Events 2024 (Archive)

We are very pleased to announce that Mr. Chia-Wei Chen, from our Visual Inspection Systems department, has just successfully completed his doctorate at the Faculty of Computer Science at the Karlsruhe Institute of Technology (KIT) with the grade »Very Good«. The title of his thesis is: »Retroreflex Ellipsometry for Nonplanar Surfaces«.
The doctorate took place with Prof. Dr.-Ing. habil. Jürgen Beyerer Institute Director, Director Fraunhofer IOSB and our department head Prof. Dr.-Ing. Thomas Längle.
 
Congratulations and congratulations! 👏👏👏
 
His work makes a decisive contribution to expanding and securing future expertise and research in the field of retro-reflection ellipsometry (RRE) at Fraunhofer IOSB. The real-time capable image acquisition and evaluation of an RRE inspection system enables the inspection of flat and curved surfaces in a continuous process.

🎅 Anyone can have a Christmas party...we from the Visual Inspection Systems department at Fraunhofer IOSB celebrate the beginning of the year together 🎉 The "former" companions were also there.

As a team, we can look back on a successful last year 2023, with projects on current topics such as

• ASKVIT  Extracting raw materials from bulky waste
• Fruit and vegetable scanner for wholesale quality assessment
• Light field technology

and look forward to mastering new challenges together in the future.

Here's to a happy new year 2024! ✨

📝 Felix Kronenwett, Georg Maier, Norbert Leiss, Robin Gruna, Volker Thome and Thomas Längle from Fraunhofer IOSB and Fraunhofer IBP explore this question in the Waste Management & Research article:

Sensor-based characterization of construction and demolition waste at high occupancy densities using synthetic training data and deep learning

Sensor-based monitoring of construction and demolition waste streams plays an important role in recycling.♻️ Knowledge of the composition of a material stream helps to identify efficient recycling routes, optimize processing plants and forms the basis for sorting. Individual construction waste particles must be recognized even with a high material throughput. Conventional algorithms often reach their limits here, e.g. because the particles overlap. Deep learning methods, such as Faster R-CNN and YOLOv3, appear promising for more precise object recognition. In the study, various deep learning architectures were therefore examined for their suitability for characterizing construction and demolition waste. These findings are an important step towards more efficient and precise recycling in the construction industry.

Find out more in the article and click the button in the box on the right.

The main topic: »Promoting biodiversity through digital agriculture: What contribution do AI & Co. make?«

This year's host, alongside the GIL, is the University of Hohenheim.

The decline in biodiversity and its impact on essential ecosystem services have now penetrated society's consciousness. »Save the bees« and similar initiatives illustrate this. The question arises as to what extent digital agriculture can make a contribution in the area of conflict between high agricultural productivity and the preservation and restoration of biodiversity.

The Visual Inspection Systems department is involved with the following presentation:

👉A Comparative Study of RGB and Multispectral Imaging for Weed Detection in Precision Agriculture

Referent: Benedikt Fischer

The symposium will present papers from industry and academia showing new developments and applications in the field of sensor-based sorting and process control, along the entire process chain from extraction to processing, for primary and secondary raw materials.

The department Visual Inspection Systems will be involved with a presentation:

👉Fusion between Event-Based and Line-Scanning Camera for Sensor Based Sorting

Referent: Paul Bäcker on 03/13/2024 at 10:40 am

👉Intelligent Sensor-Actuator-Systems Laboratory: Particle-Specific Deflection Windows for Optical Sorting by Uncertainty Quantification

Co-author: Kevin Liang on 03/13/2024 at 11:00 am

👉Potential of Deep Learning methods for image processing in sensor-based sorting: data generation, training strategies and model architectures

Referent: Felix Kronenwett on 03/13/2024 at 11:45 am

Panel Discussion: »Closing the Implementation Gap«

Speaker: Thomas Längle on 03/13/2024 at 4:10 pm

📝 ...and what is an event camera, please?

Paul Bäcker explores these questions in the recently published article in »inspect - World of Vision«, the leading trade journal for applied image processing and optical measurement technology.

📝 Abstract: An innovative method for efficient bulk material sorting with an event camera. This technology enables significantly higher selectivity at high throughput rates and is particularly suitable for difficult sorting tasks such as sorting films or herbs. At the same time, the approach saves time-consuming and cost-intensive material stabilization measures.

📝 Find out more in the article and click the button in the box on the right.

The most important international trade event for all topics relating to measuring and testing technology as well as vision technology, image processing and sensor technology is the annual Control trade fair in Stuttgart.

