Question

I was following the example given in the OpenCV for video displaying, just took out some transformations unnecessary for me. The code I have right now loads the video file and then displays it, the problem is that reproduced video have wrong colours.

Here’s the code:

using namespace std;
using namespace cv;

// The main function

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

VideoCapture cap ("ETMI 002.mpg");  // Open the file

if (!cap.isOpened ())               // Check if opening was successful
    cerr << "I have failed!" << endl;

else
{
    Mat edges;
    Mat frame;
    namedWindow("edges",1);
    while (cap.read (frame))
    {
      cvtColor(frame, edges, CV_BGR2RGB);  
      imshow("edges", edges);
      if(waitKey(30) >= 0) break;
    }
}

return 0;

}

Answer 1

to avoid transformation in reproduced video change
imshow(“edges”, edges) to
imshow(“edges”, frame)

while (cap.read (frame))
{
  cvtColor(frame, edges, CV_BGR2RGB); 

  // here is the change you are showing the converted image 
  // just simply add the original read frame

  imshow("edges", frame);   // here 
  if(waitKey(30) >= 0) break;
}