Machine Learning Approach for Multi Particle Tracking in Complex Plasmas

N. Dormagen, M. Klein (THM Univ. of Applied Sciences and Justus Liebig Univ., Germany), M.H. Thoma (Justus Liebig Univ., Germany), M. Schwarz (THM Univ. of Applied Sciences, Germany)

Detecting micron-sized particles is an essential task for the analysis of complex plasmas, because a large part of the analysis is based on the initially detected positions of the particles. Accordingly, high accuracy in particle detection is desirable. Previous studies have shown that machine learning algorithms have made great progress and outperformed classical approaches. This work presents an approach for tracking micron-sized particles in a dense cloud of particles in a dusty plasma at Plasmakristall-Experiment 4 using a U-Net. The U-net is a convolutional network architecture for fast and precise segmentation of images that was developed at the Computer Science Department of the University of Freiburg. We compare the neural network with a conventional method we used so far in terms of accuracy in analyzing artificial data. We also apply the network to data of the PK-4 experiment. The experimental data were generated under laboratory conditions.