一、安装依赖

    sudo apt install build-essential cmake libtbb2 libttb-dev libgtk-3-dev
    sudo apt install python3 python3-dev pkg-config
    sudo apt install libjpeg-dev libpng-dev libtiff-dev libjasper-dev
    sudo apt install zlib1g zlibe1g-dev libavcodec-dev libavformat-dev libswscale-dev
    sudo apt install  libv4l-dev libxvidcore-dev libx264-dev libatlas-base-dev libdc1394-22-dev
    sudo apt install libgstreamer-plugins-base1.0-dev libgstreamer1.0-dev
    sudo apt install gfortran openexr libopenexr-dev

二、下载opencv和opencv_contrib安装包

1.opencv:https://link.csdn.net/?target=https%3A%2F%2Fgithub.com%2Fopencv%2Fopencv
2.opencv_contrib:mirrors / opencv / opencv_contrib · GitCode
（注：opencv和opencv_contrib版本应一致）
三、编译opencv和opencv_contrib

        将opencv_contrib解压后放到opencv文件夹内，在opencv文件夹内新建build文件

        通过这个下载镜像下载:boostdesc_bgm.i,vgg_generated_48.i等.rar （提取码：p50x），放在 opencv_contrib/modules/xfeatures2d/src/ 路径下。

 进入build文件，首先进行预编译

cmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local -D OPENCV_GENERATE_PKGCONFIG=ON -D OPENCV_EXTRA_MODULES_PATH=../opencv_contrib/modules ..

    CMAKE_BUILD_TYPE：编译类型
    CMAKE_INSTALL_PREFIX：安装路径
    OPENCV_GENERATE_PKGCONFIG：生成opencv.pc文件
    OPENCV_EXTRA_MODULES_PATH：opencv_ccontrib编译

其次进行编译

make -j6

在编译过程中会出现两种问题：

1.问题：fatal error: features2d/test/test_detectors_regression.impl.hpp: 没有那个文件或目录

解决方案：将opencv / modules / features2d复制，然后粘贴到build目录中。

2.问题：fatal error: opencv2/xfeatures2d.hpp: 没有那个文件或目录

解决方案：将opencv_contrib/modules/xfeatures2d/include/opencv2中的xfeatures2d文件夹和  xfeatures2d.hpp文件复制到opencv/build/opencv2路径下

重新make，编译成功后，进行安装

sudo make install

四、测试

查看opencv的版本：opencv_version