This repository contains Alpine glacier crevasse information derived from aerial imagery. The crevasses were detected for the most recent available orthophotos for the following five regions: Stubai (2019/2020), Oetztal (2019/2020), Piz Palue (2022), Grossglockner (2022) and Ortler (2023). The crevasse detection is based on a multitask neural network (HED-Unet) able to simultaneously perform edge detection and semantic segmentation.

The dataset is available as binary GeoTIFF in the ETRS89 / LAEA Europe projection (EPSG:3035) with a pixel size of 0.2 m. Additionally, all files are available in the original, local projections of the aerial imagery (higher geolocation accuracy and higher spatial resolution for Piz Palue at 0.1 m).

The grey levels in the raster file indicate the location of the crevasse boundary:
0 = no crevasse
1 = crevasse boundary

For further information contact Dr. Celia A. Baumhoer: celia.baumhoer@dlr.de

References:
- Baumhoer, C., Leibrock, S., Zapf, C., Beer, W., Kuenzer, C., in review. Automated crevasse mapping for Alpine glaciers: A multitask deep neural network approach. International Journal of Applied Earth Observation and Geoinformation.
- Heidler, K., Mou, L., Baumhoer, C., Dietz, A., Zhu, X.X., 2022. HED-UNet: Combined Segmentation and Edge Detection for Monitoring the Antarctic Coastline. IEEE Trans. Geosci. Remote Sensing 60, 1–14. https://doi.org/10.1109/TGRS.2021.3064606
- Heidler, K., Mou, L., Baumhoer, C., Dietz, A., Zhu, X.X., 2021. Hed-Unet: A Multi-Scale Framework for Simultaneous Segmentation and Edge Detection, in: 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS. Presented at the 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, pp. 3037–3040. https://doi.org/10.1109/IGARSS47720.2021.9553585