Learn how MVTec’s HALCON deep-learning-based anomaly detection works. This area of their deep learning tools enables the detection of deviations from known training data.
Training a network for this only requires a relatively small number of ‘good’ images and eliminating the need for data labelling. Combined with the simple and flexible workflow HALCON enables fast and easy prototyping, allowing you to rethink existing applications. The training of anomaly detection can be performed on a standard CPU.
To maximise its potential in industrial environments, HALCON’s deep-learning-based image classification, semantic segmentation, object detection, and anomaly detection can be performed on GPUs, on x86 CPUs and on Arm® processors.