In particular, the expert team at MVTec has enhanced the deep learning technologies available in HALCON. One highlight: In version 19.05, the deep learning inference can also be executed on CPUs with the established Arm® processor architecture. This allows customers to use the latest deep learning technologies on standard embedded devices. Deep-learning-based object detection has also been improved: The method, which locates and identifies objects by their surrounding rectangles (so-called bounding boxes) now precisely detects the orientation of the objects.
Significant improvements in 2D and 3D matching
Other new features in HALCON 19.05 optimize matching processes. For example, users of HALCON’s shape-based matching can now define areas within a search model that should *not* contain any contours. In the context of repetitive structures, this leads to more robust matching results.
Moreover, the new HALCON release also offers several enhancements for surface-based 3D matching. This means that additional parameters can be used to better inspect the quality of 3D edges, resulting in even more robust matching – especially in the case of noisy 3D data.
Optimized usability in embedded environments
“With version 19.05, we are raising our proven HALCON standard machine vision software to a new level. Developers and users alike benefit from numerous new features. These not only include additional, well-thought-out deep learning functions, but also noticeably improved usability in embedded environments,” explains Johannes Hiltner, HALCON Product Manager at MVTec.
Significant improvements in 2D and 3D matching
Other new features in HALCON 19.05 optimize matching processes. For example, users of HALCON’s shape-based matching can now define areas within a search model that should *not* contain any contours. In the context of repetitive structures, this leads to more robust matching results.
Moreover, the new HALCON release also offers several enhancements for surface-based 3D matching. This means that additional parameters can be used to better inspect the quality of 3D edges, resulting in even more robust matching – especially in the case of noisy 3D data.
Optimized usability in embedded environments
“With version 19.05, we are raising our proven HALCON standard machine vision software to a new level. Developers and users alike benefit from numerous new features. These not only include additional, well-thought-out deep learning functions, but also noticeably improved usability in embedded environments,” explains Johannes Hiltner, HALCON Product Manager at MVTec.