Tackling fluffy clouds: robust field boundary delineation across global agricultural landscapes with Sentinel-1 and Sentinel-2 Time Series

TL;DR

This paper introduces a deep learning model that accurately delineates agricultural field boundaries from satellite time series data, effectively handling cloud contamination and reducing manual data curation efforts.

Contribution

The paper presents PTAViT3D and PTAViT3D-CA, novel 3D Vision Transformer architectures that fuse Sentinel-1 and Sentinel-2 data for robust, cloud-resilient field boundary delineation across global landscapes.

Findings

Achieves state-of-the-art boundary delineation accuracy.

Demonstrates high robustness in cloud-affected imagery.

Shows excellent transferability across diverse datasets.

Abstract

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.