Guardado en:
Detalles Bibliográficos
Autores principales: Diakogiannis, Foivos I., Zhou, Zheng-Shu, Wang, Jeff, Mata, Gonzalo, Henry, Dave, Lawes, Roger, Parker, Amy, Caccetta, Peter, Ibata, Rodrigo, Hlinka, Ondrej, Richetti, Jonathan, Batchelor, Kathryn, Herrmann, Chris, Toovey, Andrew, Taylor, John
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2409.13568
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909660410281984
author Diakogiannis, Foivos I.
Zhou, Zheng-Shu
Wang, Jeff
Mata, Gonzalo
Henry, Dave
Lawes, Roger
Parker, Amy
Caccetta, Peter
Ibata, Rodrigo
Hlinka, Ondrej
Richetti, Jonathan
Batchelor, Kathryn
Herrmann, Chris
Toovey, Andrew
Taylor, John
author_facet Diakogiannis, Foivos I.
Zhou, Zheng-Shu
Wang, Jeff
Mata, Gonzalo
Henry, Dave
Lawes, Roger
Parker, Amy
Caccetta, Peter
Ibata, Rodrigo
Hlinka, Ondrej
Richetti, Jonathan
Batchelor, Kathryn
Herrmann, Chris
Toovey, Andrew
Taylor, John
contents Accurate delineation of agricultural field boundaries is essential for effective crop monitoring and resource management. However, competing methodologies often face significant challenges, particularly in their reliance on extensive manual efforts for cloud-free data curation and limited adaptability to diverse global conditions. In this paper, we introduce PTAViT3D, a deep learning architecture specifically designed for processing three-dimensional time series of satellite imagery from either Sentinel-1 (S1) or Sentinel-2 (S2). Additionally, we present PTAViT3D-CA, an extension of the PTAViT3D model incorporating cross-attention mechanisms to fuse S1 and S2 datasets, enhancing robustness in cloud-contaminated scenarios. The proposed methods leverage spatio-temporal correlations through a memory-efficient 3D Vision Transformer architecture, facilitating accurate boundary delineation directly from raw, cloud-contaminated imagery. We comprehensively validate our models through extensive testing on various datasets, including Australia's ePaddocks - CSIRO's national agricultural field boundary product - alongside public benchmarks Fields-of-the-World, PASTIS, and AI4SmallFarms. Our results consistently demonstrate state-of-the-art performance, highlighting excellent global transferability and robustness. Crucially, our approach significantly simplifies data preparation workflows by reliably processing cloud-affected imagery, thereby offering strong adaptability across diverse agricultural environments. Our code and models are publicly available at https://github.com/feevos/tfcl.
format Preprint
id arxiv_https___arxiv_org_abs_2409_13568
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Tackling fluffy clouds: robust field boundary delineation across global agricultural landscapes with Sentinel-1 and Sentinel-2 Time Series
Diakogiannis, Foivos I.
Zhou, Zheng-Shu
Wang, Jeff
Mata, Gonzalo
Henry, Dave
Lawes, Roger
Parker, Amy
Caccetta, Peter
Ibata, Rodrigo
Hlinka, Ondrej
Richetti, Jonathan
Batchelor, Kathryn
Herrmann, Chris
Toovey, Andrew
Taylor, John
Computer Vision and Pattern Recognition
Accurate delineation of agricultural field boundaries is essential for effective crop monitoring and resource management. However, competing methodologies often face significant challenges, particularly in their reliance on extensive manual efforts for cloud-free data curation and limited adaptability to diverse global conditions. In this paper, we introduce PTAViT3D, a deep learning architecture specifically designed for processing three-dimensional time series of satellite imagery from either Sentinel-1 (S1) or Sentinel-2 (S2). Additionally, we present PTAViT3D-CA, an extension of the PTAViT3D model incorporating cross-attention mechanisms to fuse S1 and S2 datasets, enhancing robustness in cloud-contaminated scenarios. The proposed methods leverage spatio-temporal correlations through a memory-efficient 3D Vision Transformer architecture, facilitating accurate boundary delineation directly from raw, cloud-contaminated imagery. We comprehensively validate our models through extensive testing on various datasets, including Australia's ePaddocks - CSIRO's national agricultural field boundary product - alongside public benchmarks Fields-of-the-World, PASTIS, and AI4SmallFarms. Our results consistently demonstrate state-of-the-art performance, highlighting excellent global transferability and robustness. Crucially, our approach significantly simplifies data preparation workflows by reliably processing cloud-affected imagery, thereby offering strong adaptability across diverse agricultural environments. Our code and models are publicly available at https://github.com/feevos/tfcl.
title Tackling fluffy clouds: robust field boundary delineation across global agricultural landscapes with Sentinel-1 and Sentinel-2 Time Series
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2409.13568