SSD安装

本人环境ubuntu16.04(仅cpu)

下载SSD项目

1
2
3

git clone https://github.com/weiliu89/caffe.git
cd caffe
git checkout ssd

安装caffe相关依赖包

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
sudo apt-get install libatlas-base-dev
sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev
sudo apt-get install liblapack-dev liblapack3 libopenblas-base libopenblas-dev

配置

1	cp Makefile.config.example Makefile.config

修改Makefile.config：

1	# CPU_ONLY:=1`改为`CPU_ONLY:=1

在INCLUDE_DIRS和LIBRARY_DIRS后面分别加上/usr/include/hdf5/serial和/usr/lib/x86_64-linux-gnu/hdf5/serial

1 2	INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include /usr/include/hdf5/serial LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib /usr/lib/x86_64-linux-gnu/hdf5/serial

进入python文件夹安装需要的包

1	for req in $(cat requirements.txt); do sudo -H pip install $req; done

将该python文件夹的路径添加到PYTHONPATH中

1	export PYTHONPATH=/path/to/caffe/python:$PYTHONPATH

在Makefile文件的195行：

1 2	ifeq($(USE_OPENCV),1) LIBRARIES+=

在LIBRARIES后面添加opencv_videoio和opencv_imgcodecs

编译中途出错，建议make clean，再重新编译

编译

make -j8
make py
make test -j8
make runtest -j8

运行demo

下载训练好的模型文件放到models文件夹内，运行

1	python examples/ssd/ssd_detect.py

效果如下

终端上打印的为：[xmin,ymin,xmax,ymax,label,score,label_name]

xmin,ymin,xmax,ymax为矩形框的范围，表示方式是图像宽高的百分比