Use of AI methods for noise reduction investigated.

As part of his bachelor’s thesis at Ernst Abbe University of Applied Sciences in Jena, Leonhard Knipfelberg investigated artificial intelligence (AI) methods for improving image quality in optical coherence tomography (OCT). The aim was to compare methods that contribute to increasing the signal-to-noise ratio and thus enable clearer images.

Taking into account image quality, computational effort, and stability, tests were carried out on classic filtering methods and AI-supported approaches on specially recorded OCT images of plants.

The results show that each of the methods has specific strengths. However, these are subject to method-dependent limitations in image quality, processing time, or both. The work provides a valuable foundation for the open-source project OpenLabKI, which aims to make OCT images with noise reduction available.

We congratulate Leonhard Knipfelberg on his successful bachelor’s degree!