Towards global flood mapping onboard low cost satellites with machine learning. Mateo-Garcia, G., Veitch-Michaelis, J., Smith, L., Oprea, S. V., Schumann, G., Gal, Y., Baydin, A. G., & Backes, D. Scientific Reports, 11(1):7249, December, 2021. ![Towards global flood mapping onboard low cost satellites with machine learning [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex Abstract Spaceborne Earth observation is a key technology for flood response, offering valuable information to decision makers on the ground. Very large constellations of small, nano satellites— ’CubeSats’ are a promising solution to reduce revisit time in disaster areas from days to hours. However, data transmission to ground receivers is limited by constraints on power and bandwidth of CubeSats. Onboard processing offers a solution to decrease the amount of data to transmit by reducing large sensor images to smaller data products. The ESA’s recent PhiSat-1 mission aims to facilitate the demonstration of this concept, providing the hardware capability to perform onboard processing by including a power-constrained machine learning accelerator and the software to run custom applications. This work demonstrates a flood segmentation algorithm that produces flood masks to be transmitted instead of the raw images, while running efficiently on the accelerator aboard the PhiSat-1. Our models are trained on WorldFloods : a newly compiled dataset of 119 globally verified flooding events from disaster response organizations, which we make available in a common format. We test the system on independent locations, demonstrating that it produces fast and accurate segmentation masks on the hardware accelerator, acting as a proof of concept for this approach.
@article{mateo-garcia_towards_2021,
title = {Towards global flood mapping onboard low cost satellites with machine learning},
volume = {11},
issn = {2045-2322},
url = {http://www.nature.com/articles/s41598-021-86650-z},
doi = {10.1038/s41598-021-86650-z},
abstract = {Abstract
Spaceborne Earth observation is a key technology for flood response, offering valuable information to decision makers on the ground. Very large constellations of small, nano satellites— ’CubeSats’ are a promising solution to reduce revisit time in disaster areas from days to hours. However, data transmission to ground receivers is limited by constraints on power and bandwidth of CubeSats. Onboard processing offers a solution to decrease the amount of data to transmit by reducing large sensor images to smaller data products. The ESA’s recent PhiSat-1 mission aims to facilitate the demonstration of this concept, providing the hardware capability to perform onboard processing by including a power-constrained machine learning accelerator and the software to run custom applications. This work demonstrates a flood segmentation algorithm that produces flood masks to be transmitted instead of the raw images, while running efficiently on the accelerator aboard the PhiSat-1. Our models are trained on
WorldFloods
: a newly compiled dataset of 119 globally verified flooding events from disaster response organizations, which we make available in a common format. We test the system on independent locations, demonstrating that it produces fast and accurate segmentation masks on the hardware accelerator, acting as a proof of concept for this approach.},
language = {en},
number = {1},
urldate = {2021-12-11},
journal = {Scientific Reports},
author = {Mateo-Garcia, Gonzalo and Veitch-Michaelis, Joshua and Smith, Lewis and Oprea, Silviu Vlad and Schumann, Guy and Gal, Yarin and Baydin, Atılım Güneş and Backes, Dietmar},
month = dec,
year = {2021},
keywords = {AI4EM, AI4EarthMonitoring},
pages = {7249},
}
Downloads: 0
{"_id":"pqXBsiSWPKmnHmgc9","bibbaseid":"mateogarcia-veitchmichaelis-smith-oprea-schumann-gal-baydin-backes-towardsglobalfloodmappingonboardlowcostsatelliteswithmachinelearning-2021","author_short":["Mateo-Garcia, G.","Veitch-Michaelis, J.","Smith, L.","Oprea, S. V.","Schumann, G.","Gal, Y.","Baydin, A. G.","Backes, D."],"bibdata":{"bibtype":"article","type":"article","title":"Towards global flood mapping onboard low cost satellites with machine learning","volume":"11","issn":"2045-2322","url":"http://www.nature.com/articles/s41598-021-86650-z","doi":"10.1038/s41598-021-86650-z","abstract":"Abstract Spaceborne Earth observation is a key technology for flood response, offering valuable information to decision makers on the ground. Very large constellations of small, nano satellites— ’CubeSats’ are a promising solution to reduce revisit time in disaster areas from days to hours. However, data transmission to ground receivers is limited by constraints on power and bandwidth of CubeSats. Onboard processing offers a solution to decrease