Originally posted by: EagleKeeper
As Dave stated:
Scanned documents are of dots.
The Optical Caharacter Recognition programs attempts to take groups of dots and translate it into letters. This is exactly the way your eyes and brain works.
Depending on the algorithms within the OCR program and the quality of the scan, you can get good text translation or lousy. Most any OCR program will require you to then clean up the output text.
Some will also allow you to train them on the fly. A character pattern will be displayed that is can not recognize and the user will be asked to identifiy what character it is.
PS:
Waggy should not use alter egos to ask questions.