Hello everyone,
We are working on an automatic measurement system for livestock animals using Oak-D Pro cameras. For our purposes we require an accuracy of less than 1cm at 1m distance. Unfortunately, we have observed that at these distances the measured length varies between different cameras (we have multiple different Oak-D Pro in use) up to 1%. This error can be observed in length measurements parallel to the camera view, as well as in depth measurements themselves.
I would assume that this is because of differences in camera calibration? Do you have any references regarding the amount of error that should be anticipated between cameras or any ideas to reduce it by further calibration?
Many thanks.
Daniel
