MVTec’s latest version, Halcon 19.05 comes with outstanding new features, such as improvements for surface-based matching and enhanced shape-based matching. It also features the possibility to execute the deep learning inference on Arm-based CPUs and an enhanced object detection.
HALCON 19.05 Feature Spotlight: Improvements for Surface-Based and Shape-Based Matching
Users of shape-based matching in HALCON 19.05 can now specifically define “clutter” regions. These are areas within a search model that should not contain any contours. This leads to more robust matching results, for example in the context of repetitive structures.
Furthermore, HALCON 19.05 offers an even more robust, edge-supported surface-based matching: Users can control the impact of surface and edge information via multiple minscores. Additionally, a new parameter now allows switching off 3D edge alignment entirely, enabling users to eliminate the influence of insufficient 3D data on matching results, while retaining the valuable 2D information for surface and 2D edge alignment.
HALCON 19.05 Feature Spotlight: Inference on Arm Processors
HALCON’s deep learning based image classification has been optimised for NVIDIA GPUs and Intel CPUs since HALCON 17.12. With the latest version, HALCON 19.05, you can also execute the deep learning inference on Arm-based CPUs, which allows the deployment of deep learning applications on embedded devices without the need of any further dedicated hardware. All three deep learning technologies image classification, object detection, and semantic segmentation run out of the box on Arm-based embedded devices.