一：Opencv + Cuda源码的下载、编译

1.卸载jetson上自带的无cuda加速Opencv

运行如下代码即可：

sudo apt purge libopencv*
sudo apt autoremove
sudo apt update

（不建议）:不放心的话可以在根目录下搜索opencv关键词，删去多余文件

2.安装Opencv依赖库

运行如下代码：

sudo apt install -y build-essential checkinstall cmake pkg-config yasm git gfortran
sudo apt update
sudo apt install -y libgstreamer1.0-dev libgstreamer-plugins-base1.0-dev
sudo apt install -y libjpeg8-dev libjasper-dev libpng12-dev libtiff5-dev libavcodec-dev libavformat-dev libswscale-dev libdc1394-22-dev libxine2-dev libv4l-dev
sudo apt install -y libgstreamer1.0-dev libgstreamer-plugins-base1.0-dev libgtk2.0-dev libtbb-dev libatlas-base-dev libfaac-dev libmp3lame-dev libtheora-dev libvorbis-dev libxvidcore-dev libopencore-amrnb-dev libopencore-amrwb-dev x264 v4l-utils
sudo apt install python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev

坑一：
笔者建议先更换一下Jetson自带的镜像源，否则过程中可能会出现各种丢失问题，更换步骤如下：

修改/etc/apt/sources.list，改为下面的源

deb [arch=arm64] https://mirrors.ustc.edu.cn/ubuntu-ports/ focal main restricted universe multiverse
# deb-src http://mirrors.ustc.edu.cn/ubuntu-ports/ focal main main restricted universe multiverse
deb [arch=arm64] https://mirrors.ustc.edu.cn/ubuntu-ports/ focal-updates main restricted universe multiverse
# deb-src http://mirrors.ustc.edu.cn/ubuntu-ports/ focal-updates main restricted universe multiverse
deb [arch=arm64] https://mirrors.ustc.edu.cn/ubuntu-ports/ focal-backports main restricted universe multiverse
# deb-src http://mirrors.ustc.edu.cn/ubuntu-ports/ focal-backports main restricted universe multiverse
deb [arch=arm64] https://mirrors.ustc.edu.cn/ubuntu-ports/ focal-security main restricted universe multiverse
# deb-src http://mirrors.ustc.edu.cn/ubuntu-ports/ focal-security main restricted universe multiverse

最后sudo apt-get update即可。

坑二：
运行倒数第三条命令时，若遇到“ 无法定位libjasper-dev软件包 “报错，执行如下命令即可安装：

sudo add-apt-repository "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu-ports/ xenial main multiverse restricted universe"
sudo apt update
sudo apt install libjasper1 libjasper-dev

坑三：
运行倒数第三条命令时，若遇到“ libpng12-dev没有可安装候选 “报错，是因为ubuntu14以上已经废弃，代替为libpng-dev，安装命令如下：

sudo apt install libpng-dev

3.下载 OpenCV 和 opencv_contrib 源码

分别前往 opencv 和 Opencv_contrib 的 github 仓库下载 Latest release 源码 ( 👈 点击超链接跳转)

点击Source code（zip）下载Opencv源码

点击zip下载opencv_contrib源码

注意： OpenCV 和 opencv_contrib 的版本要对应正确

4.编译安装 OpenCV、opencv_contrib

• 解压opencv-4.7.0.zip 和 opencv_contrib-4.7.0.zip 得到两个文件夹，然后将opencv_contrib-4.7.0文件夹放在opencv-4.7.0文件中！
• 执行如下命令

cd opencv-4.5.3
mkdir build
cd build

• 执行cmake

cmake \
-DCMAKE_BUILD_TYPE=Release \
-DCMAKE_INSTALL_PREFIX=/usr/local \
-DOPENCV_ENABLE_NONFREE=1 \
-DBUILD_opencv_python2=1 \
-DBUILD_opencv_python3=1 \
-DWITH_FFMPEG=1 \
-DCUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda \
-DCUDA_ARCH_BIN=8.7 \
-DCUDA_ARCH_PTX=8.7 \
-DWITH_CUDA=1 \
-DENABLE_FAST_MATH=1 \
-DCUDA_FAST_MATH=1 \
-DWITH_CUBLAS=1 \
-DOPENCV_GENERATE_PKGCONFIG=1 \
-DOPENCV_EXTRA_MODULES_PATH=../opencv_contrib-4.7.0/modules \
..

注意点1：

DCUDA_ARCH_BIN、DCUDA_ARCH_PTX要改成cuda对应的容量值，运行如下指令查看：

cd /usr/local/cuda/samples/1_Utilities/deviceQuery
sudo make
./deviceQuery

打印出如下信息，我的cuda11.4对应的系数是8.7

注意点2：

DOPENCV_EXTRA_MODULES_PATH要改成opencv_contrib-4.7.0的对应地址。

• 进行 make前，事先调整 CPU 性能模式打开尽量多的核心，如下命令打印核心数：

cat /proc/stat | grep cpu[0-9] -c # 查看线程数，我的是8，如果占有较少，网上有指令教学

• 进行make编译，1小时左右完成

make -j4 # 四线程编译

• 编译完成后，进行安装

sudo make install

至此，Opencv的安装全部完成，如下显示yes:

