# Invariant Features for Global Crop Type Classification

arXiv: https://arxiv.org/abs/2509.03497 · HTML: https://mlsanity.com/paper/2509.03497/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 70 references that are in the map.
- First posted on arXiv 3 Sep 2025.

## Builds on (most cited first, 19 of 19)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Learning Deep Features for Discriminative Localization](https://mlsanity.com/paper/1512.04150/) (2015)
- [EfficientNetV2: Smaller Models and Faster Training](https://mlsanity.com/paper/2104.00298/) (2021)
- [A Closer Look at Memorization in Deep Networks](https://mlsanity.com/paper/1706.05394/) (2017)
- [Do Wide and Deep Networks Learn the Same Things? Uncovering How Neural Network Representations Vary with Width and Depth](https://mlsanity.com/paper/2010.15327/) (2020)
- [Efficient Object Localization Using Convolutional Networks](https://mlsanity.com/paper/1411.4280/) (2014)
- [AlphaEarth Foundations: An embedding field model for accurate and efficient global mapping from sparse label data](https://mlsanity.com/paper/2507.22291/) (2025)
- [An Empirical Survey of Data Augmentation for Time Series Classification with Neural Networks](https://mlsanity.com/paper/2007.15951/) (2020)
- [Lightweight, Pre-trained Transformers for Remote Sensing Timeseries](https://mlsanity.com/paper/2304.14065/) (2023)
- [Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series](https://mlsanity.com/paper/1811.10166/) (2018)
- [Fields of The World: A Machine Learning Benchmark Dataset For Global Agricultural Field Boundary Segmentation](https://mlsanity.com/paper/2409.16252/) (2024)
- [Time Series Data Augmentation for Neural Networks by Time Warping with a Discriminative Teacher](https://mlsanity.com/paper/2004.08780/) (2020)
- [TimeMatch: Unsupervised Cross-Region Adaptation by Temporal Shift Estimation](https://mlsanity.com/paper/2111.02682/) (2021)
- [Unlocking large-scale crop field delineation in smallholder farming systems with transfer learning and weak supervision](https://mlsanity.com/paper/2201.04771/) (2022)
- [Early- and in-season crop type mapping without current-year ground truth: generating labels from historical information via a topology-based approach](https://mlsanity.com/paper/2110.10275/) (2021)
- [Two Shifts for Crop Mapping: Leveraging Aggregate Crop Statistics to Improve Satellite-based Maps in New Regions](https://mlsanity.com/paper/2109.01246/) (2021)
- [Combining GEDI and Sentinel-2 for wall-to-wall mapping of tall and short crops](https://mlsanity.com/paper/2109.06972/) (2021)
- [Global High Categorical Resolution Land Cover Mapping via Weak Supervision](https://mlsanity.com/paper/2406.00891/) (2024)

## Built on by (most cited first, 1 of 1)

- [SAR2Agri: Learning SAR Intensity Representations for Agricultural Monitoring](https://mlsanity.com/paper/2608.11142/) (2026)

## Abstract (arXiv)

Accurate global crop type mapping supports agricultural monitoring and food security, yet remains limited by the scarcity of labeled data in many regions. A key challenge is enabling models trained in one geography to generalize reliably to others despite shifts in climate, phenology, and spectral characteristics. In this work, we show that geographic transfer in crop classification is primarily governed by the ability to learn invariant structure in multispectral time series. To systematically study this, we introduce CropGlobe, a globally distributed benchmark dataset of 300,000 samples spanning eight countries and five continents, and define progressively harder transfer settings from cross-country to cross-hemisphere. Across all settings, we find that simple spectral-temporal representations outperform both handcrafted features and modern geospatial foundation model embeddings. We propose CropNet, a lightweight convolutional architecture that jointly convolves across spectral and temporal dimensions to learn invariant crop signatures. Despite its simplicity, CropNet consistently outperforms larger transformer-based and foundation-model approaches under geographic domain shift. To further improve robustness to geographic variation, we introduce augmentations that simulate shifts in crop phenology and reflectance. Combined with CropNet, this yields substantial gains under large domain shifts. Our results demonstrate that inductive bias toward joint spectral-temporal structure is more critical for transfer than model scale or pretraining, pointing toward a scalable and data-efficient paradigm for worldwide agricultural mapping. Data and code are available at https://github.com/x-ytong/CropNet/.