BlessemFlood21
The BlessemFlood21 dataset was created by the Fraunhofer Institute for Image Processing and other institutions, focusing on high‑resolution RGB images of non‑coastal flood scenes. It contains 4,623 images, each 512×512 pixels, captured by a drone after the 2021 Erftstadt‑Blessem flood event. Detailed water masks were generated using semi‑supervised human‑in‑the‑loop techniques, primarily for training and testing deep‑learning models to support flood detection and emergency response.
Dataset description and usage context
Dataset Overview
Dataset Title
BlessemFlood21: Advancing Flood Analysis with a High‑Resolution Georeferenced Dataset for Humanitarian Aid Support
Dataset Description
The BlessemFlood21 dataset aims to promote research on flood detection tools. It comprises high‑resolution, georeferenced RGB‑NIR images captured during the 2021 Erftstadt‑Blessem flood event, supplemented with detailed water masks obtained via semi‑supervised human‑in‑the‑loop techniques. The dataset is used to evaluate deep‑learning models for semantic segmentation, providing high‑resolution RGB data and a benchmark for RGB‑based flood detection algorithms.
Dataset Type
Image
Release Date
January 2024
Identifier
- Link: https://fordatis.fraunhofer.de/handle/fordatis/379
- DOI: http://dx.doi.org/10.24406/fordatis/323
Contributors
- Polushko, Vladyslav
- Weinmann, Andreas
- März, Thomas
- Rauhut, Markus
- Jenal, Alexander
- Weber, Immanuel
- Rösch, Ronald
- Hatic, Damjan
- Bongartz, Jens
Keywords
- Remote Sensing
- Humanitarian Aid Support
- Deep Learning
- Water Detection Dataset
Files
- Filename: ortho_blessem_20210718_mask_v.01.24.tif
- Size: 21.55 MB
- Format: TIFF
- Filename: ortho_blessem_20210718_rgb_v.01.24.tif
- Size: 3.12 GB
- Format: TIFF
License
The dataset is available under the Creative Commons Attribution‑NonCommercial‑ShareAlike 4.0 International License.
Pair the dataset with AI analysis and content workflows.
Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.