Technology has changed faster and more groundbreakingly than ever before in the last few decades. The human eye has long since been replaced by machine vision in many areas. But classical image processing is also reaching its limits with and in increasingly complex processes. This is where an area of artificial intelligence (AI), Deep Learning, comes into play and makes its way into production.
The possibilities offered by Deep Learning are great. At the same time, the new technology is intended to complement image processing and not replace it. It is important to Quality Automation GmbH to always be at the cutting edge of technology and to be able to pass this benefit on to our customers. Therefore, we are continuously researching this topic within the framework of final theses or feasibility studies. We use the Halcon and Merlic software from MVTec and the ISD NXT embedded vision platform from IDS.
As a subfield of artificial intelligence, deep learning technology uses neuronal networks to imitate human intelligence. The mathematical models are characterised by the ability to learn complex correlations and to be trained on annotated examples. A neural network consists of a large number of similar processing units, the neurons, which are interconnected. As a result, Deep Learning combines the flexibility of human visual inspection with the speed and stability of a computer-based system.
AI is used to support the following applications: