Quote of the day

A person who never made a mistake never tried anything new.

- Albert Einstein

Contact

Parameter-Efficient Adaptation of Earth Observation Foundation Models for Smallholder Crop Mapping

Open Agriculture

Parameter-Efficient Adaptation of Earth Observation Foundation Models for Smallholder Crop Mapping

Parameter-Efficient Adaptation of Earth Observation Foundation Models for Smallholder Crop Mapping

Description

Evaluate and adapt Earth observation foundation models for field-boundary and crop-type mapping in fragmented smallholder landscapes under severe label scarcity.

Details

Context and Problem Statement

Earth observation foundation models are largely pre-trained on large agricultural landscapes from the United States and Europe. Their transferability to fragmented smallholder settings, where fields may approach Sentinel-2 pixel resolution and labels are scarce, remains poorly characterized.

Research Question

Which parameter-efficient adaptation strategy provides the best trade-off among performance, computational cost, and annotation requirements when transferring an Earth observation foundation model to fragmented agricultural landscapes?

Proposed Approach

Compare linear probing, low-rank adapters, normalization-only fine-tuning, and continued self-supervised learning on local unlabeled imagery for both field-boundary segmentation and crop-type classification, with robustness analysis under spatial and temporal distribution shifts.

Expected Contribution

A reproducible evaluation protocol and practical adaptation guidelines for label-scarce regions.

Expected Prototype

A geospatial pipeline producing field-boundary and crop-type maps for a Moroccan pilot region with uncertainty estimates and GIS export.

Datasets

Sentinel-2, Harmonized Landsat Sentinel, Fields of The World, and GPS field labels.

Challenges

Spectral mixing, cloud cover, computational cost, uncertain ground truth, and potential misuse of generated maps.

Research Question

To what extent do Earth observation foundation models generalize to fragmented smallholder landscapes, and what minimal adaptation is required to recover operationally useful performance?

Innovation

The project directly investigates geographic generalization to underrepresented regions under an explicit labeling budget rather than assuming abundant labeled data.

Expected Deliverable

A validated crop map for a pilot region, a benchmark comparing adaptation strategies, and a conference paper in remote sensing or geospatial AI.

Technologies

Remote Sensing Foundation Models (Prithvi, SSL4EO) Parameter-Efficient Fine-Tuning (LoRA, adapters) Self-Supervised Learning Semantic Segmentation Google Earth Engine and Copernicus

Required Skills

  • Remote Sensing and Satellite Image Processing
  • Deep Learning and Transfer Learning
  • Geographic Information Systems
  • Distributed or GPU Computing

Datasets

  • Sentinel-2 / Copernicus Data Space
  • Fields of The World field-boundary benchmark
  • GPS field surveys in a pilot area

Morocco & Africa Relevance

Smallholder agriculture dominates many Moroccan and African regions but remains poorly mapped, limiting policy targeting, index insurance, and irrigated-area estimation.