OpenCV has become one of the most widely used open-sourced library used in the field of computer vision. Thanks to the volunteers who have contributed to this library. The journey from Willow Garage to opencv.org has been inspiring. However, sometimes setting up the environment to build OpenCV from source is always challenging. With every new release as well as with a new operating system release, configuring OpenCV will be cumbersome.
In this blog post, I am going to note down the OpenCV build process in baby steps in recently released macOS Catalina. Dropping the support for 32-bit applications has created an unpleasant situation amongst some users.
Prerequisites before starting the build process:
You need to install python as we are not going to mess with the system python. For this blog post, I am going to use python 3.6. Download it from its official site and install it. Given below are the steps you need to follow to build OpenCV on macOS Catalina.
1. Install and Configure Xcode
Download Xcode from App Store. It will take a while as the size of this file is nearly 7 to 8 GB. After the successful installation of Xcode, run the following command.
$ sudo xcodebuild -license
Use the space key to move to the bottom of the agreement page and type [accept] to agree with the license. Next, we need to install the command line tools, which will install the basic compilers (gcc, g++, etc.). Use the command given below.
$ sudo xcode-select --install
2. Install homebrew for MacOS Catalina
To install homebrew, visit their official site [link]. Run the following code from the terminal to install.
$ /usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"
After installing homebrew, update the package definition.
$ brew update
[ Optional ] Run the $ echo $PATH command in the terminal to check if /usr/local/bin has been added to the system’s path. If it has not been added, update the bash_profile or bashrc or zshrc (according to your system) and add the following lines to the file. As I am using iTerm as my terminal replacement, the following command corresponds to ~/.zshrc. I use sublime text 3 for editing. You can use any editor of your choice.
$ subl ~/.zshrc
Add the following lines to the file and save it.
# Homebrew export PATH=/usr/local/bin:$PATH
Reload the .zshrc file using source command.
$ source ~/.zshrc
3. Install Python 3.6 from the official site
After installing the downloaded python file, run the following command to check the path of python3.
$ which python3 /Library/Frameworks/Python.framework/Versions/3.6/bin/python3 $ python3 --version Python 3.6.8
Install the following dependencies for OpenCV using the brew package manager. They are needed for compilation, image I/O, and optimization.
$ brew install cmake pkg-config wget $ brew install jpeg libpng libtiff openexr $ brew install eigen tbb hdf5
4. Create the python virtual environment
The next step is to install virtualenv and virtualenvwrapper using pip.
$ pip3 install virtualenv virtualenvwrapper
Configure the ~/.zshrc profile page again to set the environment variables.
# virtualenv and virtualenvwrapper export VIRTUALENVWRAPPER_PYTHON=/Library/Frameworks/Python.framework/Versions/3.6/bin/python3 source /Library/Frameworks/Python.framework/Versions/3.6/bin/virtualenvwrapper.sh
After updating the .zshrc file, reload it by using the source command.
$ source ~/.zshrc
Create a virtual environment with the name py3cv3. You can choose a suitable name for your needs.
$ mkvirtualenv py3cv3 -p python3
5. Installing Opencv 3.4.7 from Source
The only Python dependency required by OpenCV is NumPy(we will use 1.14.0), which we can install by using the following command.
$ workon py3cv3 $ pip install numpy==1.14.0
Let us compile OpenCV from the source. First, we have to download the source code for opencv and opencv_contrib ( we will use version 3.4.7). You can use the version according to your requirements. After downloading the source codes, we will unpack it and store them inside the home folder.
$ cd ~ $ wget -O opencv.zip https://github.com/opencv/opencv/archive/3.4.7.zip $ wget -O opencv_contrib.zip https://github.com/opencv/opencv_contrib/archive/3.4.7.zip $ unzip opencv.zip $ unzip opencv_contrib.zip
We will rename the extracted folder so that the version information would be removed from the extracted folder name.
$ mv opencv-3.4.7 opencv $ mv opencv_contrib-3.4.7 opencv_contrib
To prepare our compilation process, we use CMake.
Use CMake to run the build process from the source [1].
