Skip to content

Latest commit

 

History

History
243 lines (185 loc) · 6.6 KB

opencv-setup.md

File metadata and controls

243 lines (185 loc) · 6.6 KB
Raw

OpenCV

FFmpeg

OpenCV (Open Source Computer Vision Library) is an open source computer vision and machine learning software library.

UPGRADE OpenCV

  1. Uninstall old version:

    CV_DIRS=(
    "/usr/local/share"
    "/usr/local/include"
    "/usr/local/lib"
    "/usr/local/bin"
)

sudo -v

for cv_dir in ${CV_DIRS[@]}; do
    sudo find ${cv_dir} -name "*opencv*" -exec rm -rf {} \;
done

  2. Remove installation directories:

    rm -rf "${HOME}/opencv"
rm -rf "${HOME}/opencv_contrib"

  3. Install OpenCV following the instructions below!

Install OpenCV with Python and Qt 6 support

  1. Choose the version. Paste that in the terminal prompt:

    # export OPENCV_VERSION="4.X.X"
# export OPENCV_VERSION="3.4.X"
export OPENCV_VERSION="4.6.0"

  2. Easy install using mac-setup:

    bash <(curl -fsSL raw.githubusercontent.com/borgesnotes/mac-setup/master/install) opencv

  3. Install on your own:

    # Installing Opencv dependencies
brew install cmake pkg-config ceres-solver gdal eigen glog \
             harfbuzz jpeg libpng libtiff openexr tbb \
             ffmpeg gstreamer gst-plugins-base gst-plugins-good
    pyenv shell 3.10.7
mkvirtualenv opencv
pip install --upgrade pip
pip install numpy
    # Download OpenCV

# Clone OpenCV project from GitHub
git clone https://github.com/opencv/opencv "${HOME}/opencv"

# Clone OpenCV extra modules from GitHub
git clone https://github.com/opencv/opencv_contrib "${HOME}/opencv_contrib"

# Create a branch from last stable version
cd "${HOME}/opencv"
git checkout tags/${OPENCV_VERSION} -b ${OPENCV_VERSION}

# Create a branch from last stable version
cd "${HOME}/opencv_contrib"
git checkout tags/${OPENCV_VERSION} -b ${OPENCV_VERSION}

# Create a build directory
mkdir "${HOME}/opencv/build"
    # Check Qt PATH
if ls ${HOME}/Qt*/*/clang_64/bin/qmake 1>/dev/null 2>&1; then
    QT_ON_OFF=ON
else
    QT_ON_OFF=OFF
fi

QT_LIB=$(echo ${HOME}/Qt*/*/clang_64/lib/cmake)
QT_QMAKE=$(echo ${HOME}/Qt*/*/clang_64/bin/qmake)

# Check Python PATH
pyenv shell 3.10.7
workon opencv
py3_exec=$(which python)
py3_config=$(python -c "from distutils.sysconfig import get_config_var as s; print(s('LIBDIR'))")
py3_include=$(python -c "import distutils.sysconfig as s; print(s.get_python_inc())")
py3_packages=$(python -c "import distutils.sysconfig as s; print(s.get_python_lib())")
py3_numpy=$(python -c "import numpy; print(numpy.get_include())")
    # Configure OpenCV via CMake
cd "${HOME}/opencv/build"
pyenv shell 3.10.7
workon opencv

