Chapter 10: Optical character recognition

This chapter presents the features of the Optical Character Recognition (OCR) module, which performs template-based character recognition.

• The MIL OCR module
• Steps to reading or verifying a string in an image
• Basic concepts for the MIL OCR module
• Guidelines for choosing context types
• • General OCR font context type
  • Constrained OCR font context type
  • Switching between the two
  • Deciding which OCR font context type to use
• OCR font
• • User-defined MIL OCR fonts
  • Existing MIL OCR fonts
  • SEMI fonts
  • • Using a SEMI font
    • Creating a SEMI font
  • Quality and scale are important
  • Visualizing
  • Erasing characters
• Defining the target strings
• • Calibrating your font
  • • Automatic calibration
    • Manual calibration
  • Setting appropriate processing controls
  • • Blanks
    • Broken characters
    • Morphological filtering
    • Thickness and dots
    • Touching characters
  • Specifying other string information
  • • Number of strings
    • String lengths
    • Angle
    • Positional variation
  • Constraints
• Locating your text
• Determining what is a match
• • Acceptance levels
  • Unrecognized characters
• Retrieving and analyzing the results
• • A character
  • A string
  • A text
  • Understanding odd results
  • Hooking functions
• Improving search speed
• Optical character recognition example