How do I do this? First off, each page is scanned at least once in B&W. If there is a photo, this page is scanned in gray scale with corrections for printing dots. The resulting pages are then read into OmniPage 11 for OCR. The results is then saved as a Word dcoument. I could save directly as a PDF file but the size would be substandually larger and the photos would not be be a good. The resulting Word document is then converted to a PDF file. An example of the size, the last section in this document would have been 5.7Meg directly from the OCR process vs 2.1Meg from Word. Note, this process takes approximately 10 minutes per page.
From collection of Dave Pickrel.