Ubuntu下配置opencv环境

标签：lopencv 配置 LIBRARY opencv cuda Ubuntu dev CUDA

安装准备

安装cmake

sudo apt-get install cmake

安装依赖环境

sudo apt install  build-essential
 
sudo apt install cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev  
 
sudo apt install python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev

第三行中，可能会出现【无法定位软件包libjasper-dev 的错误提示】
sudo add-apt-repository "deb http://security.ubuntu.com/ubuntu xenial-security main" sudo apt update sudo apt upgrade sudo apt install libjasper1 libjasper-dev

下载opencv

https://opencv.org/releases/

安装

解压

在opencv3文件夹下新建build文件夹

cmake

进入build/

sudo cmake -D CMAKE_BUILD_TYPE=Release -D CMAKE_INSTALL_PREFIX=/usr/local ..

【由于网络问题无法下载ippicv_linux_20151201.tgz文件的解决办法】

找到3rdparty/ippicv/downloader.cmake文件
set(OPENCV_ICV_URL "https://raw.githubusercontent.com/opencv/opencv_3rdparty/${IPPICV_BINARIES_COMMIT}/ippicv") set(IPPICV_BINARIES_COMMIT "81a676001ca8075ada498583e4166079e5744668")
通过这两个语句拼接得到下载地址：https://raw.githubusercontent.com/opencv/opencv_3rdparty/81a676001ca8075ada498583e4166079e5744668/ippicv/ippicv_linux_20151201.tgz
将下载好的文件上传到服务器
替换路径为"file:///root
【The following variables are used in this project, but they are set to NOTFOUND.解决方案】
cuda9不再支持2.0架构
1.在/home/mario/Projects/opencv-3.1.0/cmake下找到FindCUDA.cmake文件
1. 找到行：
find_cuda_helper_libs(nppi)
然后替换为：
find_cuda_helper_libs(nppial) find_cuda_helper_libs(nppicc) find_cuda_helper_libs(nppicom) find_cuda_helper_libs(nppidei) find_cuda_helper_libs(nppif) find_cuda_helper_libs(nppig) find_cuda_helper_libs(nppim) find_cuda_helper_libs(nppist) find_cuda_helper_libs(nppisu) find_cuda_helper_libs(nppitc)
2. 找到行：
set(CUDA_npp_LIBRARY "${CUDA_nppc_LIBRARY};${CUDA_nppi_LIBRARY};${CUDA_npps_LIBRARY}")
然后替换为：（注意后面的""，实际输入的时候把这个反斜杠去掉，弄成一行，这里只是作为显示换行加的""）
set(CUDA_npp_LIBRARY "${CUDA_nppc_LIBRARY};${CUDA_nppial_LIBRARY};${CUDA_nppicc_LIBRARY};\ ${CUDA_nppicom_LIBRARY};${CUDA_nppidei_LIBRARY};${CUDA_nppif_LIBRARY};${CUDA_nppig_LIBRARY};\ ${CUDA_nppim_LIBRARY};${CUDA_nppist_LIBRARY};${CUDA_nppisu_LIBRARY};${CUDA_nppitc_LIBRARY};${CUDA_npps_LIBRARY}")
3. 找到行：
unset(CUDA_nppi_LIBRARY CACHE)
然后替换为：
unset(CUDA_nppial_LIBRARY CACHE) unset(CUDA_nppicc_LIBRARY CACHE) unset(CUDA_nppicom_LIBRARY CACHE) unset(CUDA_nppidei_LIBRARY CACHE) unset(CUDA_nppif_LIBRARY CACHE) unset(CUDA_nppig_LIBRARY CACHE) unset(CUDA_nppim_LIBRARY CACHE) unset(CUDA_nppist_LIBRARY CACHE) unset(CUDA_nppisu_LIBRARY CACHE) unset(CUDA_nppitc_LIBRARY CACHE)
在/home/mario/Projects/opencv-3.1.0/cmake下找到OpenCVDetectCUDA.cmake文件
找到以下几行：
... set(__cuda_arch_ptx "") if(CUDA_GENERATION STREQUAL "Fermi") set(__cuda_arch_bin "2.0") elseif(CUDA_GENERATION STREQUAL "Kepler") set(__cuda_arch_bin "3.0 3.5 3.7") ...
修改为：
... set(__cuda_arch_ptx "") if(CUDA_GENERATION STREQUAL "Kepler") set(__cuda_arch_bin "3.0 3.5 3.7") elseif(CUDA_GENERATION STREQUAL "Maxwell") set(__cuda_arch_bin "5.0 5.2") ...
找到cudafp16.h的头文件/usr/local/cuda-9.0/include/cudafp16.h，将这个文件拷贝到opencv目录的cudev下opencv-3.2.0/modules/cudev/include/opencv2/cudev，然后在该目录下的common.hpp的文件添加:
#include <cuda_fp16.h>

编译

【Unsupported gpu architecture 'compute_20’解决方案】
cmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local -D CUDA_GENERATION=Kepler ..
hint: 添加Kepler

安装

sudo make install

配置环境

打开/etc/ld.so.conf在文件中加上一行

include /usr/loacal/lib

sudo ldconfig

打开/etc/bash.bashrc，在文件中添加

PKG_CONFIG_PATH=$PKG_CONFIG_PATH:/usr/local/lib/pkgconfig
export PKG_CONFIG_PATH

source /etc/bash.bashrc

检验

pkg-config opencv --modversion

Package opencv was not found in the pkg-config search path.解决方案
创建opencv.pc文件
cd /usr/local/lib
sudo mkdir pkgconfig
cd pkgconfig
sudo touch opencv.pc
添加如下信息
prefix=/usr/local
exec_prefix=${prefix}
includedir=${prefix}/include
libdir=${exec_prefix}/lib

Name: opencv
Description: The opencv library
Version:4.0.1
Cflags: -I${includedir}/opencv4
Libs: -L${libdir} -lopencv_shape -lopencv_stitching -lopencv_objdetect -lopencv_superres -lopencv_videostab -lopencv_calib3d -lopencv_features2d -lopencv_highgui -lopencv_videoio -lopencv_imgcodecs -lopencv_video -lopencv_photo -lopencv_ml -lopencv_imgproc -lopencv_flann  -lopencv_core
~
将文件导出到环境变量
export  PKG_CONFIG_PATH=/usr/local/lib/pkgconfig

Resources

https://zhuanlan.zhihu.com/p/76737748

标签：lopencv,配置,LIBRARY,opencv,cuda,Ubuntu,dev,CUDA
From： https://blog.51cto.com/doubleZ/5986595

目录