opencv 4.2.0 编译

一、软件下载

下载opencv 4.2.0 source，cmake，opencv-contrib 4.2.0，VS2019.

二、CMake编译

1. 解压opencv source命名opencv-4.2.0，解压opencv_contrib 4.2.0更名为opencv_contrib，并放入opencv-4.2.0文件夹中，并新建文件夹build。

2. 打开cmake软件按图填写

   
   
   image.png
3. 点击configure 选择VS版本 点击继续
4. configure成功后，search BUILD_opencv_world打钩，OPENCV_EXTRA_MODULES_PATH=opencv-4.2.0\opencv_contrib\modules，再次configure
5. 点击generate，在build目录下生成opencv.sln

三、VS2019编译

1. 用VS2019打开opencv.sln，点击菜单栏生成，批生成

   
   
   image.png
2. 编译成功后，系统path变量添加E:\opencv-4.2.0\build\install\x64\vc16\bin
   新建VS项目配置属性表，
   C++附加包含目录包括：build\install\include，build\install\include\opencv2
   VC++包含目录： build\install\include，build\install\include\opencv2
   VC++库目录：build\install\x64\vc16\lib
   链接器输入：opencv_world420d.lib

四、测试代码

#include 
#include 
#include 
#include 

using namespace cv;
using namespace std;

int main()
{
    Mat image = Mat::zeros(300, 600, CV_8UC3);
    circle(image, Point(250, 150), 100, Scalar(0, 255, 128), -100);
    circle(image, Point(350, 150), 100, Scalar(255, 255, 255), -100);
    imshow("Display Window", image);
    waitKey(0);
    return 0;
}

image.png

四、error及解决措施

1. CMake编译报错

CMake Warning at cmake/OpenCVGenSetupVars.cmake:54 (message):
  CONFIGURATION IS NOT SUPPORTED: validate setupvars script in install
  directory
Call Stack (most recent call first):
  CMakeLists.txt:947 (include)

解决方法： 无视

2. CMake编译报错 ： 找不到python2

   
   
   image.png
   
   

   解决方法：Anaconda 新建一个python27环境，CMake界面search中搜索PYTHON_LIBRARY，选择anaconda新建python27虚拟环境中，选择Libs/python27.lib
   同样，把python3的也改掉。

3. CMake编译报错： ippicv_2019_win_intel64_20180723_general.zip下载失败

   
   
   image.png

4.CMake编译报错： opencv_videoio_ffmpeg.dll下载失败

image.png

5.CMake编译报错： opencv_videoio_ffmpeg_64.dll下载失败

image.png

6. CMake编译报错： ffmpeg_version.cmake下载失败

   
   
   image.png

解决方法：

以上问题由科学上网造成的，先查看opencv4.2.0文件夹下的.cache，如ffmpeg文件夹内

image.png

三个文件都是0kb，文件名称是哈希码+文件名
查看build目录下CMakeDownloadLog.txt

#match_hash_in_cmake_cache "OCV_DOWNLOAD_ADE_HASH_3rdparty_ade_v0_1_1f_zip"
#do_copy "opencv_videoio_ffmpeg.dll" "5de6044cad9398549e57bc46fc13908d" "https://raw.githubusercontent.com/opencv/opencv_3rdparty/a66a24e9f410ae05da4baeeb8b451912664ce49c/ffmpeg/opencv_videoio_ffmpeg.dll" "E:/opencv-4.2.0/build/3rdparty/ffmpeg"
#missing "E:/opencv-4.2.0/build/3rdparty/ffmpeg/opencv_videoio_ffmpeg.dll"
#check_md5 "E:/opencv-4.2.0/.cache/ffmpeg/5de6044cad9398549e57bc46fc13908d-opencv_videoio_ffmpeg.dll"
#mismatch_md5 "E:/opencv-4.2.0/.cache/ffmpeg/5de6044cad9398549e57bc46fc13908d-opencv_videoio_ffmpeg.dll" "d41d8cd98f00b204e9800998ecf8427e"
#delete "E:/opencv-4.2.0/.cache/ffmpeg/5de6044cad9398549e57bc46fc13908d-opencv_videoio_ffmpeg.dll"
#cmake_download "E:/opencv-4.2.0/.cache/ffmpeg/5de6044cad9398549e57bc46fc13908d-opencv_videoio_ffmpeg.dll" "https://raw.githubusercontent.com/opencv/opencv_3rdparty/a66a24e9f410ae05da4baeeb8b451912664ce49c/ffmpeg/opencv_videoio_ffmpeg.dll"
#try 1
# timeout on name lookup is not supported
#   Trying 199.232.4.133:443...
# Connected to raw.githubusercontent.com (199.232.4.133) port 443 (#0)
# schannel: ALPN, offering h2
# schannel: ALPN, offering http/1.1
# schannel: failed to receive handshake, SSL/TLS connection failed
# Closing connection 0
# schannel: shutting down SSL/TLS connection with raw.githubusercontent.com port 443
# Send failure: Connection was reset
# schannel: failed to send close msg: Failed sending data to the peer (bytes written: -1)

手动下载opencv_videoio_ffmpeg.dll，复制到.cache对应的文件夹下并更改名称（哈希码+文件名字）
下载地址：https://raw.githubusercontent.com/opencv/opencv_3rdparty/a66a24e9f410ae05da4baeeb8b451912664ce49c/ffmpeg/opencv_videoio_ffmpeg.dll
问题1-6都可由此方法解决

注意！！

ffmpeg_version.cmake链接由网页打开后不是直接下载，而是直接打开了这个文本，不要复制新建txt。要另存为然后该名称。

7. CMake编译报错：在安装目录中验证setupvars脚本

   
   
   image.png
   
   

   解决方法：无视掉，貌似不影响，未验证

8. VS编译错误 找不到python38.lib

解决方法：直接不编译python接口

image.png