images.

Afterward, run make clean and make GUI_BASLER_m to re-build.

Configure cameras

Plug in your camera(s). Try launching the UI with ./GUI_BASLER_m. You should get a message saying X cameras detected where X is how many cameras you plugged in, then a line that says .../SeptaCam/Config/40012345.pfs, then an error Exception on connect(). For the next steps you need this 40012345 number, which should be your camera's serial number -- you can double check that this number is printed on the camera.

• In Config/defaultROI.json, add a section for your your camera(s).
• In Config/, create a .pfs file for your camera(s).
  • The easiest option is to just copy an existing .pfs file from another camera and rename it to have your camera(s)'s serial number.
  • The "right" way is to use Pylon. To do this, open Pylon, connect to your camera, then in the top menu bar click on Camera > Save Features... and create a pfs file named {serial number}.pfs in the Config/ folder.

Now you should be able to run ./GUI_BASLER_m and successfully connect to your cameras.