Raspbian搭建python + opencv环境

环境：raspberry 3b + raspbian jessie
结果：环境搭建完以后，执行如下命令，就会进入带有opencv的python虚拟环境。

source ~/.profileworkon cv

参考：http://www.tuicool.com/articles/NZF3q2q
http://www.pyimagesearch.com/2015/10/26/how-to-install-opencv-3-on-raspbian-jessie/?utm_source=tuicool&utm_medium=referral

1. 安装依赖项

一些开发工具

sudo apt-get install build-essential git cmake pkg-config

安装一些图片的I/O包

sudo apt-get install libjpeg-dev libtiff5-dev libjasper-dev libpng12-dev

安装一些vedio的I/O包

sudo apt-get install libavcodec-dev libavformat-dev libswscale-dev libv4l-devsudo apt-get install libxvidcore-dev libx264-dev

安装一些图形工具库GTK

sudo apt-get install libgtk2.0-dev

安装一些opencv的内部库，例如矩阵操作

sudo apt-get install libatlas-base-dev gfortran

安装python2.7 和 python 3的一些开发工具

sudo apt-get install python2.7-dev python3-dev

2. 获取opencv源码

获取opencv源码

cd ~/wget -O opencv.zip https://github.com/Itseez/opencv/archive/3.0.0.zipunzip opencv.zip

获取opencv_contrib源码，务必保证与opencv的版本一致（opencv3.0.0 配 opencv_contrib3.0.0)

wget -O opencv_contrib.zip https://github.com/Itseez/opencv_contrib/archive/3.0.0.zipunzip opencv_contrib.zip

3. 创造 python 虚拟环境

3.1安装和配置virtualenv 和 virtualenvwrapper

安装pip

wget https://bootstrap.pypa.io/get-pip.pysudo python get-pip.py

安装virtualenv 和 virtualenvwrapper

sudo pip install virtualenv virtualenvwrappersudo rm -rf ~/.cache/pip

设置virtualenv的启动， 修改～/.profile文件：

sudo vim ~/.profile

在最后加上这两行

export WORKON_HOME=$HOME/.virtualenvssource /usr/local/bin/virtualenvwrapper.sh

重新加载~/.profile

source ~/.profile

3.2创建env

创建env

mkvirtualenv cv  # python2.7的环境或mkvirtualenv cv -p python3  # python3的环境

进入·env环境：

source ~/.profileworkon cv

进入以后，命令行前面会多个(cv)

安装numpy

pip install numpy

4. 编译和安装opencv

确保在虚拟cv环境中，如果不在请 workon cv

编译opencv, 先build

cd ~/opencv-3.0.0/mkdir buildcd buildcmake -D CMAKE_BUILD_TYPE=RELEASE \ -D CMAKE_INSTALL_PREFIX=/usr/local \-D INSTALL_C_EXAMPLES=ON \-D INSTALL_PYTHON_EXAMPLES=ON \-D OPENCV_EXTRA_MODULES_PATH=~/opencv_contrib-3.0.0/modules \-D BUILD_EXAMPLES=ON ..

开始编译：

make -j4   # 树莓派2以上有4个核

开始安装

sudo make installsudo ldconfig

5. 为python配置opencv

至此opencv基本安装好，并且opencv也给系统的python添加了软连接，用如下命令查看

# for python2.7ls -l /usr/local/lib/python2.7/site-packages/total 1640-rw-r--r-- 1 root staff 1677024 Oct 15 19:11 cv2.so# for python3.4ls -l /usr/local/lib/python3.4/site-packages/total 1636-rw-r--r-- 1 root staff 1674656 Oct 15 19:11 cv2.cpython-34m.so

但是我们要为virtualenv中的python引入软连接，需要手动进行：

# for python2.7cd ~/.virtualenvs/cv/lib/python2.7/site-packages/ln -s /usr/local/lib/python2.7/site-packages/cv2.so cv2.so# for python3.4cd ~/.virtualenvs/cv/lib/python3.4/site-packages/ln-s/usr/local/lib/python3.4/site-packages/cv2.socv2.so

6.验证是否安装成功

在shell中

$ workon cv$ python>>> import cv2>>> cv2.__version__'3.0.0'

这就安装成功了！

最后把安装包删了

cd ~/rm opencv_contrib.zip opencv.zip get-pip.pyrm -r opencv_contrib-3.0.0/ opencv-3.0.0/