
Temporal MMFlood Dataset This repository contains the Temporal MMFlood dataset proposed in "Portraying the Need for Temporal Data in Flood Detection via Sentinel-1", where the original MMFlood dataset is extended from single-date to multi-date. Code and access to the paper can be accessed via the following link: https://github.com/xavibou/sar_flood_detection. "The MMFlood dataset is well distributed and representative of a large set of flood events. To cope with the lack of temporal information, leading to the ill-posed definition of flood, we extend the MMFlood dataset by adding the Sentinel-1 images one year before the event and one month after. To do so, we fetch the imagery from the same relative orbit number of the original selection of MMFlood, and process the Sentinel-1 GRD imagery to obtain the same footprint and radiometry. Due to the improved geolocation accuracy of our pipeline and different DEM, there are some negligible residuals between our dataset and the original MMFlood Sentinel-1 crop. However, as discussed in Section 2, the labels of MMFlood are not suitable for a fair evaluation of flood detection methods, and we discourage their use. The proposed multi-date dataset can be used for multi-date flood mapping methods but future work is necessary to include validation labels." The dataset is provided as a set of .zip file, where each one consists in an individual scene of the original MMFlood dataset. Timeseries are provided only for the test set EMSR events of MMFlood (except EMSR162). [1] X. Bou, T. Ehret, R. Grompone and J. Anger. "Portraying the Need for Temporal Data in Flood Detection via Sentinel-1". IGARSS, 2024. [2] F. Montello, E. Arnaudo and C. Rossi. "MMFlood: A Multimodal Dataset for Flood Delineation From Satellite Imagery," in IEEE Access, vol. 10, pp. 96774-96787, 2022, doi: 10.1109/ACCESS.2022.3205419.
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 0 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Average | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |
