@incollection{fdi:010086243,
  title = {{S}ub-pixel optical satellite image registration for ground deformation using deep learning},
  author = {{M}ontagnon, {T}. and {H}ollingsworth, {J}. and {P}athier, {E}. and {M}archandon, {M}. and {D}alla {M}ura, {M}. and {G}iffard-{R}oisin, {S}ophie},
  editor = {},
  language = {{ENG}},
  abstract = {{P}recise estimation of ground displacement maps at regional scales from optical satellite imaging is fundamental for the comprehension of natural disasters, such as earthquakes. {C}urrent methods make use of correlation techniques between two acquisitions in order to retrieve a fractional pixel shift. {Y}et, differences in local lighting conditions between the two acquisitions can lead to high image differences which will bias the estimation of the displacement, and data-driven methods could have the ability to overcome these errors. {F}rom the generation of a realistic simulated database based on {L}andsat-8 satellite pairs of images with added simulated shifts, we developed a {C}onvolutional {N}eural {N}etwork ({CNN}) able to retrieve a sub-pixel displacement.},
  keywords = {},
  booktitle = {{IEEE} {I}nternational {C}onference on {I}mage {P}rocessing ({IEEE} {ICIP})},
  numero = {},
  pages = {2716--2720},
  address = {{P}iscataway},
  publisher = {{IEEE}},
  series = {},
  year = {2022},
  DOI = {10.1109/{ICIP}46576.2022.9897214},
  ISBN = {978-1-6654-9621-6},
  URL = {https://www.documentation.ird.fr/hor/fdi:010086243},
}