@inproceedings{c816b6f577e349048e4d3e622754765a,
title = "Leveraging convolutional neural networks for semantic segmentation of global floods with PlanetScope imagery",
abstract = "Constellations of small satellites are able to achieve the coverage, revisit rate, and spatial resolution needed to generate high resolution global maps of flood waters. The PlanetScope constellation achieves daily coverage at 3-4 m resolution; however, the limited spectral channels available as well as radiometric challenges inherent to large constellations of small satellites create a challenge to accurate global flood mapping. Here, we present a convolutional neural network (CNN) which combines PlanetScope imagery with information about multidecadal water dynamics from the Global Surface Water dataset to create a globally applicable water segmentation model. We also compare this model to locally-trained Random Forest models, which are commonly used for flood mapping. The CNN achieves a mean intersection over union score of 58.1% ± 9.84% across 7 biomes when evaluated on flooded areas outside of permanent water bodies, and 70.3% ± 11.6% when evaluated on all water pixels.",
keywords = "Random Forest, Sentinel-1, convolutional neural network, flood mapping, image segmentation, remote sensing",
author = "Leach, {Nicholas R.} and Philip Popien and Goodman, {Maxwell C.} and Beth Tellman",
note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2022 ; Conference date: 17-07-2022 Through 22-07-2022",
year = "2022",
doi = "10.1109/IGARSS46834.2022.9884272",
language = "English (US)",
series = "International Geoscience and Remote Sensing Symposium (IGARSS)",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "314--317",
booktitle = "IGARSS 2022 - 2022 IEEE International Geoscience and Remote Sensing Symposium",
}