116 Chapter 5 Text
Scanning and OCR
Developers often need to incorporate text that exists only in its traditional format, the
printed page. Scanning and optical character recognition (OCR) produce a computer text
file from printed sources.
Scanning text for use in multimedia applications is a three-stage process. First, the
text is scanned using one of several different types of devices (hand scanner, flatbed
scanner, etc.). In this process light and photoreceptors are used to produce, in effect, a
picture of the page. Like the graphics text described above, the scan only records the
shapes that make up the text and not the identity of the individual characters.
In the second stage, specialized software is used to identify the characters ...