AI Solutions for Digitalization and Sustainability

Machine Learning in Image Processing for Production and Industry

In recent years, Deep Learning techniques have been used extremely successfully in the field of image processing. In addition to Deep Learning, there are many other Machine Learning methods such as the support vector machine. One challenge for the developers of all these methods is: how can these algorithms be used safely and stable for the optical quality assurance in production? This challenge arises from the special features of many learning methods:
 

Hybrid Methods and »classic« Solutions for Success

Methods such as »Deep Learning« require a large amount of annotated data, e.g. of defects found in a production facility. However, in a well-functioning production environment, many images of faultless products are available, but only a few of products with defects. One possibility is then to perform data augmentation, i.e. artificial defect databases are created on the basis of the real defect data. Another solution would be to describe the defects mathematically and then learn this model.

It is also difficult to change the inspection level of a machine-learning based inspection system during production, e.g. to set certain quality levels. We therefore frequently use hybrids based on »classical« parameterizable methods (filters, morphology, edge detectors) and machine learning.

In addition to solutions for production, we also offer »typical« Machine Learning solutions for image processing. These are often projects in which huge amounts of image data are currently processed manually and this process is now to be automated by software.

Overview Example Projects

 

Surface Inspection of Wooden Panels

We automate defect detection on surfaces such as wooden panels using hybrid methods that combine image processing with Artificial Intelligence (AI).

 

Efficient Humanitarian Aid through Intelligent Image Analysis

EDDA is an AI software that analyzes drone images to determine the extent of disasters in real time.

 

Signal Analysis in the Railway Sector

The monitoring of hot-running axle bearings and fixed brakes on passenger trains and freight trains requires a non-contact measurement method.

 

»Viewpoint of Interest« (V-POI)

V-POI enables inspection experts to plan, simulate and generate without computer graphics and modeling skills.

 

Determining Wood Species Using Artificial Intelligence (AI)

In the project »KI-WOOD«, we are developing automatic image recognition systems to identify wood species using AI.

 

Project »eQuality«

In the project, we are developing »eQuality«, a digital defect library that supports companies in production with the inspection using Artificial Intelligence and the standardized recording of defects.

 

Pflegeforensik: With AI Against Billing Fraud

In the »PflegeForensik« project, we are developing AI software to combat billing fraud.

 

Fraunhofer Lighthouse Project ML4P

Machine Learning for Production

In the Fraunhofer Lighthouse Project, seven Fraunhofer Institutes bundle their extensive experience in the field of Machine Learning.

 

Hyperspectral Image Processing

In the project »Hypermath«, we have developed possibilities for the simple visualization of hyperspectral images.

 

 

Granules and Pellets

The 3D shape of particles influences further processing decisively. How can the shape be captured and described?

 

EniQmA

In the project »EniQmA« (Enabling Hybrid Quantum Applications) we work on systematizing hybrid approaches in the field of quantum computing (QC) in a targeted way.