Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

DirPA: Addressing Prior Shift in Imbalanced Few-shot Crop-type Classification

Joana Reuss, Ekaterina Gikalo, Marco Körner

Abstract

Real-world agricultural monitoring is often hampered by severe class imbalance and high label acquisition costs, resulting in significant data scarcity. In few-shot learning (FSL) -- a framework specifically designed for data-scarce settings -- , training sets are often artificially balanced. However, this creates a disconnect from the long-tailed distributions observed in nature, leading to a distribution shift that undermines the model's ability to generalize to real-world agricultural tasks. We previously introduced Dirichlet Prior Augmentation (DirPA; Reuss et al., 2026a) to proactively mitigate the effects of such label distribution skews during model training. In this work, we extend the original study's geographical scope. Specifically, we evaluate this extended approach across multiple countries in the European Union (EU), moving beyond localized experiments to test the method's resilience across diverse agricultural environments. Our results demonstrate the effectiveness of DirPA across different geographical regions. We show that DirPA not only improves system robustness and stabilizes training under extreme long-tailed distributions, regardless of the target region, but also substantially improves individual class-specific performance by proactively simulating priors.

DirPA: Addressing Prior Shift in Imbalanced Few-shot Crop-type Classification

Abstract

Real-world agricultural monitoring is often hampered by severe class imbalance and high label acquisition costs, resulting in significant data scarcity. In few-shot learning (FSL) -- a framework specifically designed for data-scarce settings -- , training sets are often artificially balanced. However, this creates a disconnect from the long-tailed distributions observed in nature, leading to a distribution shift that undermines the model's ability to generalize to real-world agricultural tasks. We previously introduced Dirichlet Prior Augmentation (DirPA; Reuss et al., 2026a) to proactively mitigate the effects of such label distribution skews during model training. In this work, we extend the original study's geographical scope. Specifically, we evaluate this extended approach across multiple countries in the European Union (EU), moving beyond localized experiments to test the method's resilience across diverse agricultural environments. Our results demonstrate the effectiveness of DirPA across different geographical regions. We show that DirPA not only improves system robustness and stabilizes training under extreme long-tailed distributions, regardless of the target region, but also substantially improves individual class-specific performance by proactively simulating priors.
Paper Structure (29 sections, 6 equations, 10 figures, 28 tables, 1 algorithm)

This paper contains 29 sections, 6 equations, 10 figures, 28 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related work
  3. Few-shot learning paradigm
  4. Class imbalance
  5. FSL in remote sensing
  6. Prior shift in FSL
  7. Dirichlet priors and distribution augmentation
  8. Dataset
  9. Dataset split
  10. Class distribution
  11. Biogeographical regions
  12. Agricultural practices
  13. Fine-tuning split
  14. Methodology
  15. DirPA: Dirichlet prior augmentation
  16. ...and 14 more sections

Figures (10)

  • Figure 1: Visualization of a label (prior) distribution shift between train dataset and a long-tailed test dataset.
  • Figure 2: Pre-training and fine-tuning country split. Countries colored in blue are used for pre-training, whereas those colored in green are used for fine-tuning. The darker the coloring, the denser the labeled agricultural fields.
  • Figure 3: EuroCropsML 2.0 binned class distribution for Germany ( and ) and Austria, shown on a logarithmic scale, with the pasture meadow grassland grass class removed.
  • Figure 4: The number of annotated crop classes that are shared and distinct between the pre-training countries---Austria and Germany---and the respective fine-tuning country.
  • Figure 5: Biogeographical regions present in EuroCropsML 2.0, as defined by the EEA16:biogeographical.
  • ...and 5 more figures