Göttingen 2025 – wissenschaftliches Programm

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

P: Fachverband Plasmaphysik

P 6: Atmospheric Plasmas and their Applications II

P 6.4: Vortrag

Dienstag, 1. April 2025, 12:00–12:15, ZHG006

Open-source tools for interactive preselection and analysis of large image datasets — Philipp Mattern¹, Richard Krieg², Hans Höft¹, •Torsten Gerling¹, and Markus M. Becker¹ — ¹Leibniz Institute for Plasma Science and Technology (INP), Greifswald, Germany — ²University of Greifswald, Greifswald, Germany

To gain a holistic understanding of complex phenomena in plasma processes, it is often necessary to combine several high-resolution diagnostics and extensive parameter variations. This results in large data sets that are difficult to access using conventional methods and analysis tools. This contribution introduces two open-source tools developed at INP Greifswald for efficient data handling and image analysis: WOLKE and BLITZ. BLITZ enables rapid loading, visualization, and statistical evaluation of large image collections---handling more than 20,000 images (exceeding 20 GB) in under a minute---without requiring specialized hardware. Its matrix-based approach allows swift calculation of key parameters even for massive datasets. WOLKE provides a web-oriented layout and filtering framework for interactively preselecting image data based on user-defined criteria (e.g., mean, entropy, sharpness, operation parameter combinations, timestamps, EXIF information or any pre-calculated value). Filtered subsets identified in WOLKE can be seamlessly examined and further analyzed within BLITZ. This combination creates a highly adaptable workflow for data exploration, evaluation, and presentation, effectively responding to evolving research demands in plasma physics and beyond.

Keywords: Big Data; Image Analysis; Interactive Data Exploration; Rapid Statistical Evaluation