BLITZ treats images as structured data.
A high-performance, matrix-based image viewer designed for efficiently exploring both massive image datasets and single-image analysis workflows.
Download the latest release for Windows and Ubuntu
No installation required — simply download and run the executable.
BLITZ (Bulk Loading and Interactive Time series Zonal analysis) is a high-performance, matrix-based image exploration and analysis tool designed for efficiently managing both massive datasets and single-image analysis.
It was originally developed and initially implemented by Philipp Mattern during his time at INP Greifswald.
It is actively maintained and further developed as part of his independent engineering work at M.E.S.S. – Mattern Engineering & Software Solutions.
Version 2.0 introduces a fully refactored architecture with improved performance, stability, and maintainability.
BLITZ is the interactive viewer in the WETTER framework: Raw Data → DAMPF → KEIM → WOLKE → BLITZ. For the full pipeline, ecosystem overview, and links to all modules, see:
WETTER Framework — wetter.mess.engineering
DPG Symposium presentation (architecture and BLITZ–WOLKE integration):
📄 BLITZ_WOLKE_DPG25V2_Compact.pdf
- High-Performance Data Handling: Efficiently processes very large datasets (e.g. loading, scaling, and converting ~21,000 RGB images (~2.5 GB raw data) into ~6.2 GB of grayscale matrix data in ~30 s on a standard gaming laptop).
- Easy Data Handling: Drag-and-drop support for image, video, and NumPy matrix (*.npy) formats.
- Easy to Use: Automatic resource management for small and large datasets.
- User-Friendly Interface: Intuitive GUI with mouse-based navigation and shortcuts.
- Advanced Image Processing: Matrix-based processing with fast, Numba-accelerated statistics.
- Live View: Support for real USB cameras and simulated data streams.
- Built on Python: Using Qt and PyQtGraph for high performance and flexibility.
(Click if animation is not playing)
- Full Documentation Index
- Quick Start Guide
- Features & Tabs Explained
- Missing & Planned Features
- Optimization Report
- Data Sources & Build Variants
Run BLITZ in a browser via Docker. See: docker/README.md
To compile and develop locally:
-
Clone the repository:
$ git clone https://github.com/pimav/BLITZ.git $ cd BLITZ -
Set up a virtual environment and install dependencies with uv:
$ pip install uv $ uv sync $ uv run python -m blitz
Early development of BLITZ was supported by Richard Krieg (student assistant) until v1.3.0 / January 2025, including refactoring, bug fixing, and feature development during the INP-funded project phase.
- Example Dataset: KinPen Science Example Set
- Explore more datasets or contribute your own on INPTDAT.
BLITZ is licensed under the GNU General Public License v3.0.