Opencv 使用cmake和mingw 编译安装

末蓝、 2022-05-21 02:23 390阅读 0赞

## 转自 [opencv 使用cmake he mingw编译安装][opencv _cmake he mingw] ##

## 软件环境 ##

*  Windows-10-64bit
 *  [Qt-5.9.3][]
 *  [MinGW-5.3.0-32bit][]
 *  [CMake-3.9.2][]
 *  [OpenCV-3.3.1 / 3.4.1(适用)][OpenCV-3.3.1 _ 3.4.1]

## 编译好的 OpenCV（MinGW 版）： ##

[Github . huihut/OpenCV-MinGW-Build][Github . huihut_OpenCV-MinGW-Build]

## OpenCV 的 MSVC 版及 MinGW 版 ##

### MSVC 版 ###

下载的 OpenCV 文件夹会有：

*  build （已编译好的库）
 *  sources （源码）

使用 MSVC 的话，直接在

build/x64/vc14

里面就有了，配置好路径就可以使用。

### MinGW 版 ###

OpenCV 没有为我们编译好 MinGW 版，所以我们只能自己编译，下面就是介绍 MinGW 版的编译流程。

也可以直接下载使用我编译好了的 OpenCV （在上文）。

## 安装及配置 Qt、MinGW、CMake ##

### 安装 ###

CMake 自行安装，Qt 和 MinGW 可以直接使用`qt-opensource-windows-x86-5.9.3.exe` 安装包安装，注意选择安装的`组件(components)`的时候勾选 `MinGW` :

*  Qt-Qt5.9-MingGW 5.3.0 32 bit
 *  Qt-Tools-MinGW 5.3.0

### 配置 Qt、MinGW ###

安装好后打开QtCreator，在`工具`\-`选项`\-`构建和运行`\-`构建套件`，选中`Desktop Qt 5.9.3 MinGW 32bit`，设为默认，OK。

## 添加 MinGW 到环境变量 ##

为系统变量 `Path` 添加 `E:\Qt\Qt5.9.3\Tools\mingw530_32\bin`

## 使用 CMake 生成 OpenCV 的 Makefile ##

打开 cmake-gui，设置源码和生成路径：

*  Where is the source code: `E:/OpenCV_3.3.1/opencv/sources`
 *  Where to build the binaries: `E:/OpenCV_3.3.1/opencv-build`

点击 Configure，设置编译器

*  Specify the generator for this project: `MinGW Makefiles`
 *  Specify native compilers
 *  Next
 *  Compilers C: `E:\Qt\Qt5.9.3\Tools\mingw530_32\bin\gcc.exe`
 *  Compilers C++: `E:\Qt\Qt5.9.3\Tools\mingw530_32\bin\g++.exe`
 *  Finish

编译配置：

*  勾选 `WITH_QT`
 *  勾选 `WITH_OPENGL`
 *  不勾选 `WITH_IPP`

点击 Configure，再次配置：

*  设置 `QT_MAKE_EXECUTABLE` 为 `E:\Qt\Qt5.9.3\5.9.3\mingw53_32\bin\qmake.exe`
 *  设置 `Qt5Concurrent_DIR` 为 `E:\Qt\Qt5.9.3\5.9.3\mingw53_32\lib\cmake\Qt5Concurrent`
 *  设置 `Qt5Core_DIR` 为 `E:\Qt\Qt5.9.3\5.9.3\mingw53_32\lib\cmake\Qt5Core`
 *  设置 `Qt5Gui_DIR` 为 `E:\Qt\Qt5.9.3\5.9.3\mingw53_32\lib\cmake\Qt5Gui`
 *  设置 `Qt5Test_DIR` 为 `E:\Qt\Qt5.9.3\5.9.3\mingw53_32\lib\cmake\Qt5Test`
 *  设置 `Qt5Widgets_DIR` 为 `E:\Qt\Qt5.9.3\5.9.3\mingw53_32\lib\cmake\Qt5Widgets`
 *  设置 `Qt5OpenGL_DIR` 为 `E:\Qt\Qt5.9.3\5.9.3\mingw53_32\lib\cmake\Qt5OpenGL`
 *  设置 `CMAKE_BUILD_TYPE` 为 `Release` 或者 `RelWithDebInfo`

