Microscan, a global technology leader in barcode, machine vision, and
lighting solutions, announces the latest innovation from its award-winning
AutoVISION machine vision suite: IntelliText OCR. With advanced optical
character recognition (OCR) functionality, AutoVISION’s IntelliText OCR tool is
capable of converting human-readable characters into machine-readable characters
with the most aggressive algorithms available. Allowing user control of
customizable parameters, IntelliText OCR can be quickly adjusted to recognize
characters regardless of marking or printing method, including low contrast
text on poor backgrounds.
Optical character recognition is a process by which software converts
human-readable text into characters that can be stored, interpreted, and
segmented by machines. Optical character recognition technology has been used
extensively in commercial applications since the 1970s, and today plays a role
in the automation of tasks from document processing to consumer goods packaging
(batch codes, lot codes, expiration dates) to clinical applications. OCR is
accomplished when an image captured by a camera is interpreted by OCR software,
such as a machine vision system, which has additional capabilities such as
barcode reading and product inspection.
With the release of Microscan’s latest user-friendly machine vision
software, AutoVISION 3.0, Microscan has introduced the IntelliText OCR tool
with advanced OCR functionality for reading the most difficult characters on
parts and products in automated identification, tracking, and inspection
applications. IntelliText OCR is capable of reading text printed by various methods,
including inkjet, Drop on Demand (DOD), direct part marking, and more. The
tool’s multi-neural network allows it to train on character variations and
store these in a font library for increased OCR speed as the library grows.
Advanced character segmentation in IntelliText OCR allows the software to
easily parse characters regardless of uniformity of each character or the
precision of the print region (useful when print consistency, label placement,
or text location is subject to variation). To aid segmentation in difficult
reading environments, IntelliText OCR offers image pre-processing, enabling the
software to run filters on an image taken by a machine vision camera to produce
the cleanest image possible for OCR.
Unique to IntelliText OCR is the software’s image binarization process,
which converts the grayscale image taken by the camera to a binary image. The
binary image allows the user to see the image features that the software is
able to recognize as characters and gives users the ability to set tolerances
that determine how much of the image is in view. This allows the most difficult
text to be adjusted for and read with ease. With adjustments to image
binarization, users have the power to tailor the software to anticipate and
enhance dark text on dark backgrounds, light text on light backgrounds, text on
damaged surfaces, or even text printed on challenging surfaces such as
transparent packaging. The tool can even interpret rotated text in cases where
product codes are printed or labeled at variable angles.
IntelliText OCR supports advanced string matching, which is the process
of comparing an interpreted string of text to match an expected string. This
includes using regular expressions (TRE) as a match string. A regular
expression is a standardized way of defining variable text, for instance when
checking the proper formatting of a date code. The regular expression match
string can be set to check a range of acceptable numerals for an expiration
date, including limits for year numerals and limits for the initial character
in a month or date numeral (for instance, a month numeral beginning with 0 or
1).
The AutoVISION machine vision family continues to be the industry’s most
flexible machine vision solution, allowing users to build a single, scalable
vision system from a range of tools. Now with IntelliText OCR, AutoVISION also
offers the industry’s leading character recognition technology.
