Five things your optical character recognition tool must do
John Stamos, 06-04-2013
The pharmaceutical industry has been compelled by government regulations to adhere to standardized solutions across all production and packaging operations. So as you can imagine, that industry has lots of experience with optical character recognition and optical character verification (OCR/OCV) applications. But most of you are in industries that don’t use standardized fonts, software, or hardware. This lack of standardization can create problems for OCR/OCV tools.
You can save yourself a lot of trouble later by asking these five questions before you set up an OCR tool to solve applications on your production line:
- Can the tool learn and read any printed font?
- Can the tool read text with little contrast between type and background?
- Can the tool read text with significant variations in width and height?
- Can the tool read text when letters are touching, skewed, and distorted?
- Can the tool differentiate between similar shapes, such as the letter “O” and the number ”0?”
Here’s a bonus tip: if you can’t set up your application in about 5 minutes, it may be too complex and could cause headaches for you down the line. OCR shouldn’t be that hard for good machine vision software – make sure your vision system can do the work for you.