请帮我看看我的代码是哪里出了问题？是与RealStk SDK 2和OpenCV有关的问题

你好
我不确定问题是来自英特尔Realsense库还是来自OpenCV：
我有一个简单的程序，只是从英特尔Realsense D435获取RGB图像，使用OpenCV函数cvtColor将图像转换为BGR，并使用OpenCVs imshow显示它。
这一直有效，直到我同时激活英特尔Realsense D435的深度相机和RGB相机。
然后我得到一些帧似乎是OpenCV没有完成RGB图像的转换。
所以红色显示为蓝色，反之亦然。
这种情况不规则地发生。
我基于github上的OpenCV imshow示例编写代码。
我使用的是最新的OpenCV版本3.4.3和Intel Realsense SDK 2.0版本2.16.0。
我还在answers.opencv.org上发布了这个问题，因为如果这是英特尔Realsense或OpenCV的问题，我不会感到害羞。
这是我的代码：
#include librealsense2 / rs.hpp> //包括RealSense跨平台API #include opencv2 / opencv.hpp> //包含OpenCV APIint main（int argc，char * argv []）try {//声明RealSense管道，封装实际
设备和传感器rs2 ::管道管道;
rs2 :: config cfg;
//深度流配置cfg.enable_stream（RS2_STREAM_DEPTH，1280,720，RS2_FORMAT_Z16,30）;
// color stream config cfg.enable_stream（RS2_STREAM_COLOR，1280,720，RS2_FORMAT_RGB8,30）;
//开始流式传输pipe.start（cfg）;
const auto window_name =“显示图像”;
cv :: namedWindow（window_name，cv :: WINDOW_AUTOSIZE）;
while（cv :: waitKey（1）0＆amp;＆amp; cvGetWindowHandle（window_name））{rs2 :: frameset data = pipe.wait_for_frames（）;
//等待相机的下一组帧rs2 :: frame color = data.get_color_frame（）;
//查询帧大小（宽度和高度）const int w = colour.asrs2 :: video_frame>（）。get_width（）;
const int h = colour.asrs2 :: video_frame>（）。get_height（）;
//从着色深度数据cv :: Mat image（cv :: Size（w，h），CV_8UC3，（void *）colour.get_data（），cv :: Mat创建大小为（w，h）的OpenCV矩阵：
：AUTO_STEP）;
//从RGB转换为BGR cv :: cvtColor（image，image，cv :: COLOR_RGB2BGR）;
//使用新数据imshow更新窗口（window_name，image）;
} return EXIT_SUCCESS;} catch（const rs2 :: error＆amp; e）{std :: cerr“RealSense错误调用”e.get_failed_function（）“（”e.get_failed_args（）“）： n”e.what（）
的std :: ENDL;
return EXIT_FAILURE;} catch（const std :: exception＆amp; e）{std :: cerr e.what（）std :: endl;
返回EXIT_FAILURE;}



以下为原文

Hello
I'm not sure if the problem comes from the Intel Realsense library or from OpenCV:
I have a simple program that just gets a RGB image from the Intel Realsense D435, converts the image to BGR with the OpenCV function cvtColor and shows it with OpenCVs imshow.

This works until I activate the depth camera and the RGB camera of the Intel Realsense D435 at the same time. Then I get some frames where it seems like the conversion of the RGB image was not done by OpenCV. So red appears blue and vice versa. This happens in irregular intervals.

I based the code on the OpenCV imshow example from github.
I'm using the newest OpenCV version 3.4.3 and Intel Realsense SDK 2.0 version 2.16.0.

I also posted this question on answers.opencv.org because I'm not shure if this is a problem with Intel Realsense or OpenCV.

Here is my code:

#include // Include RealSense Cross Platform API
#include    // Include OpenCV API

int main(int argc, char * argv[]) try
{
   // Declare RealSense pipeline, encapsulating the actual device and sensors
  rs2::pipeline pipe;

  rs2::config cfg;

   //depth stream config
  cfg.enable_stream(RS2_STREAM_DEPTH, 1280, 720, RS2_FORMAT_Z16, 30);

   //colour stream config
  cfg.enable_stream(RS2_STREAM_COLOR, 1280, 720, RS2_FORMAT_RGB8, 30);

   // Start streaming
  pipe.start(cfg);

   const auto window_name = "Display Image";
  cv::namedWindow(window_name, cv::WINDOW_AUTOSIZE);

   while (cv::waitKey(1) < 0 && cvGetWindowHandle(window_name))
   {
  rs2::frameset data = pipe.wait_for_frames(); // Wait for next set of frames from the camera
  rs2::frame colour = data.get_color_frame();

   // Query frame size (width and height)
   const int w = colour.as().get_width();
   const int h = colour.as().get_height();

   // Create OpenCV matrix of size (w,h) from the colorized depth data
  cv::Mat image(cv::Size(w, h), CV_8UC3, (void*)colour.get_data(), cv::Mat::AUTO_STEP);
   //convert from RGB to BGR
  cv::cvtColor(image, image, cv::COLOR_RGB2BGR);

   // Update the window with new data
  imshow(window_name, image);
   }

   return EXIT_SUCCESS;
}
catch (const rs2::error & e)
{
  std::cerr << "RealSense error calling " << e.get_failed_function() << "(" << e.get_failed_args() << "):n " << e.what() << std::endl;
   return EXIT_FAILURE;
}
catch (const std::exception& e)
{
  std::cerr << e.what() << std::endl;
   return EXIT_FAILURE;
}

0