二：Opencv 的环境配置

具体可以参考本人之前的Ubuntu20.04 配置Opencv+VScode+cmake，里面有配置细节，这里我基本展示一下：

--------------------------------------------------------------------------------------------
环境配置
---------------------------------------------------------------------------------------------

#配置pkg-config环境
#opencv4.pc文件的默认路径为/usr/local/lib/pkgconfig/opencv4.pc，如果此目录下没有，可以使用以下命令搜索：
find / -iname opencv4.pc

#将路径加入到PKG_CONFIG_PATH，打开是个空文件
sudo vim /etc/profile.d/pkgconfig.sh

#在文件后面加入下面一行：
export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig:$PKG_CONFIG_PATH

#保存并退出后激活，激活后关闭终端
source /etc/profile

#验证是否成功，这里的代码貌似有些问题，可以跳过。。。
$ pkg-config --libs opencv4
-L/usr/local/lib -lopencv_dnn -lopencv_highgui -lopencv_ml -lopencv_objdetect -lopencv_photo 
-lopencv_stitching -lopencv_video -lopencv_calib3d -lopencv_features2d -lopencv_flann 
-lopencv_videoio -lopencv_imgcodecs -lopencv_imgproc -lopencv_core

#配置动态库环境，打开文件（可能为空文件）
sudo vim /etc/ld.so.conf.d/opencv4.conf

#在该文件末尾加上OpenCV的lib路径
/usr/local/lib

#使配置的路径生效
sudo ldconfig

#测试安装成功与否，输出版本信息
pkg-config --modversion opencv4

三：Vscode 中的Opencv环境配置

话不多说，直接上三种json配置文件内容：

c_cpp_properties.json：

{
    "configurations": [
        {
            "name": "Linux",
            "includePath": [
                "${workspaceFolder}/**",
                "/usr/local/include/opencv4"
            ],
            "defines": [],
            "compilerPath": "/usr/bin/clang-8",
            "cStandard": "c17",
            "cppStandard": "c++14",
            "intelliSenseMode": "linux-clang-arm64"
        }
    ],
    "version": 4
}

task.json：

{
    "version": "2.0.0",
    "tasks": [
        {
            "type": "cppbuild",
            "label": "C/C++: g++ 生成活动文件",
            "command": "/usr/bin/g++",
            "args": [
                "-fdiagnostics-color=always",
                "-g",
                "-std=c++11",
                "${file}",
                "-o",
                "${fileDirname}/a.out",
                "-I",
                "${workspaceFolder}/",
                "-I",
                "/usr/local/include/",
                "-I",
                "/usr/local/include/opencv4/",
                "-I",
                "/usr/local/include/opencv4/opencv2",
                "-L",
                "/usr/local/lib",
                "-L",
                "/usr/local/lib/libopencv_*",
            ],
            "options": {
                "cwd": "${fileDirname}"
            },
            "problemMatcher": [
                "$gcc"
            ],
            "group": "build",
            "detail": "编译器: /usr/bin/g++"
        }
    ]
}

launch.json：
（调试文件其实什么都未添加）

{
    "version": "0.2.0",
    "configurations": [
        {
            "name": "(gdb) 启动",
            "type": "cppdbg",
            "request": "launch",
            "program": "${fileDirname}/a.out",
            "preLaunchTask": "C/C++: g++ 生成活动文件",
            "args": [],
            "stopAtEntry": false,
            "cwd": "${fileDirname}",
            "environment": [],
            "externalConsole": false,
            "MIMode": "gdb",
            "setupCommands": [
                {
                    "description": "为 gdb 启用整齐打印",
                    "text": "-enable-pretty-printing",
                    "ignoreFailures": true
                },
                {
                    "description": "将反汇编风格设置为 Intel",
                    "text": "-gdb-set disassembly-flavor intel",
                    "ignoreFailures": true
                }
            ],
            "miDebuggerArgs": "-q -ex quit; wait() { fg >/dev/null; }; /bin/gdb -q --interpreter=mi",
        }

    ] 
}

四：Opencv的demo测试

最后，检测一下是否能跑通：

#include<opencv2/opencv.hpp>
#include<iostream>
#include<iostream>
using namespace std;
using namespace cv;
int main(int argc,char** argv){
  Mat src = imread("./ceshi.jpg"); //读取图片
  Mat src2;
  if (src.empty()) {                     //判断是否找到图片
    printf("没有找到图片 ");           //输出文字
    return -1;
  }
  else
  {
    namedWindow("input",WINDOW_AUTOSIZE);// 显示窗口命名为input ；WINDOW_AUTOSIZE显示大小为图片自定义大小，且不可以更改大小
    resize(src, src2, Size(612, 612));
    imshow("input",src2); //显示
    waitKey(0);//显示的毫秒时间，如果函数参数<=0表示一直显示。>0表示显示的时间
    destroyAllWindows();

  }
  cout << "success" << endl;
  return 0;
}

我在jetson终端上跑有一个warning“ Failed to load module canberra-gtk-module ”，运行“sudo apt install libcanberra-gtk-module”重装该模块即可。

完结撒化 ！！！！！

---

---

我是刚出cv泥潭不久，又入jetson终端部署泥潭的江南咸鱼，希望大家能从一而终，不留遗憾！！！

觉得文章有用的话，各位客官点个赞吧！！