Multi-Region Transfer Learning for Segmentation of Crop Field Boundaries in Satellite Images with Limited Labels

TL;DR

This paper introduces a multi-region transfer learning approach for segmenting crop field boundaries in satellite images, especially effective in regions with limited labeled data, outperforming existing methods and providing a public benchmark.

Contribution

The paper proposes a novel multi-region transfer learning method for crop boundary segmentation that works well with scarce labeled data and demonstrates significant performance improvements.

Findings

Outperforms existing segmentation methods in low-label scenarios

Multi-region transfer learning boosts model accuracy across architectures

Provides publicly available datasets and benchmarks for future research

Abstract

The goal of field boundary delineation is to predict the polygonal boundaries and interiors of individual crop fields in overhead remotely sensed images (e.g., from satellites or drones). Automatic delineation of field boundaries is a necessary task for many real-world use cases in agriculture, such as estimating cultivated area in a region or predicting end-of-season yield in a field. Field boundary delineation can be framed as an instance segmentation problem, but presents unique research challenges compared to traditional computer vision datasets used for instance segmentation. The practical applicability of previous work is also limited by the assumption that a sufficiently-large labeled dataset is available where field boundary delineation models will be applied, which is not the reality for most regions (especially under-resourced regions such as Sub-Saharan Africa). We present an approach for segmentation of crop field boundaries in satellite images in regions lacking labeled data that uses multi-region transfer learning to adapt model weights for the target region. We show that our approach outperforms existing methods and that multi-region transfer learning substantially boosts performance for multiple model architectures. Our implementation and datasets are publicly available to enable use of the approach by end-users and serve as a benchmark for future work.