Major New Features of HALCON 19.11.0.0 Progress


Convenient Inspection with Anomaly Detection

The new anomaly detection feature significantly facilitates the automated surface inspection for, e.g., detection and segmentation of defects. With HALCON 19.11, you only need a low number of high quality images for training. The technology is able to localize deviations, i.e. defects of any type, on subsequent images. This means, defects of varying appearance can be detected without any previous knowledge or any preceding labeling efforts. The training of anomaly detection can be performed on a standard CPU.


More Transparency with the Grad-CAM Heatmap

Deep learning networks are often considered as a black box because users do not know what happens with the data during the inspection process. Therefore, it is very difficult to debug in case of misclassifications. HALCON’s newly implemented Grad-CAM (Gradient-weighted Class Activation Mapping) supports you in analyzing, which parts of an image have a strong influence for the inference into a certain class. The so-called heatmap based on Grad-CAM is very fast compared to the formerly offered method.


Support of the ONNX Format

HALCON 19.11 is able to read data in ONNX format, allowing to use previously created 3rd party networks for classification and object detection within HALCON.


Speedup of the ECC 200 Data Code Reader

In HALCON 19.11, the code reader for ECC 200 codes has been significantly accelerated for multi-core systems. This affects especially codes that are difficult to detect and read. This speedup also increases the viability of embedded-based code readers by taking full advantage of existing hardware capacities.


Generic Box Finder for Pick-And-Place Applications

The generic box finder allows users to locate boxes of different sizes within a predefined range of height, width, and depth, removing the need to train a model.


Calibration of Telecentric Line Scan Cameras

HALCON has been extended with a new camera model: It provides the functionality to calibrate line scan cameras with telecentric lenses. Images acquired with such telecentric line scan cameras can now be rectified. Thus, measurements can be performed with high precision.