But having worked with digital scanning technology and OCR, has anyone investigated if these layers in the PDF are actually created by the scanning process itself? OCR scanning is now above 99% accurate, but it is not perfect. This means that all characters in an scanned physical piece of paper are not always converted to text. This would explain the "R" in BA_ACK that is on a different layer than the text layer.
(I am referring to the link NoGrayZone used in post 6 of this thread.)
Maybe someone should scan a physical BC themselves and see what if an OCR-capable scanner creates these layers as well.
If I had a scanner, I would scan mine. Perhaps when I go to work on Tuesday, I’ll scan it there and see what happens.