简体   繁体   English

在Ubuntu 14.04中安装OpenCV

[英]Installation of OpenCV in Ubuntu 14.04

I built OpenCV from Source but I can't manage to use it. 我从Source构建了OpenCV,但我无法使用它。 Every time I try to even load an image with the next code I get No module named cv2.cv . 每当我尝试使用下一个代码加载图像时,我得到的是No module named cv2.cv Why is that happening? 为什么会这样? How can I fix it? 我该如何解决?

from cv2.cv import *

img = LoadImage("/home/User/Desktop/Image.png")
NamedWindow("opencv")
ShowImage("opencv",img)
WaitKey(0)

The procedure I did was the following... 我做的程序如下......

I downloaded the zip file from the main page of GitHub and while being in a destination directory I created, I built OpenCV using 我从GitHub的主页面下载了zip文件,在我创建的目标目录中,我使用了OpenCV

cmake OpenCV_Source_Directory

Then on the destination directory I run 然后在我运行的目标目录上

make

sudo make install

I found the solution to my problem. 我找到了解决问题的方法。 I had to install python-opencv as follows: 我必须安装python-opencv如下:

sudo apt-get install python-opencv

After that OpenCV works fine. 之后OpenCV工作正常。

You've likely installed opencv 3, which doesn't have the cv2.cv module. 你可能已经安装了opencv 3,它没有cv2.cv模块。 It's all in cv2 now. 它现在都在cv2

To verify run this in a python interpreter 验证在python解释器中运行它

import cv2
print cv2.__version__ 

Anything like 3.0.0 or 3.1.0 means that the cv2.cv module doesn't exist. 3.0.03.1.0这样的cv2.cv意味着cv2.cv模块不存在。

Presumable you did: 推测你做了:

git clone git@github.com:opencv/opencv.git
mkdir build; cd build
cmake ../opencv && make && sudo make install

But if you do this: 但是如果你这样做:

cd opencv
git describe

You will get something like 你会得到类似的东西

3.1.0-1374-g7f14a27

That is, the default branch is OpenCV 3, which doesn't have the cv2.cv module. 也就是说,默认分支是OpenCV 3,它没有cv2.cv模块。 Either change your code to work with the OpenCV 3 cv2 module. 更改代码以使用OpenCV 3 cv2模块。 Or downgrade to OpenCV2. 或者降级到OpenCV2。 You can do this by: 你可以这样做:

cd opencv
git checkout 2.4.13.1
cd ../build && so on ...

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM