OpenCV 开发环境配置
| Name | Version |
|---|---|
| 系统版本 | Ubuntu20.04LTS |
| OpenCV | 4.7.0 |
| CUDA | 12.0 |
| CUDNN | 8.8.1 |
OpenCV
从源码编译OpenCV - 开启GPU支持
安装前先安装显卡驱动以及CUDA、CUDNN
可以参考这里👉
获取 OpenCV 源码
从官网下载压缩包
下载完成后解压
tar zxvf opencv-4.7.0.tar.gz ~/workspace/opencv/4.7.0/
tar zxvf opencv_contrib-4.7.0.tar.gz ~/workspace/opencv/4.7.0/
tar zxvf opencv_extra-4.7.0.tar.gz ~/workspace/opencv/4.7.0/
安装各种依赖
sudo apt-get install \
build-essential \
cmake \
git \
libgtk2.0-dev \
pkg-config \
libavcodec-dev \
libavformat-dev \
libswscale-dev \
libtbb2 \
libtbb-dev \
libjpeg-dev \
libpng-dev \
libtiff-dev \
libdc1394-22 \
libglew-dev \
zlib1g-dev \
libavutil-dev \
libpostproc-dev \
libeigen3-dev \
libopenexr-dev \
libwebp-dev \
libgtk-3-dev \
qtbase5-dev \
libgstreamer1.0-dev \
libgstreamer-plugins-base1.0-dev
libjasper-dev
sudo add-apt-repository "deb http://security.ubuntu.com/ubuntu xenial-security main"
sudo apt update
sudo apt install libjasper-dev
编译
新建 build/ 目录存放编译生成的文件
cd ~/workspace/opencv/4.7.0/;mkdir build;cd build
- [可选] 配置并生成
makefile过程中会下载一些依赖内容，建议开启网络代理
export http_proxy=socks5://127.0.0.1:7890;export https_proxy=socks5://127.0.0.1:7890;export all_proxy=socks5://127.0.0.1:7890
- [可选] 如果你的 Python 环境使用
Anaconda或者是其他的 python 环境管理应用，记得启用你的虚拟环境，这里会配置 python 包的路径，编译支持 CUDA 的 opencv-python 版本
conda activate {env-name}
配置 cmake 参数，修改 opencv_extra 和 opencv_contrib 的路径
cmake -DCMAKE_BUILD_TYPE=Release \
-DCMAKE_INSTALL_PREFIX=/usr/local \
-DOPENCV_GENERATE_PKGCONFIG=YES \
-DWITH_QT=ON \
-DWITH_OPENGL=ON \
-DWITH_TBB=ON \
-DWITH_OPENCL=ON \
-DWITH_FFMPEG=ON \
-DWITH_CUDA=ON \
-DOPENCV_DNN_CUDA=ON \
-DCUDA_ARCH_BIN=5.0,5.2,6.0,6.1,7.0,7.5,8.0,8.6,8.9,9.0 \
-DENABLE_FAST_MATH=ON \
-DCUDA_FAST_MATH=ON \
-DWITH_CUBLAS=ON \
-DWITH_GTK=ON \
-DCUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda \
-DCUDA_ARCH_PTX="" \
-DBUILD_WITH_DEBUG_INFO=ON \
-DBUILD_opencv_python3=ON \
-DPYTHON3_NUMPY_INCLUDE_DIRS=$(python3 -c "import numpy; print(numpy.get_include())") \
-DPYTHON3_PACKAGES_PATH=$(python3 -c "from distutils.sysconfig import get_python_lib; print(get_python_lib())") \
-DPYTHON3_LIBRARY=$(python3 -c "from distutils.sysconfig import get_config_var;from os.path import dirname,join ; print(join(dirname(get_config_var('LIBPC')),get_config_var('LDLIBRARY')))") \
-DINSTALL_TESTS=ON \
-DOPENCV_TEST_DATA_PATH=../opencv_extra-4.7.0/testdata \
-DOPENCV_EXTRA_MODULES_PATH=../opencv_contrib-4.7.0/modules \
../opencv-4.7.0
配置完成后的部分终端输出内容
-- NVIDIA CUDA: YES (ver 12.0, CUFFT CUBLAS FAST_MATH)
-- NVIDIA GPU arch: 50 52 60 61 70 75 80 86 89 90
-- NVIDIA PTX archs:
--
-- cuDNN: YES (ver 8.8.1)
--
-- OpenCL: YES (no extra features)
-- Include path: ~/workspace/opencv/4.7.0/opencv-4.7.0/3rdparty/include/opencl/1.2
-- Link libraries: Dynamic load
--
-- Python 3:
-- Interpreter: ~/anaconda3/envs/opencv-cuda/bin/python3 (ver 3.8.16)
-- Libraries: ~/anaconda3/envs/opencv-cuda/lib/libpython3.8.so (ver 3.8.16)
-- numpy: ~/anaconda3/envs/opencv-cuda/lib/python3.8/site-packages/numpy/core/include (ver 1.24.2)
-- install path: ~/anaconda3/envs/opencv-cuda/lib/python3.8/site-packages/cv2/python-3.8
--
-- Python (for build): ~/anaconda3/envs/opencv-cuda/bin/python3
--
-- Java:
-- ant: NO
-- JNI: NO
-- Java wrappers: NO
-- Java tests: NO
--
-- Install to: /usr/local
-- -----------------------------------------------------------------
--
-- Configuring done
-- Generating done
生成完之后开始编译
NUM_CPU=$(nproc)
make -j$(($NUM_CPU - 1))
信息
这里编译时间更久，你可以叫几个朋友一起喝茶 (= =!)
等待编译完成后
sudo make install
- 将编译好的
.so文件以及头文件安装到/usr/local目录下
测试 opencv-python 安装
python -c "import cv2;print(cv2.getBuildInformation())"
python 包安装成功
配置环境变量
sudo gedit /etc/ld.so.conf.d/opencv.conf
- 打开
opencv.conf在其中添加/usr/local/lib
sudo gedit /etc/bash.bashrc
- 打开
bash.bashrc在最后面添加
PKG_CONFIG_PATH=$PKG_CONFIG_PATH:/usr/local/lib/pkgconfig
export PKG_CONFIG_PATH
sudo ldconfig使配置生效
测试安装
pkg-config --libs opencv4
安装完成