电商网站分析,努力把网站建设成为,服务佳的广州网站建设,国外vps私人系统版本#xff1a;ubuntu18.04 库版本#xff1a;opencv4.6.0 opencv3.6.0 一、多版本安装前准备
1. 卸载已经安装的opencv版本[可选] 1.1 卸载从软件仓库中安装的opencv sudo apt-get purge libopencv* 1.2 卸载使用source自行编译安装的opencv
首先进入原先编译… 系统版本ubuntu18.04 库版本opencv4.6.0 opencv3.6.0 一、多版本安装前准备
1. 卸载已经安装的opencv版本[可选] 1.1 卸载从软件仓库中安装的opencv sudo apt-get purge libopencv* 1.2 卸载使用source自行编译安装的opencv
首先进入原先编译opencv的build目录在该目录下打开终端再执行以下代码。注意如果build已经删除了可以重新编译一遍该版本的opencv然后在最后一步执行sudo make uninstall sudo make uninstall
cd ..
sudo rm -r build
sudo rm -r /usr/local/include/opencv2 /usr/local/include/opencv /usr/include/opencv /usr/include/opencv2 /usr/local/share/opencv /usr/local/share/OpenCV /usr/share/opencv /usr/share/OpenCV /usr/local/bin/opencv* /usr/local/lib/libopencv* 以上可能有error可忽略最后执行如下操作 sudo apt-get --purge remove opencv-doc opencv-data python-opencv 二、安装opencv3.4.15
2.1 下载opencv3.4.15
链接1 GitHub - opencv/opencv_contrib at 3.4.15
链接2Releases - OpenCV
2.2 下载opencv_contrib
注意这里要与上面下载的opencv版本一致 链接
GitHub - opencv/opencv_contrib at 3.4.15 2.3 编译源码
2.3.1 下载好后将opencv_contrib放到解压后opencv文件夹里面
2.3.2 安装依赖项 sudo apt-get updatesudo apt-get upgradesudo apt-get install build-essentialsudo apt-get install cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-devsudo apt-get install python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev2.3.3 cmake编译
2.3.3.1 在该文件夹opencv-3.4.15下打开终端
后续步骤参见下文3.3.3节
三、安装opencv4.6.0
3.1 下载opencv4.6.0
链接1GitHub - opencv/opencv at 4.6.0
链接2Releases - OpenCV
3.2 下载opencv_contrib
注意这里要与上面下载的opencv版本一致
链接https://github.com/opencv/opencv_contrib/tree/4.6.0
3.3 编译源码 3.3.1 两个都下载好以后把opencv_ opencv_contrib放到解压后opencv文件夹里面
3.3.2 安装依赖 sudo apt-get updatesudo apt-get upgradesudo apt-get install build-essentialsudo apt-get install cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-devsudo apt-get install python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev3.3.3 cmake编译
3.3.3.1 在该文件夹opencv-4.6.0下打开终端 mkdir build cd build
(1) 不编译opencv_contrib、cuda
注意CMAKE_INSTALL_PREFIX/usr/local/opencv-4.6.0
cmake -D CMAKE_INSTALL_PREFIX/usr/local/opencv-4.6.0 -D CMAKE_BUILD_TYPERelease -D OPENCV_GENERATE_PKGCONFIGON -D OPENCV_ENABLE_NONFREETrue ..注解 1. CMAKE_INSTALL_PREFIX 是opencv的安装地址 默认安装在 usr/local 2. CMAKE_BUILD_TYPE 是opencv安装的版本Release和Debug两种可选默认安装Release 3. OPENCV_ENABLE_NONFREE 是否使用部分被申请了专利的算方法 这里选True的话就可以使用了 4. OPENCV_GENERATE_PKGCONFIG 强烈建议开启这个 设置为ON OPENCV_GENERATE_PKGCONFIG 因为opencv4默认不生成.pc文件所以加上这句用于生成opencv4.pc文件支持pkg-config功能。opencv4版本及以上 这里用ON. (2) 编译opencv_contrib不编译cuda
