Hi! I'm very new to arduino and electronics in general so my knowledge is very limited and i would like some help. I was watching a chess tournament on twitch and got inspired to start a project. There were several matches going on at the same time and on the right side of the screen there was an image of a chess board and the current placement of the chess pieces on the board. When the camera changed to a different match so did the chess pieces on the right side of the screen. I was wondering what kind of sensor / tecnique was being used to accomplish this?
You can put a unique code on the bottom of each chess piece. It can be optical or RFID based.
More likely, the system is using the rules of chess, where it knows the starting position of every piece and only one piece moved at a time. Any time a piece arrives at a new place, a previously-identified piece which is now missing must be that piece.
Or it could be a person watching the monitor punching it in to a computer.
It could be a bar code or RFID built into the chess pieces and board.
I think it's more likely to be an overhead camera. It would be fairly easy (for someone skilled in image processing) to detect what has moved between two images.
They could also have a team of people in a low cost country watching the games and updating the positions.
One possible approach is to use an IR-transparent board and symbols on the bottom of
each piece - an IR camera into a bit of CV code to match the symbols would identify what is where.
Has the advantage of a single sensor (camera) per board.
I know some types of black plastic are transparent at near-IR wavelengths (discovered with a
PiNoir camera!), but white tends to be opaque at many wavelengths by scattering, so the IR-transparaent
board technology might be challenging.
More advanced CV could directly recognize pieces, but that would require machine learning and
training and is much harder challenge.
Reply #6 has the best advice you will ever get. Computer chess boards have been around for decades and thus their computer had less power than the humblest Arduino. I have no idea how they work, but it might be no more than a matrix of Hall effect switches and a magnet below each piece.