Install the latest OpenCV4.3 under Centos8

Preface##

OpenCV is a cross-platform Computer Vision and Machine Learning software library based on the BSD license (open source) release, written in C++ language, it has C++, [Python] (https://baike.baidu.com/item/Python/407313), Java and MATLAB interfaces, and supports Windows, Linux, Android and Mac OS. OpenCV mainly tends to real-time visual applications, and uses MMX and SSE instructions when available. Provide support for C#, Ch, Ruby, GO.

installation##

Centos8 can install opencv in two ways

dnf installation

One is installed through dnf, which is simple to install, the version is 3.4.1, and the version is lower but stable.

dnf install opencv opencv-devel opencv-python

Execute the above command to install successfully, check whether the installation is successful through the command

pkg-config --modversion opencv

If it returns

3.4.1

Prove that the installation was successful.

Source installation opencv

If you want to install the latest version, you can install it through the source code, which is recommended here.

1、 Install dependencies

 dnf install epel-release git gcc gcc-c++ cmake3 qt5-qtbase-devel \

　python3 python3-devel python3-pip cmake python3-devel python3-numpy \

　gtk2-devel libpng-devel jasper-devel openexr-devel libwebp-devel \

　libjpeg-turbo-devel libtiff-devel tbb-devel libv4l-devel \

 freeglut-devel mesa-libGL mesa-libGL-devel \

　boost boost-thread boost-devel gstreamer1-plugins-base

It should be noted here that eigen3-devel cannot be installed directly through dnf, it needs to be installed through the following command

dnf --enablerepo=PowerTools install eigen3-devel

2、 Download OpenCV and OpenCV contrib library#####

Create a temporary directory

mkdir -p /tmp/opencv && cd /tmp/opencv
//This is slower, recommend the following domestic sources
git clone https://github.com/opencv/opencv.git
git clone https://github.com/opencv/opencv_contrib.git

**Note **, the download speed of github is too slow, the reason is not listed. If you want to be faster, you can try to set /etc/hosts to increase

13.229.188.59 github.com
52.216.146.19 github-cloud.s3.amazonaws.com
199.232.4.133 raw.githubusercontent.com

The actual situation is not much faster, the download speed makes people crash. Here to recommend domestic sources (update github official code daily, caring:)), execute the following commands.

git  clone https://gitee.com/mirrors/opencv.git
git  clone https://gitee.com/mirrors/opencv_contrib.git

3、 Create a temporary directory and enter

cd /tmp/opencv/opencv && mkdir build && cd build

4、 Configure OpenCV build, run the following command

cmake3 -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=/tmp/opencv/opencv_contrib/modules 　-D BUILD_EXAMPLES=ON ..

After the execution is complete, the output will be as follows:

- - Build files have been written to:/tmp/opencv/opencv/build

5、 Compile and run the following command

make -j1

Please note here, please modify the value of -j according to the processor, you can type nproc to view, the number of cpu in my system is 1.

6、 Install the OpenCV library, run the following command

make install

7、 Create the symlink opencv4.pc file to the /usr/share/pkgconfig directory, and then run ldconfig to rebuild the library cache

ln -s /usr/local/lib64/pkgconfig/opencv4.pc /usr/share/pkgconfig/
ldconfig

8、 Check OpenCV version

pkg-config --modversion opencv4

Will return normally

4.3.0

9、 To verify that the Python cv2 module is running, please run the following command

python3 -c "import cv2; print(cv2.__version__)"

Will return if there are no problems

4.4.0- pre