Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

HK: Fachverband Physik der Hadronen und Kerne

HK 44: Instrumentation IX

HK 44.6: Vortrag

Mittwoch, 20. März 2019, 18:00–18:15, HS 11

Particle-Track Reconstruction with Artificial Neural Networks — Lukas Bierwirth, Laura Fabbietti, Martin Jan Losekamm, Stephan Paul, and •Thomas Pöschl — Technische Universität München

Finding the parameters of a particle's track in a detector in real time is a resource-intensive pattern recognition task. Artificial neural networks are a promising approach to this problem because of their ability to self-learn complex features from training data while still achieving a short reconstruction time per event.

We develop a neural network to analyze the data of the RadMap Telescope, which will measure the radiation environment aboard the International Space Station. We compare the performance of the neural network to a classical algorithm based on the Hough transform using simulated data and measurements from a test campaign at Paul Scherrer Institute.