OPC UA Vision expands the possibilities of information exchange between image processing systems as well as between image processing and machine control, line controllers, software systems or to SCADA, MES and ERP systems.

© Asentics

According to Dr. Heinol-Heikkinen, the Companion Specification is more than just an ordinary interface: "As a digital representation of a vision system, OPC UA Vision defines all the properties and characteristics of vision systems. Accordingly, it is not just about data and a data interface, but rather about information, i.e. the interpretation and meaning of data. This is mapped in a so-called data model for image processing systems. According to this model, information becomes the structural properties of a vision system and is the first component of the Companion Specification."

OPC UA Vision consists of two essential elements: the 'state machine' and the information model.

© Asentics

The data model was initially kept very 'simple' and general in order to cover the variety of existing image processing systems and their applications. It describes the standardized management of recipes, configurations and results. These three central management objects themselves and the methodical handling of them are obligatory, whereby the contents of these objects remain manufacturer-specific and are treated as black boxes. Although this generic description of the information model is still somewhat 'fuzzy', it also offers an important advantage: it forms a common bracket around the multitude of different existing systems and covers all common products in monochrome and color, from the vision sensor to the intelligent camera to the complex inspection system, from the line scan camera to area scan cameras to the 3D camera. Even the light barrier, which can be regarded as an image processing system with just one pixel, is covered.

However, as not all of the model's information about a vision system is of interest or actually available at all times, there is a second key component of OPC UA Vision: the state machine. This abstracts the behavior of image processing systems, i.e. the possible operating states and state transitions of an image processing system are mapped on an abstract state model with a fixed number of states. This means that every interaction with the image processing system and the associated concrete access to information in the information model is also dependent on the respective state of this model. The information model in conjunction with the state model thus forms the digital representation of any image processing system.

Image processing users benefit

The scope of OPC UA Vision is not only to supplement or replace existing interfaces between vision systems and the process environment with OPC UA, but also to create non-existent horizontal and vertical integration capabilities to exchange relevant data with other authorized process participants - such as the IT enterprise level. To this end, the OPC UA vision interface enables the transfer of information between different vision systems as well as the transfer of information between vision and machine control, line controllers or any software systems at the control level as well as at the higher level of MES, SCADA or ERP systems.

The added value of OPC UA generally lies in the platform- and manufacturer-independent networking of automation components. All systems used should communicate with each other in a simple, standardized and secure manner, and OPC UA Vision also meets this requirement. For users, this means faster development of machine vision systems as well as simplified configuration and integration into existing automation systems. As sensor technology and image processing are among the most important sources of information in modern Industry 4.0 production systems, OPC UA Vision is a crucial building block on the way to more efficient production.

Karlheinz Hohm, Vice President R&D Industrial Automation at Isra Vision, shares this view: "The simplified integration of image processing into automation makes production processes more efficient and at the same time more flexible. In particular, expert knowledge is no longer required for integration and use, so users can automate their production quickly and easily themselves."

In view of these obvious advantages, it is not surprising that OPC UA Vision is strongly supported by the industry. Karlheinz Hohm, for example, sees the current developments as a positive approach towards a general, manufacturer-independent interface. "We attach particular importance to the activities in the area of OPC UA Vision, as this simplifies the integration of image processing systems for users and thus opens up further fields for their use and further automation." Isra Vision was consequently actively involved in the 'Core Working Group' of the 'VDMA OPC UA Vision Initiative' in the development of the 'OPC UA Vision Companion Specification'.

Expansion plans

The definitions described in Part 1 of OPC UA Vision are aimed at the basic capabilities of the system and not at the image processing capabilities, i.e. not at the application software that distinguishes, for example, a code reading system from a surface inspection system or a robot vision solution. In Dr. Heinol-Heikkinen's opinion, the next step is certainly to describe this level, i.e. to create a terminology and taxonomy, in order to map it in capability-based information models: "For me personally, this would be a goal worth striving for, which would help machine vision to further expand its product status."

According to Dr. Heinol-Heikkinen, with the publication of OPC UA Vision, "a start has now been made and an umbrella has been built under which many different systems can stand. Sharpening the description is one facet that will certainly follow."

Author:
Peter Stiefenhöfer is the owner of PS Marcom Services in Olching.

The 'OPC UA Vision' working group

The companies Asentics, Bosch, Isra Vision, Kuka, Peer Group, SAC, Silicon Software, Stemmer Imaging, Unified Automation and Vitronic were involved in the definition of 'OPC UA Vision'. This 'Core Working Group' held intensive working meetings at the VDMA in Frankfurt over a period of around 18 months and also held numerous web meetings. The entire working group, known as the 'Total Working Group', consisted of around 60 members. There were three joint meetings within the 15-month working period.