Ubuntu 下python3 安装opencv3.1.0及问题解决

方法/步骤

1. 1

   由于ubuntu14.04/16.04都自带了python2.7和python3.5，所以无需在安装python。

   检查python是否安装：

   ~ whereis python

2. 2

   首先需要的是 pip：

   ~ sudo apt-get install python3-setuptools python3-dev sudo easy_install3 pip

   然后是 numpy：

   ~ pip3 install numpy

   安装依赖：

   ~ sudo apt-get install build-essential sudo apt-get install cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev

3. 3

   安装opencv3:

   ~ wget https://codeload.github.com/opencv/opencv/zip/3.2.0

   ~ unzip 3.2.0 && cd 3.2.0

   cmake -D CMAKE_BUILD_TYPE=Release -D CMAKE_INSTALL_PREFIX=/usr/local  PYTHON3_EXECUTABLE=/usr/bin/python3 PYTHON_INCLUDE_DIR=/usr/include/python3.4 PYTHON_LIBRARY=/usr/lib/x86_64-linux-gnu/libpython3.4m.so PYTHON3_NUMPY_INCLUDE_DIRS=/usr/local/lib/python3.4/dist-packages/numpy/core/include ..

   ~ make -j4

   ~ sudo make install

4. 4

   测试一下：

   python3.5

   >>> import cv2

   >>> cv2.__version__

   '3.2.0'

   >>>

   恭喜您，现在可以在python3.5中调用opencv3了！

   错误：在编译安装opencv3.1时可能会报错

   modules/cudalegacy/CMakeFiles/opencv_cudalegacy.dir/build.make:290: recipe for target ‘modules/cudalegacy/CMakeFiles/opencv_cudalegacy.dir/src/graphcuts.cpp.o’ failed 
   make[2]: *** [modules/cudalegacy/CMakeFiles/opencv_cudalegacy.dir/src/graphcuts.cpp.o] Error

   这是CUDA配置出错。

   解决办法1：不用CUDA在cmake时设置-DWITH_CUDA=OFF

   解决办法2:GraphCut在cuda7.5中弃用，在cuda8.0中被移除。

   修改 ～/opencv-3.1.0/modules/cudalegacy/src/graphcuts.cpp文件 
   把第45行修改成：

   #if !defined (HAVE_CUDA) || defined (CUDA_DISABLER) || (CUDART_VERSION >= 8000)

   即在 \#if !defined (HAVE_CUDA) || defined (CUDA_DISABLER) 加上 || (CUDART_VERSION >= 8000)