Machine Learning as a Key Technology for Quality Assurance with Image Processing  

This time Fraunhofer IOSB presents with the department Visual Inspection Systems the exhibit:

Light Field Inspection-light field based light source for flexible use

In the presented inspection technology, objects with complex geometry are illuminated by a specially adapted "light recipe" in such a way that relevant structures can be detected with maximum contrast. The optimum light field for this task is determined by an algorithm that intelligently adapts the illumination modalities to the individual object geometry. This approach is groundbreaking for a future generation of inspection systems, as here image evaluation and illumination are jointly optimized for the most efficient inspection system possible. 

Joint booth of the Fraunhofer Vision Business Unit

We would like to invite you to get an overview of the new developments of the Fraunhofer Business Unit Vision and the exhibits at the joint booth of the Fraunhofer Business Unit Vision.

Visit us: Hall 8/Stand 8201
Your contact person: Christian Kludt

One evening, one topic, many guests from practice and research, two moderators - in the Experimental Fields Talks 2024, researchers and farmers get into conversation, exchange figures, opinions and arguments, illuminate different aspects of a topic from different perspectives.

🍇 Grape ripening under the magnifying glass: advances in mobile NIR spectroscopy for viticulture.

An interactive panel discussion with Dr. Anna Kicherer from the Julius Kühn Institute and Pascal Gauweiler from our SPR department on the topic oft »Mobile tire measurement«.

Through our involvement in the research project »DigiVine« (sponsored by BMEL) and in the development of FoodScannern, we can report on our experience in determining the degree of freshness and ripeness with the help of various sensors and AI, e.g. to determine the optimal harvest time by the winegrower or in the ripeness and quality assessments of the winegrowers' cooperative areas.

✅ Participation: The talks take place in Zoom and anyone interested can take part free of charge. Registration via the blue button on the right.

The International Measurement Confederation_IMEKO, is a global organisation dedicated to the advancement of measurement science and technology. Through conferences, symposia and publications, IMEKO provides a platform for researchers and practitioners to share their latest findings and innovations, contributing to the advancement of measurement science and its applications in various fields.

Host 2024: Physikalisch-Technische Bundesanstalt (PTB) / National Metrology Institute.

The Department Visual Inspection Systems (SPR) is involved with the following presentation:

👉 Simultaneous measurement of surface geometry and film thickness distribution by retroreflex ellipsometry.

Referent: Chia-Wei Chen

The VISION 2024 trade fair in Stuttgart is the international trade fair for industrial image processing and identification technologies. This time, we are participating in the VDMA Technology Days and presenting the Karlsruhe Center for Material Signatures-KCM. This is a merger of the Karlsruhe Institute of Technology (KIT) and the Inspection and Optronic Systems business unit of the Fraunhofer IOSB. Every two years, the KCM organizes the Image Processing Forum and the OCM Conference. 
Visit us in Stuttgart in hall 8/booth.

Your contact person: Henning Schulte

Already for the 17th time representatives from science and industry meet to discuss current waste management and waste technology topics. The current conference topics and the quality of the lectures have made Recy & DepoTech known far beyond the Austrian borders. 

We are represented with two presentations:

👉 assigned to the topic block »Artificial intelligence in preparation & sorting«:

Towards a Comprehensive Recycling Chain: Advancing Road Debris Management through Innovative Sensor-Based Detection and Thermo-Chemical Treatment Techniques

Reverent: Paul Bäcker on 15/11/2024 from 10:00-10:20 a.m.

👉 assigned to the topic block »Artificial intelligence in sorting & recycling«:

Implementation and evaluation of a real-time capable approach to sensor-based sorting using CNNs

 Reverent: Felix Lehnerer on 15/11/2024 from 11:00-11:20 a.m.

Image processing plays a key role in many areas of technology for fast and contactless data acquisition. For example, in the quality assurance of industrial production processes, in robotics and for driver assistance, image processing systems have gained an indispensable place. This development is supported by the availability of high-quality and inexpensive sensor systems, by the increase in the performance of computer systems, and by advances in the methodology for evaluating image data.

The "Image Processing forum" aims to report on current trends towards solutions in image processing and to contribute to the professional exchange between the participants.

Image Processing for Surfaces - Advances in technology not only enable ever higher inspection speeds and more compact systems, but also, in addition to the evaluation of two-dimensionally recorded textures, the complete recording of complex 3D surface structures down to the nanometer range. Furthermore, the rapid evaluation of colored, patterned, transparent, highly reflective or mirroring surfaces is possible. In the seminar, different inspection and analysis methods will be presented for this purpose.

We are involved with the following contributions:

Image acquisition in surface inspection

Thomas Längle, at November 27, 2024

Lightfield technologies

Christian Kludt, at November 27, 2024

Theory and methods of color measurement

Robin Gruna, at November 27, 2024

Internship station with:

Inspection of transparent materials

Christian Kludt, at November 28, 2024 on 09:00 am to 03:30 pm