简易方法
下载sh脚本文件Easy Install by@zeakey
只安装opencv：bash install-opencv.sh
安装caffe(cpu_only)+opencv：bash install-caffe.sh
如果安装GPU版本，则修改如下
#!/bin/bash
# install common deps
set -e
sudo apt-get update
sudo apt-get upgrade
sudo apt-get install -y build-essential cmake make
sudo apt-get install --no-install-recommends libboost-all-dev
sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libhdf5-serial-dev protobuf-compiler
sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev
sudo apt-get install libatlas-base-dev
sudo apt-get install python-dev
# install OpenCV
if [[ "$(which opencv_version)" == '' ]]; then
echo "Installing OpenCV ..."
sudo bash install-opencv.sh
fi
# install python deps
if [[ "$(which pip)" == '' ]]; then
# curl https://bootstrap.pypa.io/get-pip.py
# alternative link for CN users
curl http://zhaok-data.oss-cn-shanghai.aliyuncs.com/service/get-pip.py | sudo -H python
else
sudo -H pip install --upgrade pip
fi
sudo -H pip install numpy scipy Cython
if [ ! -d /usr/lib/python2.7/dist-packages/numpy ]; then
sudo ln -s /usr/local/lib/python2.7/dist-packages/numpy /usr/lib/python2.7/dist-packages/numpy
fi
echo '#numpy include path' >> ~/.bashrc
echo 'export CPATH=/usr/local/lib/python2.7/dist-packages/numpy/core/include:$CPP_INCLUDE_PATH' >> ~/.bashrc
echo 'export CPATH=/usr/local/lib/python2.7/dist-packages/numpy/core/include:$C_INCLUDE_PATH' >> ~/.bashrc
. ~/.bashrc
sudo apt-get install python-tk # tkinter cannot install with pip
sudo -H pip install scikit-image scikit-learn
sudo -H pip install matplotlib
sudo -H pip install ipython jupyter
sudo -H pip install protobuf
sudo -H pip install h5py leveldb lmdb
sudo -H pip install networkx nose
sudo -H pip install pandas
sudo -H pip install python-dateutil
sudo -H pip install python-gflags pyyaml Pillow six pyzmq singledispatch
sudo -H pip install backports_abc certifi jsonschema graphviz qtawesome pydot
# clone latest caffe source code
cd ~/ && git clone https://github.com/BVLC/caffe --depth 1 && cd caffe
mkdir build
cd build
cmake ..
make all
make install
make runtest
## some other cmake examples ##
# build with cuda, cudnn and nccl
# cmake -USE_NCCL=ON -DNCCL_INCLUDE_DIR=/pkg/nccl/include -DNCCL_LIBRARIES=/pkg/nccl/lib \
# -DCUDNN_INCLUDE=/usr/local/cudnn/include -DCUDNN_LIBRARY=/usr/local/cudnn/lib64 ..
echo "Done! Built caffe source at '~/caffe'"
# try caffe MNIST example:
# cd ~/caffe
# bash data/mnist/get_mnist.sh
# bash examples/mnist/create_mnist.sh
# bash examples/mnist/train_lenet.sh
次选方法
官网指引
注意：装Caffe前先安装OpenCV
OpenCV安装
下载opencv源码,opencv.zip和opencv_contirb.zip下载解压，放在根目录下
opencv官网
第一步：安装源码前先安装好需要的第三方环境
[compiler] sudo apt-get install build-essential
[required] sudo apt-get install cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev
[optional] sudo apt-get install python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev
第二步：使用cmake编译源码（包括opencv_contrib部分）
~$ cd opencv
~/opencv$ mkdir build
~/opencv$ cd build
~/opencv/build$ cmake -D CMAKE_BUILD_TYPE=Release -D OPENCV_EXTRA_MODULES_PATH=~/opencv_contirb/modules/ -D CMAKE_INSTALL_PREFIX=/usr/local ..
常见问题
解决方法：把需要的文件从百度网盘源码资料中替换到对应的目录（<opencv目录>/3rdparty/ippicv/downloads/linux-808b791a6eac9ed78d32a7666804320e/ippicv_linux_20151201.tgz）重新编译
第三步：编译安装
make -j7
sudo make install
ls /usr/local/lib
sudo vim /etc/ld.so.conf
include /etc/ld.so.conf.d/*.conf
/usr/local/lib
:wq!退出
sudo ldconfig
vim test_opencv.cpp
:q!退出
g++ -o test_opencv test_opencv.cpp -lopencv_core -lopencv_highgui -lopencv_imgcodecs
./test_opencv
显示图片
Caffe选择Github上的主分支，下载解压到某个位置
General dependencies
sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler
sudo apt-get install --no-install-recommends libboost-all-dev
BLAS
sudo apt-get install libatlas-base-dev
Remaining dependencies, 14.04
sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev
CMake Build
在caffe的根目录下
mkdir build
cd build
cmake ..
make all
make install
make runtest
如果是CPU模式，cmake ..加上
cmake -DCMAKE_BUILD_TYPE=Release -DCPU_ONLY=ON ..
-D表示参数定义
安装成功