cmake -D CMAKE_BUILD_TYPE=RELEASE \
      -D CMAKE_INSTALL_PREFIX="/usr/local" \
      -D OPENCV_EXTRA_MODULES_PATH="${HOME}/opencv_contrib/modules" \
      -D ENABLE_PRECOMPILED_HEADERS=OFF \
      -D OPENCV_GENERATE_PKGCONFIG=ON \
      -D WITH_CUDA=OFF \
      -D WITH_QT="${QT_ON_OFF}" \
      -D CMAKE_PREFIX_PATH="${QT_LIB}" \
      -D QT_QMAKE_EXECUTABLE="${QT_QMAKE}" \
      -D QT5_DIR="${QT_LIB}/Qt5" \
      -D QT5Core_DIR="${QT_LIB}/Qt5Core" \
      -D QT5Gui_DIR="${QT_LIB}/Qt5Gui" \
      -D QT5Test_DIR="${QT_LIB}/Qt5Test" \
      -D QT5Widgets_DIR="${QT_LIB}/Qt5Widgets" \
      -D Qt5Concurrent_DIR="${QT_LIB}/Qt5Concurrent" \
      -D OPENCV_ENABLE_NONFREE=ON \
      -D WITH_GTK=OFF \
      -D WITH_OPENGL=ON \
      -D WITH_OPENVX=ON \
      -D WITH_OPENCL=ON \
      -D WITH_TBB=ON \
      -D WITH_EIGEN=ON \
      -D WITH_GDAL=ON \
      -D WITH_FFMPEG=ON \
      -D BUILD_TIFF=ON \
      -D BUILD_opencv_python2=OFF \
      -D BUILD_opencv_python3=ON \
      -D PYTHON_DEFAULT_EXECUTABLE="${py3_exec}" \
      -D PYTHON3_EXECUTABLE="${py3_exec}" \
      -D PYTHON3_INCLUDE_DIR="${py3_include}" \
      -D PYTHON3_PACKAGES_PATH="${py3_packages}" \
      -D PYTHON3_LIBRARY="${py3_config}" \
      -D PYTHON3_NUMPY_INCLUDE_DIRS="${py3_numpy}" \
      -D INSTALL_PYTHON_EXAMPLES=ON \
      -D INSTALL_C_EXAMPLES=ON \
      -D BUILD_EXAMPLES=ON \
      -D CMAKE_CXX_STANDARD=14 ..
    # make opencv
cd "${HOME}/opencv/build"
pyenv shell 3.10.7
workon opencv

make -j 4
    # make install opencv
sudo make -j 4 install

Test OpenCV installation with Python

workon opencv
python -c "import cv2;  print(cv2.__version__)"

Test OpenCV installation with C++

We can run some sample code located at opencv/samples/cpp:

cd "${HOME}/opencv/samples/cpp"
workon opencv

g++ -std=c++11 $(pkg-config --cflags --libs opencv4) facedetect.cpp -o ${TMPDIR}/test

${TMPDIR}/test

Test OpenCV installation with Qt

  1. Add :/usr/local/bin to PATH

    Go to Project, expand Build Environment and add :/usr/local/bin to PATH.

  2. Add a new variable called PKG_CONFIG_PATH with the value /usr/local/lib/pkgconfig

    Qt build settings for OpenCV

  3. Add a few lines to the project file .pro to tell qmake to use pkg-config for OpenCV:

    # Set up Qmake to use pkg-config
QT_CONFIG -= no-pkg-config
CONFIG += link_pkgconfig
PKGCONFIG += opencv4

# Add OpenCV Libraries
LIBS += $(pkg-config --libs opencv4)

  4. You may receive a runtime error similar to the message below:

    dyld: Symbol not found: __cg_jpeg_resync_to_restart
Referenced from: /System/Library/Frameworks/ImageIO.framework/Versions/A/ImageIO
Expected in: /usr/local/lib/libJPEG.dylib
in /System/Library/Frameworks/ImageIO.framework/Versions/A/ImageIO
The program has unexpectedly finished.

    To fix it, go to Project -> Run -> Run Environment -> Unset DYLD_LIBRARY_PATH.

    DYLD problem

Sym-link OpenCV to a virtual environment (if necessary)

  1. Activate the TARGET Python virtual environment:

    workon TARGET_ENV

  2. Create a symbolic link to OpenCV:

    # Get the OpenCV library path
opencv_lib=$(echo ${HOME}/opencv/build/lib/python3/*.so)

# Get the site-packages path of the virtual environment
site_packages=$(python3 -c "import distutils.sysconfig as s; print(s.get_python_lib())")

# Create a symbolic link to OpenCV library
ln -s ${opencv_lib} ${site_packages}