Deep Learning model for predicting fracture features in cracked glass
In this project, we harness the power of deep learning models and neural networks to decipher the intricate patterns of cracked glass. The goal is to teach artificial intelligence (AI) to predict and understand the nuanced features of cracks in damaged glass.
Through the lens of sophisticated neural networks, you will delve into the realm of deep learning, where complex algorithms unravel the subtleties of experimental data for you. The heart of the approach lies in processing extensive sets of experimental data, meticulously collected from damaged glass scenarios. For that an existing deep learning model is to be extended and feeded with physical relationships so that it is able to backpropagate multiple geometric parameters which are not/or only determinable with time-costly effort.
Join us on this journey at the intersection of deep learning and material science, where we aim to unlock the secrets hidden within fractured glass, ultimately reshaping the landscape of damage prediction and prevention.
Additional information
| Capacity | Two IREP students |
| Project available for | Spring and Summer 2024 |
| Credits | 12 ECTS |
| Available via Remote | Yes |
| Project Supervisor | Isabell Ayvaz M.Sc. |
