OCR software functionality in four steps

January 2nd 2017

OCR software is used to ‘read’ text which is captured on a digital image.
OCR stands for Optical Character Recognition and that is exactly what it does: Recognizing Characters by their optical appearance.
The OCR process requires four main steps:

Step one is to capture a digital image with a camera with text that is visible. The better the contrast between text and surrounding, the better the OCR software will be able to distinguish the characters.

Step two is to identify the text in the picture, as said, the contrast helps to do this.

Step three is the interpretation of the individual characters. In order to do this, the software algorithm has been given intelligence.
But intelligence alone is not enough, it is crucial that the camera resolution has the right ratio in comparison to the filed of view which it covers. If the resolution is relatively low, the image will not have enough pixels assigned to a character to make a reliable decision.

ocr software
It requires few pixels to recognize an ‘i’ from a ‘a’, but in order to be sure that a ‘B’ is not interpreted as an ‘8’, we assign at least 20 pixels to the minimal character height of the text we wish to read.

In these three steps the software can convert a picture into information and export the digital characters to a database.

Step four is optionally usable for redundance: syntax.
When syntax-rules are known in advance, the software can compare the read characters to the syntax-rules.
If for example the syntax-rule is the presence of 8 characters of wich the first two are numerical, it will:
– report an error if not 8 characters are found
– match the first two to 0-9 (and never to A-Z)
– match the last 6 to A-Z (and never to 0-9)


Industrial applications of OCR software

Practical industrial applications of OCR software can be found in three areas of interest:ocr software
1) ANPR, Automated Number Plate Recognition
2) industrial track & trace of printed codes
) traceability in supply chain
ocr software

In our 30+ years of machine vision experience we have built up practical knowledge to ensure that best results in the above four steps. This experience translates into one important property: system-reliability.
No thanks, it’s our pleasure!

Application Request Form

  • Drop files here or