2020-06-04, 08:38 AM
Might it be possible that OpenCV might play a gateway role here? Attempting to integrate things at a native protocol level is obviously hard. In a case like this where output is already being displayed visually, wouldn't it be possible (and much easier in the long run) to train a camera (a cheap CCD would suffice) on the instrument and perform the conversion is OpenCV? https://github.com/opencv/opencv