Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

PRUE: A Practical Recipe for Field Boundary Segmentation at Scale

Gedeon Muhawenayo, Caleb Robinson, Subash Khanal, Zhanpei Fang, Isaac Corley, Alexander Wollam, Tianyi Gao, Leonard Strnad, Ryan Avery, Lyndon Estes, Ana M. Tárano, Nathan Jacobs, Hannah Kerner

Abstract

Large-scale maps of field boundaries are essential for agricultural monitoring tasks. Existing deep learning approaches for satellite-based field mapping are sensitive to illumination, spatial scale, and changes in geographic location. We conduct the first systematic evaluation of segmentation and geospatial foundation models (GFMs) for global field boundary delineation using the Fields of The World (FTW) benchmark. We evaluate 18 models under unified experimental settings, showing that a U-Net semantic segmentation model outperforms instance-based and GFM alternatives on a suite of performance and deployment metrics. We propose a new segmentation approach that combines a U-Net backbone, composite loss functions, and targeted data augmentations to enhance performance and robustness under real-world conditions. Our model achieves a 76\% IoU and 47\% object-F1 on FTW, an increase of 6\% and 9\% over the previous baseline. Our approach provides a practical framework for reliable, scalable, and reproducible field boundary delineation across model design, training, and inference. We release all models and model-derived field boundary datasets for five countries.

PRUE: A Practical Recipe for Field Boundary Segmentation at Scale

Abstract

Large-scale maps of field boundaries are essential for agricultural monitoring tasks. Existing deep learning approaches for satellite-based field mapping are sensitive to illumination, spatial scale, and changes in geographic location. We conduct the first systematic evaluation of segmentation and geospatial foundation models (GFMs) for global field boundary delineation using the Fields of The World (FTW) benchmark. We evaluate 18 models under unified experimental settings, showing that a U-Net semantic segmentation model outperforms instance-based and GFM alternatives on a suite of performance and deployment metrics. We propose a new segmentation approach that combines a U-Net backbone, composite loss functions, and targeted data augmentations to enhance performance and robustness under real-world conditions. Our model achieves a 76\% IoU and 47\% object-F1 on FTW, an increase of 6\% and 9\% over the previous baseline. Our approach provides a practical framework for reliable, scalable, and reproducible field boundary delineation across model design, training, and inference. We release all models and model-derived field boundary datasets for five countries.

Paper Structure

This paper contains 60 sections, 3 equations, 10 figures, 6 tables, 3 algorithms.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Semantic segmentation with post-processing.
  4. Instance and panoptic segmentation.
  5. Universal segmentation.
  6. Geospatial Foundation Models (GFMs).
  7. Methods
  8. Dataset
  9. Model architecture search
  10. Semantic segmentation baselines.
  11. Instance and panoptic segmentation baselines.
  12. Geospatial foundation models (GFMs).
  13. Model design space exploration
  14. Architectures and backbones.
  15. Loss functions.
  16. ...and 45 more sections

Figures (10)

  • Figure 1: Planting Season Heuristic Algorithm
  • Figure 2: Example visualization of predictions over Illinois, USA (top, MGRS tile 16TDL) and Mato Grosso, Brazil (bottom, MGRS tile 20LRQ) with the FTW Baseline, Terramind and PRUE. The FTW Baseline and Terramind models show strong sensitivity to scene characteristics, producing discontinuous and noisy field boundaries. The PRUE model achieves stable boundaries across regions and imaging conditions.
  • Figure 2: Harvest Season Heuristic Algorithm
  • Figure 3: Spatial consistency. Four overlapping crops from each image corner are independently segmented. The consistency mask shows pixel-level agreement, with yellow indicating unanimity across all four predictions, and purple disagreement. This metric quantifies grid artifact resistance for large-scale field delineation.
  • Figure 3: Greedy Scene Selection Algorithm
  • ...and 5 more figures