Object Recognition With Pat Genius; Pat Genius Object Recognition - Siemens SIMATIC MV500 Operating Instructions Manual

Combination of decoder and OCR steps
Note the following points when you use decoder as well as OCR/locator steps in a program:
• The standard report belonging to the selected verification standard or, for "Uncalibrated",
to the read code type, is used.
• The OCR/locator grades can also be output via the relevant String inputs.
3.3

Object recognition with PAT Genius

Definition
The term "object recognition" describes methods for identifying a known object within an
object space using optical, acoustic or other physical recognition processes. The presence of
an object and its position and orientation are determined in an image or section of the image.
Optical recognition process PAT Genius
PAT Genius is an optical recognition process for object recognition. Prior to recognition, the
object is stored as a model in a so-called model library. With the help of this model, the object
can then be searched for and recognized in the image.
Application
Object recognition is a function often required in industrial image processing. With PAT
Genius object recognition, the SIMATIC MV500 reader has a very extensive, powerful and
flexible image processing library.
3.3.1

PAT Genius object recognition

As of firmware V2.1, PAT Genius object recognition can be integrated seamlessly into the
existing operator control and application concept of the SIMATIC MV500 optical readers.
• For objects to be recognized, you can create so-called models in the WBM of the reader
("Program > Locator"). These models are stored in a model library that can be used in all
steps of a program.
In compatibility mode, a model library can also be used for multiple programs.
Depending on their size, several models and also very different objects can be stored in
each model library.
• With the license-free version of the locator ("Locator light"), simple applications with a
reduced scope of functions can be created. For example, only one model can be created
per program step.
• In one program, up to twelve object recognition steps can activate different models of a
model library. This means that a single program can flexibly implement a task that can be
divided into a maximum of twelve subtasks. The subtasks can be very different and
SIMATIC MV500
Operating Instructions, 03/2021, C79000-G8976-C494-05
Image processing
3.3 Object recognition with PAT Genius
75