STEMMER IMAGING

HALCON 22.11 provides a wide range of new and improved machine vision features:

• 3D Gripping Point Detection
  The new 3D Gripping Point Detection technology highlights the new HALCON 22.11 release. It can be used to robustly detect surfaces on any object suitable for gripping with suction. In contrast to classic bin-picking applications, this new technology eliminates the need to train object surfaces. This means that no prior knowledge of the specific objects is required, enabling the much faster and cost-efficient implementation of typical bin-picking applications, such as those in the logistics industry.

• Data Type Memory Block
  The HALCON 22.11 edition also introduces a new data type called “memory block”, which can be used to store and transfer binary data in MVTecs software HALCON as well as further process it with other applications. This increases the software's compatibility with machine communication protocols, such as OPC UA or image acquisition interfaces, e.g. for storing camera configuration files. Furthermore, there is the option to encrypt all data that can be serialised, which significantly increases data security. In this way, trained deep learning models can now also be protected. This allows users to protect their expertise and investment that has gone into collecting the data and training the models. Moreover, it can be ensured that only authorised users use the respective model.

• Guided GradCam
  The Guided GradCam supplies even more precise clues, in the form of a heat map, as to which regions of the image are relevant for the decision made by the deep learning network. For example, this makes it possible to investigate misclassification.

• HAILO AI Acceleration Hardware
  MVTec's machine vision software HALCON 22.11 now also supports HAILO AI acceleration hardware, which can be used via a plug-in to execute deep learning inferences very quickly. This broadens the range of available hardware and increases flexibility when it comes to using the best components for each application.

• Network Licenses
  With HALCON 22.11, MVTec expands the licensing possibilities by adding the option to license HALCON via a network. A license server allows the use of floating licenses. Here, developers share a predefined number of licenses using a network connection. Customers benefit from cost savings due to multi-usage and greater flexibility in user allocation, developers enjoy greater independence and flexibility regarding their work location. Especially for distributed or remotely working development teams, this is the perfect way to effectively make use of HALCON’s powerful machine vision algorithms. Besides this, the new mechanism enables users to work in virtualised environments without a permanent physical host ID.

HALCON 22.05 provides a wide range of new and improved machine vision features.

• Global Context Anomaly Detection
  The next evolution of HALCON’s Anomaly Detection technology is a solution for even the most complex anomaly detection applications. Global Context Anomaly Detection is now able to detect unknown anomalies in a global context, check the completeness and detect defects. The development of quality inspection for unknown anomalies is fast. It is ideally suited for quality control and more.

• Training for Deep OCR
  HALCON`s Deep OCR enables users to efficiently solve text reading applications in a multitude of use cases. With HALCON 22.05, this technology is extended by training functionality, enabling application-specific training. It retrains Deep OCR to your specific application. Furthermore, this advanced OCR technology can now read uncommon printing styles and has improved readability in bad lighting conditions and contrast. Users get better results with less parameter tuning and solve even the most complex OCR tasks.

• Enhanced ECC200 Print Quality Inspection
  The print quality inspection for ECC200 codes has been further improved to help solve code reading/printing applications more effectively. It is now up to 150% faster. The grid determination module is now more robust and the usability has been improved by introducing additional procedures. This tool can be used in the pharmaceutical industry or in other code-printing industries.

• Notable Quality-of-Life Improvements
  A high number of smaller improvements complete HALCON 22.05, including new operators e.g. for smoothing with arbitrarily shaped regions, a new histogram equalization operator and speedups. In addition, Generic Shape Matching is now included in HDevelop’s Matching Assistant.

HALCON 21.11 provides a wide range of new and improved machine vision features:

• AI Accelerator Interface (AI²)
  A generic, documented interface similar to MVTec’s Acquisition Interface which allows to implement "Plug-ins" that can be used with MVTec products. This generic interface provides full flexibility with deep learning hardware. So far, TensorRT and OpenVINO toolkits are available, with more to follow.

• OpenVINO toolkit plugin
  OpenVINO is a toolkit for high-performance deep learning on Intel hardware. The toolkit extends workloads across Intel hardware (including accelerators) and maximises performance. OpenVINO itself provides plugins for certain Intel hardware such as Intel CPUs, Intel GPUs, Intel VPUs (e.g., Intel Movidius Neural Compute Stick). HALCON 21.11 supports the OpenVINO toolkit to speed up the inference. Using the OpenVINO toolkit plugin can drastically shorten the runtime and lower the memory consumption which in turn offers more flexibility in terms of the hardware.

