Linux使用CMake构建OpenCV、onnxruntime模型部署项目

Linux使用CMake构建模型部署项目

---

一、安装opencv和下载onnxruntime

• 安装opencv可参考Linux环境下OpenCV的安装与配置
• 请到onnxruntime release下载onnxruntime
  
• 下载好onnxruntime后上传到linux上，输入以下命令

tar -xzvf onnxruntime-linux-x64-gpu-1.11.1.tgz

解压后的内容为：

二、构建Cmake项目

构建项目结构

为了使整个项目更加条理，我们的文件夹采用如下组织方式：

onnx
—bin
—build
—CMakeLists.txt
—src

构建CMakeLists.txt

# project name
PROJECT(onnx)
# requirement of cmake version
cmake_minimum_required(VERSION 3.5)

# set the directory of executable files
set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${onnx_SOURCE_DIR}/bin)

# find required opencv
find_package(OpenCV REQUIRED)
# directory of opencv headers
include_directories(${OpenCV_INCLUDE_DIRS})

#onnxruntime
set(ONNXRUNTIME_ROOT_PATH /home/linwis/Code/onnxruntime-linux-x64-gpu-1.11.1/)
set(ONNXRUNTIME_INCLUDE_DIRS ${ONNXRUNTIME_ROOT_PATH}/include/)
set(ONNXRUNTIME_LIB ${ONNXRUNTIME_ROOT_PATH}lib/libonnxruntime.so)

# name of executable file and path of source file
add_executable(onnx src/imageSeg.cpp)
# directory of opencv library
link_directories(${OpenCV_LIBRARY_DIRS})
# opencv libraries
target_link_libraries(onnx ${OpenCV_LIBS})

include_directories(${ONNXRUNTIME_INCLUDE_DIRS})
target_link_libraries(onnx ${ONNXRUNTIME_LIB})

记得改自己的onnxruntime的root path

Cmake编译

src下面放你写好的模型部署推理代码，执行以下命令：

cd build
cmake ..
make
cd ../bin
./onnx

参考资料：
https://blog.csdn.net/github_30605157/article/details/79839177