2

I know theres a similar question about this but mine is the continuation of this post. ROI in Video file

I already find a way to get the ROI and is now running with my program but only in Rectangular(cv::Rect)

Now, I want to get the ROI in this manner so that the actual areas of interest will be the only one processed.

Any idea on how to get this kind of ROI?

enter image description here

Thank you very much.

Community
  • 1
  • 1
mabg
  • 73
  • 4
  • 1
    Answers to this question may help you: http://stackoverflow.com/questions/7946383/using-roi-in-opencv – Mahm00d Sep 04 '13 at 14:36
  • 1
    Why not using a binary mask? – Michele mpp Marostica Sep 05 '13 at 09:20
  • OpenCV only suports rectangular ROI's. The best solution to your problem depends on what you are going to use the image for. Depending on the algorithms you are going to use, you could proceess it in two steps, i.e. defining two ROI's and processing each separatedely, or you could go for an approach where you use a binary mask to select the pixels of interest. Maybe you could give us more details? – brunocodutra Sep 05 '13 at 18:49

0 Answers0