Newest features of MERLIC 5.6
With the new version, MVTec continues it's established path with MERLIC - easy process integration combined with powerful machine vision methods. That's why MERLIC 5.6 again includes an interface that further simplifies the process integration of MERLIC. There will also be exciting new features.
What's changing?
In terms of improved connectivity, version 5.6 will feature a plug-in for the Siemens SIMATIC S7 PLC. In addition, the existing plug-in for the REST API has been expanded to include new functionalities for direct access to image results. The new machine vision methods include the option for quality control of 2D data codes and surface reconstruction with photometric stereo. The latter is also used for quality control. Customers can also look forward to enhancements in AI-based deep learning applications.
Communicator plug-in Siemens S7
The new version of MERLIC includes a plug-in for communicating with Siemens SIMATIC S7 PLCs. This direct access to the widely used Siemens PLC further increases the connectivity of MERLIC. For customers, the plug-in offers improved usability as well as faster and direct integration into the production environment.
Simplified image source parameterization
The parameterization of the image acquisition makes it possible to easily switch between different predefined camera parameter sets. This leads to improved usability and time savings and enables new use cases, as machine vision applications in MERLIC (so-called MVApps) can be executed with specific camera settings.
Image transfer with REST plug-in
The REST plug-in has been available since MERLIC 5.4. This interface opens the door to the world of REST API web services. In the new MERLIC version, images can now also be accessed via the interface. This is useful, for example, for integrating images from MERLIC directly into HTML websites which enable users to control and monitor production processes. That means, customers can integrate images from the REST API directly into their applications. Furthermore, an extended image memory enables retrieving images from MERLIC asynchronously via the plug-in.
Support of optimized models in the deep learning tools
Deep learning models optimized for AI2 accelerators can be used directly in the corresponding MERLIC tools. The typical workflow is a training of the model in the Deep Learning Tool. When exported, the model can be optimized for a target hardware and then used directly in MERLIC. In MERLIC, no further adjustment of the model is necessary, so that a significant time saving is achieved when loading programs.
Print quality of 2D data codes
Quality control of 2D data codes is particularly important for the production industry. The new method in MERLIC 5.6 does not only read the code, but also evaluates its print quality. The print quality of the codes is determined in accordance with the AIM DPM-1-2006 and ISO/IEC 15415 standards. The new method speeds up the process of checking the print quality of labels on goods, which is very important for many companies.
Searching tool within the Tool Flow
This feature improves the clarity of sophisticated tool flows within MERLIC. This allows users to search explicitly for specific tools.
Reconstruct Surface with Photometric Stereo
Photometric stereo is a technique in machine vision used to reconstruct the 3D surface structure of an object. This approach is often used for detecting defects. The corresponding feature “Reconstruct Surface with Photometric Stereo” is initially available in MERLIC as a Concept Tool. The lighting required for this technology can be easily controlled within MERLIC via GenICam.
Provide MERLIC documentation online
From version 5.6 onwards, the MERLIC documentation is available on the MVTec website www.merlic.help. This allows users to share references to the documentation directly as links on the corresponding web pages.
New machine vision tools
In MERLIC 5.6, some methods that were previously available on a test basis have been transferred to completely new tools. These tools are: “Count with Deep Learning”, “Segment Image Pixel-Precisely", and “Recognize Color”.
With “Count with Deep Learning”, also known as Deep Counting, it is possible to count a large number of objects quickly and robustly with little training effort. There are many possible use cases for this feature, like for example completeness checks.
“Segment Image Pixel-Precisely" makes it possible to localize trained defect classes with pixel precision. This enables users to solve inspection tasks that previously could not be solved at all or only with considerable programming effort.
With the feature “Recognize Color”, users can reliably recognize colors under different conditions. In addition, the results can be further improved by setting the parameters accordingly. Among other things, the method contributes to quality assurance, for example in applications such as inspection tasks or selecting the right components.
