HALCON 20.05 has been released today and once again, the HALCON Progress Edition shows the advantages of a short release cycle: Besides innovative new features you will find optimized functionalities from the last release and incorporated customer feedback.
The new version comes with good news for all machine vision users: With HALCON 20.05, training for all deep learning technologies can now be performed on the CPU of standard industrial PCs. This greatly increases customers' flexibility regarding the implementation of deep learning, as training can now be carried out directly on the production line. The new independence from dedicated GPUs makes it possible to adjust the application to changing environmental conditions “on the fly”.
But that’s not all!
In HALCON 20.05, the Grad-CAM-based heatmap (Gradient-weighted Class Activation Mapping) calculation can also be performed on the CPU. The heatmap supports you in analyzing which parts of an image influence the classification decision. Since this can be done without significant speed drops, users are now able to analyze their deep learning network's class prediction also during operation.
HALCON 20.05 has many more highlights in store!
To name a few:
- Reading very small codes with the subpixel bar code reader
- More accurate and robust matching results with surface-based 3D matching
- More robust generic box finder for pick-and-place applications
- Significant improvements for anomaly detection
- Support of the language interface .Net Core
All technical details can be found in the documentation