Fraunhofer Institute for Industrial Mathematics ITWM

Qualitätsprüfung — © iStockphoto
Image compositing between photography and 3D background. Hand turning a knob with a dial on the right side, blur effect. Concept of TQM, Total Quality Management or improvement.

Quality through Algorithms

Due to the algorithm portfolio, the department can safely inspect products for defects.

Inline Quality Control for Production

The focus of this field of activity is the development of efficient and innovative image-based complete solutions for automated quality control and quality assurance in the inline production process. At the core of our image processing software, we offer mathematical methods and algorithms for image analysis, image processing and their implementation.

Our systems offer customized and individual solutions that emulate the quality control established in the company. In addition to the use of classic image processing algorithms for surface inspection, we have been working with machine learning methods for many years and also develop so-called hybrid systems, which consist of a mixture of both methods. This mixture is particularly suitable for applications with a broad product portfolio and low production errors. We also carry out feasibility and preliminary studies, define the necessary hardware components, develop the system software and the graphical user interface for operation in close consultation with our customers and optimize the system through to project acceptance.

Selected Examples of 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).

MASC-DISQ – Surface Inspection of Gaskets

The image processing system DISQ was developed for surface inspection of metal gaskets.

MASC-Expansion Cells

MASC expansion cell is a robust, and simultaneously high-precision inspection system for expansion cells, which is used in an industrial environment.

MASC-STEX – Ceiling Panels

Since 2003, MASC-STEX has been successfully operating at a large German manufacturer of ceiling panels.

Machine Learning for Production

Seven Fraunhofer Institutes are pooling their extensive experience in the field of machine learning in the ML4P lighthouse project.

Automated Blisk Inspection

We have developed a fully automatic solution for the surface inspection of BLISKs (bladed disks) as part of the research program »Clean Sky EU Aviation«.

Cloudiness – The Homogeneity of Filter Media

We developed a theoretically sound cloudiness index and a method to measure it from image data.

Process Control and Monitoring Using Edge Computing

The »EMILIE« project focuses on improving decentralized data acquisition and processing through edge gateways.

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.

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.

Hyperspectral Image Processing

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

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.