I captured images of a charuco board on the openMV H7 cam and ran an openCV program on my computer to get the intrinsic calibration camera matrix and distortion coefficients. Is there any place for me to input the intrinsic calibration camera matrix and use it to undistort the camera image to increase localization accuracy (similar to cv.undistort())?
Unfortunately we do not. We only have the lens corr method which lets you manually type in values.
I have a goal of implementing a function like the OpenCV one. But, given my day job work load demands and the current backlog I don’t know when I will be able to write it.
How did you set things up to capture the image? I’m trying to do a checkerboard calibration and all I can see is a black smudge. I’m using the super telephoto lens with the global shutter module. Is my image likely too small?
too many reasons can cause this…
try pure imaging techniques at first like focus, lighting, working distance, dirty lenses e.t.c…