点击 Generate 生成 Makefile

![OpenCVCMakeGenerate][]

## 编译 OpenCV ##

打开终端进行编译：（`-j` 是使用 `8` 个线程进行编译，请根据你的计算机配置合理设置线程数）

E:
    cd E:\OpenCV_3.3.1\opencv-build
    mingw32-make -j 8mingw32-make install

如果 `mingw32-make -j 8` 遇到错误，请看下面的 编译 OpenCV 常见错误，否则执行 `mingw32-make install`，完成安装。

## 编译 OpenCV 常见错误 ##

### 0. 多线程编译错误信息不明确 ###

#### 表现 ####

如果使用了多线程编译，导致错误，但是错误信息不明确，如：

Makefile:161: recipe for target 'all' failed
    mingw32-make: *** [all] Error 2

![MakeOpenCV\_cap\_dshow][MakeOpenCV_cap_dshow]

#### 解决 ####

使用单线程编译：

mingw32-make

以查看详细的错误提示，再根据具体情况解决。

### 1. RC 错误 ###

#### 表现 ####

... windres.exe: unknown option -- W ...

或者

FORMAT is one of rc, res, or coff, and is deduced from the file name
    extension if not specified.  A single file name is an input file.
    No input-file is stdin, default rc.  No output-file is stdout, default rc.

![MakeOpenCVRCError][]

#### 解决 ####

在 cmake-gui 编译配置中：

*  不勾选 `ENABLE_PRECOMPILED_HEADERS`

然后重新`Configure`\-`Generate`\-`mingw32-make`

### 2. sprintf\_instead\_use\_StringCbPrintfA\_or\_StringCchPrintfA 错误 ###

#### 表现 ####

...opencv/sources/modules/videoio/src/cap_dshow.cpp...
    ... 'sprintf_instead_use_StringCbPrintfA_or_StringCchPrintfA' was not declared in this scope ...

或者

Makefile:161: recipe for target 'all' failed
    mingw32-make: *** [all] Error 2

![MakeOpenCV\_cap\_dshow][MakeOpenCV_cap_dshow]

#### 解决 ####

修改`E:\OpenCV_3.3.1\opencv\sources\modules\videoio\src\cap_dshow.cpp`文件，在`#include "DShow.h"`这行的上面加一行`#define NO_DSHOW_STRSAFE`，如：

#define NO_DSHOW_STRSAFE
    #include "DShow.h"

然后重新`Configure`\-`Generate`\-`mingw32-make`

### 3. identifier ‘nullptr’ is a keyword in C++11 错误【2018年3月2日编译OpenCV 3.4.1时遇到并解决】 ###

#### 表现 ####

D:\opencv-3.4.1\opencv-3.4.1\3rdparty\protobuf\src\google\protobuf\stubs\io_win32.cc:94:3: warning: identifier 'nullptr' is a keyword in C++11 [-Wc++0x-compat]
       return s == nullptr || *s == 0;
       ^
    D:\opencv-3.4.1\opencv-3.4.1\3rdparty\protobuf\src\google\protobuf\stubs\io_win32.cc: In function 'bool google::protobuf::internal::win32::{ anonymous}::null_or_empty(const char_type*)':
    D:\opencv-3.4.1\opencv-3.4.1\3rdparty\protobuf\src\google\protobuf\stubs\io_win32.cc:94:15: error: 'nullptr' was not declared in this scope
       return s == nullptr || *s == 0;
                   ^
    3rdparty\protobuf\CMakeFiles\libprotobuf.dir\build.make:412: recipe for target '3rdparty/protobuf/CMakeFiles/libprotobuf.dir/src/google/protobuf/stubs/io_win32.cc.obj' failed
    mingw32-make[2]: *** [3rdparty/protobuf/CMakeFiles/libprotobuf.dir/src/google/protobuf/stubs/io_win32.cc.obj] Error 1
    CMakeFiles\Makefile2:710: recipe for target '3rdparty/protobuf/CMakeFiles/libprotobuf.dir/all' failed
    mingw32-make[1]: *** [3rdparty/protobuf/CMakeFiles/libprotobuf.dir/all] Error 2
    Makefile:161: recipe for target 'all' failed
    mingw32-make: *** [all] Error 2

![MakeOpenCV\_nullptr\_Error][MakeOpenCV_nullptr_Error]

