Measuring position of micro-objects by machine learning algorithms

Active: 
yes

You will be developing algorithms for a sensor measuring positions of micro-objects in 3D. The sensor mainly consists of an image sensor capturing diffraction patterns encoding positions of some micro-objects. We have algorithms which can extract the positions from the diffraction patterns, but they are slow. We have some ideas how to make the algorithms faster by implementing them on a graphical card (GPU) and/or by approximating them by methods from machine learning. Your goal will be to explore these possibilities.

This topic will fit a student who is more into coding and applied mathematics as the work is more on the software side.

Contact person: 
Martin Gurtner
Image: Lab setup for measuring the position and levitation height of a particle using a CMOS chip