HALCON 18.11 - the newest version of MVTec HALCON - has been released today and is now available for download.
Many highlights in the new release are the result of MVTec's ongoing development of AI technologies, such as deep learning and convolutional neural networks (CNNs).
For example, object-, feature-, or error classes trained with deep learning can now be segmented with pixel accuracy. Similar to classification, segmentation can be performed on a GPU as well as a CPU. MVTec provides pretrained networks based on millions of royalty-free images for training the required network. Customers can train new objects more easily, since they do not require hundreds of thousands of their own application images. This paves the way for an entirely new range of applications, which previously could not be realized, or only with significant programming effort. As the images contained are free of third-party rights, they can be used in commercial applications without reservations.
Localizing object-, feature-, and error classes
Another helpful feature of the new HALCON version is the localization of trained object-, feature-, or error classes in an image – also with the help of highly-developed deep learning algorithms. In contrast to pixel-precise semantic segmentation, the sought-after objects are each marked in the image by a surrounding rectangle (a so-called bounding box). This leads to new possible applications in the area of quality control. A wide range of complex machine vision localization tasks can be re-evaluated and performed with far less effort than conventional methods. The object detection inference also runs on both a GPU and a CPU, which offers significant advantages for use in industrial environments. Pretrained networks are likewise available to make it easier to "train" new objects. Furthermore, this feature can be used to reliably identify and thus count objects that touch each other or partially overlap.
Moreover, HALCON 18.11 provides a crucial advantage for the world of embedded vision. The new release now also runs on the increasingly popular 64-bit Arm® architecture right out of the box.
A USB3 Vision interface for both 32 and 64 bit is included as well. Further improvements are the optimization and the greater flexibility of data code reading. For example, the ECC200 data code reader is much faster and capable of reading codes with missing or damaged/disrupted "quiet zones”. In addition, codes against complex backgrounds can be found and read faster and more robustly.
Valuable new features also for developers
Developers too will benefit from the new release. HDevelop, HALCON's integrated development environment, now displays detailed information about important handle variables, which is very helpful for debugging: It allows users to easily identify and check the current features of complex data structures at a glance. Double-clicking on a handle variable displays all the assigned parameters and their current settings. Even complex parameters with multiple key-value pairs, such as the camera parameters in 3D shape-based matching, can be easily checked.
The field bus connection is another useful feature for developers. Thanks to the Hilscher CifX interface, HALCON can communicate with almost all industrial field bus protocols. Profibus, ProfiNet, and Ethernet/IP, among others, are supported.
Moreover, HALCON 18.11 fully supports UTF-8 encoding, which means that the software can be easily used in multiple languages simultaneously. This particularly benefits companies that operate across different countries, or that have employees working in different languages. As a result, Western alphabets and Asian characters are now displayed in the same encoding and no longer have to be converted from one to another, reducing error rates.
Available in two editions
The new HALCON 18.11 release comes in a Steady edition and a Progress edition. While the latter is available by subscription and has a six-month release cycle, the Steady version is offered for regular purchase with a release cycle of two years.