HALCON's
OCR Assistant lends a helping hand
Go to MVTec
Go to Machine vision
software
Go to Halcon 11
Go to Halcon 12
The OCR Assistant, one
of the assistants in HALCON's HDevelop software, is a front-end
to HALCON's optical character recognition software. |
|
|
Using
the OCR Assistant will enable you to: |
• |
easily and quickly set parameters
for optical character recognition (OCR) with the Quick Setup, |
• |
segment text by choosing parameters
suitable for the characters appearance |
• |
choose a pretrained classifier
or train your own classifier |
• |
choose the kind of results you
need |
• |
generate code for your OCR application |
|
|
Simple
to use and customise, if needed |
HALCON uses a classification approach
for optical character recognition that requires segmentation of
individual characters from their background in an image and then
reads each character using a pre-trained classification.
HALCON includes many pre-trained fonts for classification, but
also enables users to train their own custom fonts.
A GUI tool is designed that:
• automatically determines or optimizes the appropriate
parameters for character segmentation
• tests and optimizes different classifiers for
desired results
• trains custom fonts
• automatically generates HALCON code for OCR applications.
|
|
Getting started
Setting up an OCR task is very easy with the OCR Assistant's Quick
Setup. It is done in five steps. The quick setup tool uses basic
image information, and so Halcon offers the option to further optimise for more challenging
OCR applications.
If the Quick Setup iss successful, the characters
found are read and the Results tab is automatically opened. Otherwise,
the segmentation tab opens and allows you to adapt parameters while
simultaneously observing the segmentation results. Helping
to customise
The OCR assistant supports customisation by providing a graphical
interface for manually selecting the best parameters for segmentation.
Users can adjust settings like character appearance, size, shape
features, orientation range, fragmentation, and layout (number of
lines of text, expected number of characters, etc). Results of the
parameter settings are displayed in real-time on the sample image.
|
How it works
The general process of an OCR application is seen in the figure
below which shows how a sample is found in an image via segmentation
and can be directly classified if an OCR classifier is available.
The sample is then assigned to a certain class, a symbol.
If no suitable classifier is available, samples can be added to
a training file from which a classifier can be trained that can
subsequently be used to classify a sample. The symbol class is typically
equivalent to a simple character. |
|
Adept Turnkey Pty Ltd are "The Machine Vision and Imaging Specialists"
and distributor of MVTec products in Australia and New Zealand.
To find out more about MVTec's products or any machine vision
product, please contact
us or call Perth (08) 9242 5411 / Sydney (02) 9905 5551 / Melbourne
(03) 9384 1775
|