Introduction

Track ArUco markers using MATLAB and OpenCV.

To date, MATLAB is not providing any toolbox which can detect ArUco markers. The only way to have any support on ArUco markers, is via the OpenCV C++ API.

Using the library

Prerequisites

OpenCV

You need to have OpenCV together with the opencv_contrib projects installed in your computer.

Most Linux distros provide OpenCV via their package manager. You can however also compile manually:

Here is an example on how you can manually compile and install OpenCV 4.1.2:

# Close OpenCV
git clone https://github.com/opencv/opencv.git && cd opencv && git checkout 4.1.2 && cd .. &&

# Clone opencv_contrib
git clone https://github.com/opencv/opencv_contrib.git && cd opencv_contrib && git checkout 4.1.2 && cd ..

# Build OpenCV
cd opencv && mkdir build && cd build && \
cmake -D CMAKE_BUILD_TYPE=RELEASE \
      -D CMAKE_INSTALL_PREFIX=/usr/local \
      -D INSTALL_C_EXAMPLES=OFF \
      -D INSTALL_PYTHON_EXAMPLES=OFF \
      -D WITH_V4L=ON \
      -D WITH_QT=ON \
      -D WITH_OPENGL=ON \
      -D WITH_CUBLAS=1 \
      -D WITH_TBB=ON \
      -D WITH_OPENMP=ON \
      -D WITH_CUDA=ON \
      -D ENABLE_PRECOMPILED_HEADERS=OFF \
      -D OPENCV_EXTRA_MODULES_PATH=../../opencv_contrib/modules \
      -D BUILD_EXAMPLES=OFF \
      .. && make -j4

# Install OpenCV
sudo make install

# Remove repos
cd ../.. rm -rf /opencv && rm -rf /opencv_contrib

MATLAB Computer Vision Toolbox OpenCV Interface

The Computer Vision Toolbox OpenCV Interface of MATLAB also needs to be installed. For that, you need to have a mathworks account.

To install it, just use the "Add-Ons" button of the "Home" tab:

This is the addon you should install:

Build MEX Function

Before you can build the mex function, you need to locate:

• the OpenCV include files (header files), e.g. by using sudo find / -name 'libopencv_aruco*' 2>/dev/null
• the OpenCV library files (lib files), e.g. by using sudo find / -name 'aruco.hpp' 2>/dev/null

In the below example the OpenCV 4.1.2

• headers are located in /usr/local/include/opencv4 and the
• lib files are located in /usr/local/lib.

To build the aruco_detect.cpp mex function, navigate within MATLAB to the root of this project and run the following command (in MATLAB of course):

mexOpenCV -L/usr/local/lib -lopencv_aruco -I/usr/local/include/opencv4 aruco_detect.cpp

Run the mex function

Add missing OpenCV libs in MATLAB's path

Before you can run the mex function you need to create a symbolic link of the libopencv_aruco.so.* file in the MATLAB search path, otherwise MATLAB cannot locate it in run-time.

In the below example the libopencv_aruco.so.* file is located in /usr/local/lib/ (as also mentioned above).

ln -s /usr/local/lib/libopencv_aruco.so.4.1 <MATLAB_ROOT_FOLDER_PATH>/bin/glnxa64/libopencv_aruco.so.4.1

If you are using a newer version of gcc than MATLAB is, then you need to link the libstdc++.so.6 as well.

First create a backup of the existing one:

mv <MATLAB_ROOT_FOLDER_PATH>/sys/os/glnxa64/libstdc++.so.6 <MATLAB_ROOT_FOLDER_PATH>/sys/os/glnxa64/libstdc++.so.6.backup

then link the systems libstdc++.so.6 to MATLAB's search-path:

ln -s /usr/lib/libstdc++.so.6 <MATLAB_ROOT_FOLDER_PATH>/sys/os/glnxa64/libstdc++.so.6

If you still getting errors when running the mex function you can also try to start MATLAB as:

LD_PRELOAD=/usr/lib/libstdc++.so.6 <MATLAB_ROOT_FOLDER_PATH>/bin/matlab -desktop

Run the example

Now you can run the example test/aruco_detect_test.m file.

Issues and Contributing

• Please let us know by filing a new issue.
• You can contribute by opening a pull request.