#### 解决 ####

在 cmake-gui 编译配置中：

*  勾选 `ENABLE_CXX11`

然后重新`Configure`\-`Generate`\-`mingw32-make`

## 添加 OpenCV 编译的库到环境变量 ##

*  为系统变量 `Path` 添加 `E:\OpenCV_3.3.1\opencv-build\install\x86\mingw\bin`

## 新建 OpenCV 的 Qt 项目 ##

在 `.pro` 文件里面添加：

win32 { 
    INCLUDEPATH += E:\OpenCV_3.3.1\opencv-build\install\include
    LIBS += E:\OpenCV_3.3.1\opencv-build\install\x86\mingw\bin\libopencv_*.dll
    }

或者：（区分 debug 和 release 是因为 OpenCV 对其两者有不同的库，你需要把路径改为你自己的，我编译 MinGW 的 OpenCV 只有 release 库）

win32 { 
    INCLUDEPATH += E:\OpenCV_3.3.1\opencv-build\install\include
    CONFIG(debug, debug|release): { 
    LIBS += E:\OpenCV_3.3.1\opencv-build\install\x86\mingw\bin\libopencv_*d.dll
    } else:CONFIG(release, debug|release): { 
    LIBS += -LE:\OpenCV_3.3.1\opencv-build\install\x86\mingw\bin \     -llibopencv_core331 \     -llibopencv_highgui331 \     -llibopencv_imgcodecs331 \     -llibopencv_imgproc331 \     -llibopencv_features2d331 \     -llibopencv_calib3d331
    }
    }

配置也可以这样写

打开pro文件 添加如下：

INCLUDEPATH += G:\opencv\opencv\cmake4\install\include\

INCLUDEPATH += G:\opencv\opencv\cmake4\install\include\opencv\

INCLUDEPATH += G:\opencv\opencv\cmake4\install\include\opencv2\

LIBS += G:\opencv\opencv\cmake4\lib\libopencv_*.a

![OpenCVMinGWBin][]

然后在 MainWindow 中如下：

#include "mainwindow.h"
    #include "ui_mainwindow.h"
    
    #include <opencv2/core/core.hpp>
    #include <opencv2/highgui/highgui.hpp>
    
    MainWindow::MainWindow(QWidget *parent) :
        QMainWindow(parent),
        ui(new Ui::MainWindow)
    {
        ui->setupUi(this);
    
        // read an image
        cv::Mat image = cv::imread("E:/Pictures/H_white.png", 1);
        // create image window named "My Image"
        cv::namedWindow("My Image");
        // show the image on window
        cv::imshow("My Image", image);
    }
    
    MainWindow::~MainWindow()
    {
        delete ui;
    }

最后运行起来了，效果如图：

![OpenCVQtDemo][]

## 官方教程 ##

*  [How to setup Qt and openCV on Windows][]

[opencv _cmake he mingw]: https://blog.csdn.net/huihut/article/details/78701814
[Qt-5.9.3]: https://download.qt.io/archive/qt/
[MinGW-5.3.0-32bit]: https://sourceforge.net/projects/mingw/
[CMake-3.9.2]: https://cmake.org/download/
[OpenCV-3.3.1 _ 3.4.1]: https://opencv.org/releases.html
[Github . huihut_OpenCV-MinGW-Build]: https://github.com/huihut/OpenCV-MinGW-Build
[OpenCVCMakeGenerate]: http://ojlsgreog.bkt.clouddn.com/OpenCVCMakeGenerate.png
[MakeOpenCV_cap_dshow]: http://ojlsgreog.bkt.clouddn.com/MakeOpenCV_cap_dshow.png
[MakeOpenCVRCError]: http://ojlsgreog.bkt.clouddn.com/MakeOpenCVRCError.png
[MakeOpenCV_nullptr_Error]: http://ojlsgreog.bkt.clouddn.com/MakeOpenCV_nullptr_Error.png
[OpenCVMinGWBin]: http://ojlsgreog.bkt.clouddn.com/OpenCVMinGWBin.png
[OpenCVQtDemo]: http://ojlsgreog.bkt.clouddn.com/OpenCVQtDemo.png
[How to setup Qt and openCV on Windows]: https://wiki.qt.io/How_to_setup_Qt_and_openCV_on_Windows