注意CMAKE_INSTALL_PREFIX/usr/local/opencv-4.6.0
OPENCV_EXTRA_MODULES_PATH../opencv_contrib-4.6.0/modules
cmake -D CMAKE_BUILD_TYPERELEASE \-D CMAKE_INSTALL_PREFIX/usr/local/opencv-4.6.0 \-D INSTALL_PYTHON_EXAMPLESON \-D INSTALL_C_EXAMPLESOFF \-D OPENCV_GENERATE_PKGCONFIGON \-D OPENCV_ENABLE_NONFREEON \-D OPENCV_DNN_CUDAON \-D ENABLE_FAST_MATH1 \-D OPENCV_EXTRA_MODULES_PATH../opencv_contrib-4.6.0/modules \-D HAVE_opencv_python3ON \-D PYTHON_EXECUTABLEusr/bin/python3 \-D BUILD_EXAMPLESON ..cmake -DCMAKE_BUILD_TYPERELEASE -DOPENCV_EXTRA_MODULES_PATH ../opencv_contrib-4.6.0/modules -DCMAKE_INSTALL_PREFIX/usr/local/opencv-4.6.0 -DWITH_CUDAOFF -DBUILD_EXAMPLESON -DBUILD_DOCSOFF -DBUILD_PERF_TESTSOFF -DBUILD_TESTSOFF -DOPENCV_GENERATE_PKGCONFIGYES -DOPENCV_ENABLE_NONFREETrue -DBUILD_opencv_worldOFF ..
可能出现的错误
ippicv_2020_win_intel64_20191018_general.zip等文件下载失败的解决办法找了好久还是这位老哥的方法解决了问题-CSDN博客
(3) 编译opencv_contrib、cuda
cmake -D CMAKE_BUILD_TYPERELEASE \-D CMAKE_INSTALL_PREFIX/usr/local/opencv-4.6.0 \-D INSTALL_PYTHON_EXAMPLESON \-D INSTALL_C_EXAMPLESOFF \-D OPENCV_GENERATE_PKGCONFIGON \-D OPENCV_ENABLE_NONFREEON \-D WITH_CUDAON \-D WITH_CUDNNON \-D OPENCV_DNN_CUDAON \-D ENABLE_FAST_MATH1 \-D CUDA_FAST_MATH1 \-D CUDA_ARCH_BIN7.5 \-D WITH_CUBLAS1 \-D OPENCV_EXTRA_MODULES_PATH../opencv_contrib-4.6.0/modules \-D HAVE_opencv_python3ON \-D PYTHON_EXECUTABLEusr/bin/python3 \-D BUILD_EXAMPLESON .. 从上WITH_CUDA和WITH_CUDNN项指的是用CUDA和cuDNN。OPENCV_DNN_CUDA项表示我们要编译后的dnn模块支持CUDA。ENABLE_FAST_MATH, CUDA_FAST_MATH和WITH_CUBLAS项是为了优化运算而开启的。 3.3.3.2 cmake结束后make
注终端输入nproc 命令可以查看自己电脑有多少线程。我的电脑有8个。
sudo make -j8
3.3.3.3 安装指令
sudo make install3.3.3.4 配置环境
sudo gedit /etc/ld.so.conf.d/opencv.conf
新增 /usr/local/opencv-4.6.0/lib
保存后终端输入
sudo ldconfig
更新共享链接库
sudo gedit /etc/bash.bashrc
在文末新增 PKG_CONFIG_PATH$PKG_CONFIG_PATH:/usr/local/opencv-4.6.0/lib/pkgconfig export PKG_CONFIG_PATH 保存后退出终端输入以下命令使配置生效
source /etc/bash.bashrc更新一下
sudo updatedb查看是否安装成功成功则会输出版本号
pkg-config --modversion opencv四、配置菜单选项
打开终端输入以下指令
sudo update-alternatives --install /usr/local/opencv opencv /usr/local/opencv-3.4.15/ 3415
sudo update-alternatives --install /usr/local/opencv opencv /usr/local/opencv-4.6.0/ 4600 切换指令
sudo update-alternatives --config opencv