$ cd ~/opencv
$ mkdir build
$ cd build
$ cmake -D CMAKE_BUILD_TYPE=RELEASE \
-D CMAKE_INSTALL_PREFIX=/usr/local \
-D OPENCV_EXTRA_MODULES_PATH=~/opencv_contrib/modules \
-D PYTHON3_LIBRARY=`python -c 'import subprocess ; import sys ; s = subprocess.check_output("python-config --configdir", shell=True).decode("utf-8").strip() ; (M, m) = sys.version_info[:2] ; print("{}/libpython{}.{}.dylib".format(s, M, m))'` \
-D PYTHON3_INCLUDE_DIR=`python -c 'import distutils.sysconfig as s; print(s.get_python_inc())'` \
-D PYTHON3_EXECUTABLE=$VIRTUAL_ENV/bin/python \
-D BUILD_opencv_python2=OFF \
-D BUILD_opencv_python3=ON \
-D INSTALL_PYTHON_EXAMPLES=ON \
-D INSTALL_C_EXAMPLES=OFF \
-D OPENCV_ENABLE_NONFREE=ON \
-D BUILD_EXAMPLES=ON ..
Check for the output of cmake. It should look as follows:
$ make -j4
After a successful make, we can install it by using the following command.
$ sudo make install
To check the successful installation inside the system, you can search for the opencv.pc configuration file. Generally, it can be found in the following path.
/usr/local/lib/pkgconfig/opencv.pc
Run the following command to check the installed version.
$ pkg-config --modversion opencv 3.4.7 $ pkg-config --libs --cflags opencv -I/usr/local/include/opencv -I/usr/local/include -L/usr/local/lib -lopencv_stitching -lopencv_superres -lopencv_videostab -lopencv_aruco -lopencv_bgsegm -lopencv_bioinspired -lopencv_ccalib -lopencv_dnn_objdetect -lopencv_dpm -lopencv_face -lopencv_fuzzy -lopencv_hdf -lopencv_hfs -lopencv_img_hash -lopencv_line_descriptor -lopencv_optflow -lopencv_reg -lopencv_rgbd -lopencv_saliency -lopencv_stereo -lopencv_structured_light -lopencv_phase_unwrapping -lopencv_surface_matching -lopencv_tracking -lopencv_datasets -lopencv_text -lopencv_highgui -lopencv_videoio -lopencv_dnn -lopencv_plot -lopencv_xfeatures2d -lopencv_shape -lopencv_video -lopencv_ml -lopencv_ximgproc -lopencv_xobjdetect -lopencv_objdetect -lopencv_calib3d -lopencv_imgcodecs -lopencv_features2d -lopencv_flann -lopencv_xphoto -lopencv_photo -lopencv_imgproc -lopencv_core
After the installation, we have to create a symbolic link of the cv2.so file into the virtual environment. Let’s first check whether the .so file exists inside the installation path or not.
$ cd /usr/local/lib/python3.6/site-packages/cv2/python-3.6 $ ls cv2.cpython-36m-darwin.so
If we want to install multiple versions of OpenCV, its better to rename it, to avoid confusion later. After renaming, we can create the symbolic link of the .so file into the virtual environment (e.g., py3cv3).
$ sudo mv cv2.cpython-36m-darwin.so cv2.opencv3.4.7.so $ cd ~/.virtualenvs/py3cv3/lib/python3.6/site-packages $ ln -s /usr/local/lib/python3.6/site-packages/cv2/python-3.6/cv2.opencv3.4.7.so cv2.so
Finally, we can test the installation by running the following commands.
$ cd ~ $ workon py3cv3 $ python >>> import cv2 >>> cv2.__version__ '3.4.7'
If you are able the see the version of opencv, it has been successful in your system, and you are ready to go. Congratulations!!
6. Tensorflow and Keras Installation on macOS Catalina
We are going to install TensorFlow and Keras inside the virtual environment py3cv3. To activate the virtual environment, just run the following code.
$ workon py3cv3
Using pip, we will install some of the famous computer vision, machine learning, and image processing packages.
$ pip install scipy pillow $ pip install imutils h5py requests progressbar2 $ pip install scikit-learn scikit-image
Next, we will install matplotlib and update the rendering background.
$ pip install matplotlib $ mkdir ~/.matplotlib $ vim ~/.matplotlib/matplotlibrc $ echo "backend: TkAgg" >> ~/.matplotlib/matplotlibrc
If you ever face any difficulties in getting a final output in matplotlib, follow the official documentation page to set up the virtual environment for matplotlib [Link]. Now it’s the time to install TensorFlow and Keras. Follow the TensorFlow APIs page to select the required version. I am using version r1.15 here.
$ pip install tensorflow==1.15 $ pip install keras $ python >>> import tensorflow as tf >>> tf.__version__ '1.15.0' >>> import keras Using TensorFlow backend.
The Keras configuration file is present inside the home folder with a path ~/.keras/keras.json. You can open the file to have a look into the contents. Make sure the “image_data_format” field is set to channels_last and that the “backend” is set to tensorflow.
Congratulations!! You have successfully installed OpenCV on the macOS Catalina.
Random Blabbering 