Generalized Classification of Satellite Image Time Series with Thermal Positional Encoding - Archive ouverte HAL Access content directly
Conference Papers Year :

Generalized Classification of Satellite Image Time Series with Thermal Positional Encoding

Joachim Nyborg
  • Function : Author
  • PersonId : 1211554
Ira Assent
  • Function : Author

Abstract

Large-scale crop type classification is a task at the core of remote sensing efforts with applications of both economic and ecological importance. Current state-of-the-art deep learning methods are based on self-attention and use satellite image time series (SITS) to discriminate crop types based on their unique growth patterns. However, existing methods generalize poorly to regions not seen during training mainly due to not being robust to temporal shifts of the growing season caused by variations in climate. To this end, we propose Thermal Positional Encoding (TPE) for attentionbased crop classifiers. Unlike previous positional encoding based on calendar time (e.g. day-of-year), TPE is based on thermal time, which is obtained by accumulating daily average temperatures over the growing season. Since crop growth is directly related to thermal time, but not calendar time, TPE addresses the temporal shifts between different regions to improve generalization. We propose multiple TPE strategies, including learnable methods, to further improve results compared to the common fixed positional encodings. We demonstrate our approach on a crop classification task across four different European regions, where we obtain state-of-the-art generalization results. Our source code is available at https://github.com/jnyborg/tpe.
Fichier principal
Vignette du fichier
Nyborg_Generalized_Classification_of_Satellite_Image_Time_Series_With_Thermal_Positional_CVPRW_2022_paper.pdf (1010.38 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03922196 , version 1 (04-01-2023)

Identifiers

  • HAL Id : hal-03922196 , version 1

Cite

Joachim Nyborg, Charlotte Pelletier, Ira Assent. Generalized Classification of Satellite Image Time Series with Thermal Positional Encoding. Conference, Jun 2022, New Orleans, United States. ⟨hal-03922196⟩
18 View
1 Download

Share

Gmail Facebook Twitter LinkedIn More