c++ - 无法在 docker 容器中打开目录运行代码

Question

我是 docker 新手，我正在编写一个代码来使用带有 docker 的 opencv 检测车道线。但是，当我尝试在挂载到本地目录的容器中运行代码时，终端给了我以下错误：

root@26a27d6346fe:~/eec193_lab1/lane_line_detection/build# ./perspective_transform_tester /Users/evangeng/eec_193/eec193_lab1/docker_container/udacity_parameters.yaml /Users/evangeng/eec_193/eec193_lab1/docker_container/lane_line_detection/images/test_images/\*.jpg
terminate called after throwing an instance of 'cv::Exception'
  what():  OpenCV(3.4.4-dev) /root/opencv/modules/core/src/glob.cpp:267: error: (-204:Requested object was not found) could not open directory: /Users/evangeng/eec_193/eec193_lab1/docker_container/lane_line_detection/images/test_images in function 'glob_rec'

Aborted

我确信我的路径是正确的，图像格式也是正确的。如果这是一个明显的问题，我深表歉意。

任何意见将不胜感激！

// draws the four source or destination points on an image
void draw_viewing_window(cv::Mat& dst, std::vector<cv::Point2f>& points, cv::Scalar color, int thickness);

int main(int argc, char* argv[]) {

  if (argc != 3) {
    std::cerr << "Usage: ./perspective_transform_tester <yaml_file> <path/to/images/\\*.png> \n";
    return 1;
  }

  // create save path for final birds-eye view images
  std::string save_path = "../images/warped_images";
  if (!boost::filesystem::exists(save_path)) {
    boost::filesystem::create_directory(save_path);
  }

  // read in yaml file
  cv::FileStorage file(argv[1], cv::FileStorage::READ);
  // declare camera matrix and list of distortion coefficients
  cv::Mat dist, mtx;
  // declare source points and destination points
  std::vector<cv::Point2f> src_points, dst_points;

  file["distortion coefficients"] >> dist;
  file["camera matrix"] >> mtx;
  file["source points"] >> src_points;
  file["destination points"] >> dst_points;

  std::vector<cv::String> image_paths;
  cv::glob(argv[2], image_paths);

  cv::Mat img, undist_img;
  std::vector<cv::Mat> axs(image_paths.size());
  for (unsigned int i = 0; i < image_paths.size(); i++) {
    img = cv::imread(image_paths[i]);
    cv::undistort(img, undist_img, mtx, dist);
    draw_viewing_window(undist_img, src_points, cv::Scalar(0,255,0), 5); 
  }
  subplot("source points", axs, 4, 2);
}

编辑：以下是我的dockerfile：

FROM spmallick/opencv-docker:opencv
RUN apt-get update -y
RUN apt-get install libboost-all-dev -y
RUN apt-get install ffmpeg -y
RUN apt-get install python-numpy -y
RUN cd ~
RUN git clone https://github.com/lava/matplotlib-cpp.git
RUN cp matplotlib-cpp/matplotlibcpp.h /usr/include/
RUN rm -rf matplotlib-cpp

当我创建容器时，我使用了：

docker run -it --name=lanelines -v <path_where_I_store_my_codes>:/root/eec193_lab1 eec193_lab1

标签： c++dockeropencvvirtual-machine