• Code128 reader improvements
  With HALCON 21.11, HALCON’s bar code reader is improved in terms of robustness in case of blurred Code 128/GS1-128 codes. Now, codes with a larger amount of blur can be read. Blur on such codes can occur due to motion or due to limitations in depth of focus. The Code 128/GS1-128 is a widely used bar code type that is frequently used in logistics due to its compact size and high data density.

• Deep Learning Instance Segmentation
  With HALCON 21.11 MVTec extends the functional scope of its deep learning features with a new technology called “instance segmentation”. This combines the advantages of semantic segmentation and object detection. With the help of instance segmentation, objects can be assigned to different classes with pixel accuracy. This technology is particularly useful in applications where objects are very close to each other, touch or overlap. Typical use cases also include grabbing randomly arranged objects from boxes (bin picking) as well as identifying and measuring naturally grown structures.

• Simplified syntax for dictionaries (HDict)
  Dictionaries make it easy and convenient to manage complex data in HALCON. For example, different data types such as images, ROIs and parameter settings can be bundled in a single dictionary. This allows programs to be structured in a logical way, for example when passing many parameters to a procedure. HALCON 21.11 includes several improvements that make the handling of dictionaries even easier and faster. For example, dictionaries can now be initialised with a single operator call, and the syntax for adding and retrieving elements has been simplified. In addition, the auto-completion now also suggests the keys contained in the dictionary, which further speeds up and simplifies working with dictionaries.

• Future-proof interface for shape matching
  With Generic Shape Matching, HALCON offers user-friendly access to MVTec's industry-proven shape matching technologies. Thanks to the significant reduction in the number of required operators, users can implement solutions more easily and quickly. With HALCON 21.11, existing functionalities are enhanced based on customer feedback to further increase usability. For example, the clutter feature has been integrated, handle inspection has been optimised, and additional parameters have been integrated and included in the automatic parameter estimation.

The new version implements new features in the areas of deep learning and matching as well as improvements to the subpixel bar code reader, Deep OCR, and the integrated development environment HDevelop. HALCON 21.05 Progress gives users the benefit of even more robust machine vision processes and improved usability.

• HDevelop usability improvements
  HDevelop’s new window docking has been improved. Users have now more options to control the position where floating windows are opened.

• Deep OCR improvements
  With HALCON 21.05, the performance and usability of Deep OCR have been improved. Big images are now handled more robustly and the result now contains a list of character candidates with corresponding confidence values, which can be used to further improve the recognition results. Customers also benefit from overall improved stability and the fact that they can address a wider range of possible applications, thanks to additional character support.

• Deep Learning Framework
  HALCON 21.05 introduces a first version of the HALCON Deep Learning Framework. This framework allows experienced users to create their own models within HALCON. With this feature, experts can now realise even the most demanding and highly complex applications in HALCON without having to rely on pre-trained networks or third-party frameworks.

• Generic Shape Matching
  HALCON 21.05 introduces Generic Shape Matching, which makes MVTec's industry-proven shape-matching technologies even more user-friendly and future-proof. By significantly reducing the number of required operators, users can now implement their solution much faster and a lot easier. Moreover, thanks to the unification of HALCON’s different shape-matching methods into a single set of operators, users can now integrate new shape-matching-related features more smoothly.

• Subpixel bar code reader improvements
  HALCON’s subpixel bar code reader is capable of reading codes with very thin bars. In HALCON 21.05, the subpixel bar code reader has been improved regarding low-resolved codes. The decoding rate for those can now increase up to 50%.

• Improvements of basic operators in 2d and 3d for fast and robust preprocessing
  3D point cloud sampling now supports a new mode called “furthest point”, which typically results in a more uniform sampling of a 3D object. The user sets the number of output points and then iteratively adds to the output object the point of the input object that is furthest from all points already added to the output model. The furthest point mode usually allows a reasonably uniform sampling. The 3D point cloud smoothing has been extended by a new mode that uses information from the XYZ mappings. 3D point cloud smoothing can be used as a preprocessing step to smooth point clouds and remove noise. This mode usually leads to a much faster processing time.

Further improvements in HALCON 21.05 include general speedups of basic image operators.