[ksaj]

It would be interesting to train OpenCV on the different UPC / QR / etc. code formats, and have it capture whatever codes it sees (even multiple formats at the same time, which is pretty common in shipping) and send the images to a reader to be processed.

This would probably be a good Raspberry Pi project for someone thus inclined.

Edit: I found this project that says it can, so it's at least a 2-year old still-unanswered question of who will program it: https://www.pyimagesearch.com/2018/05/21/an-opencv-barcode-a...