I am trying to track the locations of the corners of a sheet of paper as I move it relative to an Android camera (you can assume the the sheet of paper will be a completely different color than the background). I want to find the x, y coordinates of each corner on the android screen. I also want to be able to change the angle of the paper so it won't necessarily appear perfectly rectangular all the time.
I am using opencv 2.4.1 for Android, but I could not find cvgoodfeaturetotrack
or cvfindcornersubpix
in the packages. Right now I am thinking of using the CvCanny algorithm to find the edges, then use the edges with cvfindcontours
to find the main intersections of the lines to find the corners.
Any suggestions or source code would be greatly appreciated.
I suggest you two options:
1- Use other OpenCV version where you have those functions (You can check the online documentation)
2- Use the FAST detector and SIFT descriptors. It's a widely used method for this kind of task, really up to date. It will find the best features multi-scale, robust to light conditions, etc. You have to train the marker (the sheet of paper) to extract the features with SIFT. Then use FAST detector on the camera scene to detect and track those features.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With