the amount of data to transmit by reducing large sensor images to smaller data products. The ESA’s recent PhiSat-1 mission aims to facilitate the demonstration of this concept, providing the hardware capability to perform onboard processing by including a power-constrained machine learning accelerator and the software to run custom applications. This work demonstrates a flood segmentation algorithm that produces flood masks to be transmitted instead of the raw images, while running efficiently on the accelerator aboard the PhiSat-1. Our models are trained on WorldFloods : a newly compiled dataset of 119 globally verified flooding events from disaster response organizations, which we make available in a common format. We test the system on independent locations, demonstrating that it produces fast and accurate segmentation masks on the hardware accelerator, acting as a proof of concept for this approach.","language":"en","number":"1","urldate":"2021-12-11","journal":"Scientific Reports","author":[{"propositions":[],"lastnames":["Mateo-Garcia"],"firstnames":["Gonzalo"],"suffixes":[]},{"propositions":[],"lastnames":["Veitch-Michaelis"],"firstnames":["Joshua"],"suffixes":[]},{"propositions":[],"lastnames":["Smith"],"firstnames":["Lewis"],"suffixes":[]},{"propositions":[],"lastnames":["Oprea"],"firstnames":["Silviu","Vlad"],"suffixes":[]},{"propositions":[],"lastnames":["Schumann"],"firstnames":["Guy"],"suffixes":[]},{"propositions":[],"lastnames":["Gal"],"firstnames":["Yarin"],"suffixes":[]},{"propositions":[],"lastnames":["Baydin"],"firstnames":["Atılım","Güneş"],"suffixes":[]},{"propositions":[],"lastnames":["Backes"],"firstnames":["Dietmar"],"suffixes":[]}],"month":"December","year":"2021","keywords":"AI4EM, AI4EarthMonitoring","pages":"7249","bibtex":"@article{mateo-garcia_towards_2021,\n\ttitle = {Towards global flood mapping onboard low cost satellites with machine learning},\n\tvolume = {11},\n\tissn = {2045-2322},\n\turl = {http://www.nature.com/articles/s41598-021-86650-z},\n\tdoi = {10.1038/s41598-021-86650-z},\n\tabstract = {Abstract\n \n Spaceborne Earth observation is a key technology for flood response, offering valuable information to decision makers on the ground. Very large constellations of small, nano satellites— ’CubeSats’ are a promising solution to reduce revisit time in disaster areas from days to hours. However, data transmission to ground receivers is limited by constraints on power and bandwidth of CubeSats. Onboard processing offers a solution to decrease the amount of data to transmit by reducing large sensor images to smaller data products. The ESA’s recent PhiSat-1 mission aims to facilitate the demonstration of this concept, providing the hardware capability to perform onboard processing by including a power-constrained machine learning accelerator and the software to run custom applications. This work demonstrates a flood segmentation algorithm that produces flood masks to be transmitted instead of the raw images, while running efficiently on the accelerator aboard the PhiSat-1. Our models are trained on\n WorldFloods\n : a newly compiled dataset of 119 globally verified flooding events from disaster response organizations, which we make available in a common format. We test the system on independent locations, demonstrating that it produces fast and accurate segmentation masks on the hardware accelerator, acting as a proof of concept for this approach.},\n\tlanguage = {en},\n\tnumber = {1},\n\turldate = {2021-12-11},\n\tjournal = {Scientific Reports},\n\tauthor = {Mateo-Garcia, Gonzalo and Veitch-Michaelis, Joshua and Smith, Lewis and Oprea, Silviu Vlad and Schumann, Guy and Gal, Yarin and Baydin, Atılım Güneş and Backes, Dietmar},\n\tmonth = dec,\n\tyear = {2021},\n\tkeywords = {AI4EM, AI4EarthMonitoring},\n\tpages = {7249},\n}\n\n","author_short":["Mateo-Garcia, G.","Veitch-Michaelis, J.","Smith, L.","Oprea, S. V.","Schumann, G.","Gal, Y.","Baydin, A. G.","Backes, D."],"key":"mateo-garcia_towards_2021","id":"mateo-garcia_towards_2021","bibbaseid":"mateogarcia-veitchmichaelis-smith-oprea-schumann-gal-baydin-backes-towardsglobalfloodmappingonboardlowcostsatelliteswithmachinelearning-2021","role":"author","urls":{"Paper":"http://www.nature.com/articles/s41598-021-86650-z"},"keyword":["AI4EM","AI4EarthMonitoring"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"article","biburl":"https://bibbase.org/zotero/iagogv","dataSources":["ovn29uG6Mbp3JWCRR"],"keywords":["ai4em","ai4earthmonitoring"],"search_terms":["towards","global","flood","mapping","onboard","low","cost","satellites","machine","learning","mateo-garcia","veitch-michaelis","smith","oprea","schumann","gal","baydin","backes"],"title":"Towards global flood mapping onboard low cost satellites with machine learning","year":2021}