Master's Thesis "Classification of Road Surfaces Using AI-Based Image Analysis"

We are committed to creating an inspiring working environment in which all employees can develop. To this end, we promote and value the uniqueness of each person and create a culture of respect, openness and appreciation in which everyone is welcome:

• Teamwork at eye level: an open-minded team in which you can contribute your own ideas. New Work, diversity and inclusion as well as cooperative teamwork are central components of our culture.
• Theory meets practice: The opportunity to combine your knowledge from your studies with practical experience at the interface between business and science.
• Flex work: Your studies have priority - we offer you flexible working hours that adapt to your everyday study routine.
• Students@UMSICHT: Comprehensive onboarding, good support and networking opportunities as well as versatile support for your career development through training and keynote speeches.
• Health & well-being: Health offers for a good balance in everyday life.
• You can find even more attractive benefits here: Fraunhofer UMSICHT as an employer

Remuneration according to the general works agreement for employing assistant staff.

Application:

Interested? Apply online now via our career portal and join the Fraunhofer UMSICHT team! We look forward to getting to know you. Please submit at least the following documents:

• Curriculum vitae
• Current certificate of enrollment
• Current transcript of records
• Bachelor’s degree certificate

• Currently pursuing a master’s degree in computer science, data science, physics, mechanical engineering, or a related field
• Programming skills in Python
• Basic knowledge of machine learning and image processing
• Ideally, some experience with CNNs/Vision Transformers and clustering algorithms, as well as basic knowledge of optical measurement technology and lighting technology
• Good English skills would be an advantage

The Fraunhofer-Gesellschaft (www.fraunhofer.com) is one of the world’s leading organizations for application-oriented research. 74 institutes develop pioneering technologies for our economy and society - more precisely: 30 000 people from technology, science, administration and IT. They know: Anyone who comes to Fraunhofer wants to and can make a difference. For themselves, for us and for the markets of today and tomorrow.

We want an earth that is worth living on. Our applied research in Circular Economy, Carbon Management, Green Hydrogen and Local Energy Systems makes concrete contributions to achieving the 17 Sustainable Development Goals of the United Nations. As an institute of the Fraunhofer-Gesellschaft, we show how to not only master challenges, but grow with them. We push boundaries, think ahead and shape the future with applied research. Diverse perspectives and approaches strengthen creativity and innovation in our scientific work.

Would you like to help us shape a sustainable world? We are looking for reinforcement!

Tire abrasion particles nowadays account for the majority of a vehicle’s harmful fine particulate matter emissions. The amount of abrasion generated depends not only on the vehicle and driving style but, above all, on the structure of the road surface. Fraunhofer UMSICHT has been modeling tire abrasion for years (TWM) - but one key component is missing: a reliable classification of road surfaces based on their abrasiveness. This would require large-scale surveys of roads, which is virtually impossible with conventional measurement techniques. This is where your work comes in. Our hypothesis: Photographic data collection and AI-supported image analysis make it possible to classify road surfaces on a large scale. With support from the Fraunhofer Institute for Industrial and Technological Mathematics (ITWM), you will investigate whether typical abrasion classes can be derived from surface images - thereby laying the foundation for a method that can later be used on a mobile measurement platform (REDAR).

Be part of change

• You will set up a laboratory measurement system consisting of camera technology and lighting and optimize it for the reproducible capture of surface structures
• You will create a training dataset consisting of surface images with known abrasiveness or abrasiveness to be determined as part of this work
• You will develop an AI-based image analysis method and cluster the surfaces according to characteristic wear features (in collaboration with the Fraunhofer Institute for Industrial Mathematics ITWM)