# Deep learning on edge: extracting field boundaries from satellite images with a convolutional neural network

arXiv: https://arxiv.org/abs/1910.12023 · HTML: https://mlsanity.com/paper/1910.12023/

## Key facts

- Cited by 33 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Oct 2019.

## Built on by (most cited first, 30 of 33)

- [Panoptic Segmentation of Satellite Image Time Series with Convolutional Temporal Attention Networks](https://mlsanity.com/paper/2107.07933/) (2021)
- [SustainBench: Benchmarks for Monitoring the Sustainable Development Goals with Machine Learning](https://mlsanity.com/paper/2111.04724/) (2021)
- [Global and Local Contrastive Self-Supervised Learning for Semantic Segmentation of HR Remote Sensing Images](https://mlsanity.com/paper/2106.10605/) (2021)
- [Unlocking large-scale crop field delineation in smallholder farming systems with transfer learning and weak supervision](https://mlsanity.com/paper/2201.04771/) (2022)
- [AgriFM: A Multi-source Temporal Remote Sensing Foundation Model for Agriculture Mapping](https://mlsanity.com/paper/2505.21357/) (2025)
- [Multi-Region Transfer Learning for Segmentation of Crop Field Boundaries in Satellite Images with Limited Labels](https://mlsanity.com/paper/2404.00179/) (2024)
- [PRUE: A Practical Recipe for Field Boundary Segmentation at Scale](https://mlsanity.com/paper/2603.27101/) (2026)
- [GANmapper: geographical data translation](https://mlsanity.com/paper/2108.04232/) (2021)
- [Combining GEDI and Sentinel-2 for wall-to-wall mapping of tall and short crops](https://mlsanity.com/paper/2109.06972/) (2021)
- [Self-supervised learning -- A way to minimize time and effort for precision agriculture?](https://mlsanity.com/paper/2204.02100/) (2022)
- [A systematic review of the use of Deep Learning in Satellite Imagery for Agriculture](https://mlsanity.com/paper/2210.01272/) (2022)
- [Farmland Extent and Visible Boundary Mapping from 1 m NAIP Imagery Using Residual U-Net and Text-Prompted SAM 3 Refinement](https://mlsanity.com/paper/2607.21881/) (2026)
- [Investigating the Segment Anything Foundation Model for Mapping Smallholder Agriculture Field Boundaries Without Training Labels](https://mlsanity.com/paper/2407.01846/) (2024)
- [Agricultural Landscape Understanding At Country-Scale](https://mlsanity.com/paper/2411.05359/) (2024)
- [GTPBD: A Fine-Grained Global Terraced Parcel and Boundary Dataset](https://mlsanity.com/paper/2507.14697/) (2025)
- [A Comprehensive Review of Agricultural Parcel and Boundary Delineation from Remote Sensing Images: Recent Progress and Future Perspectives](https://mlsanity.com/paper/2508.14558/) (2025)
- [Productive Crop Field Detection: A New Dataset and Deep Learning Benchmark Results](https://mlsanity.com/paper/2305.11990/) (2023)
- [Taking it further: leveraging pseudo labels for field delineation across label-scarce smallholder regions](https://mlsanity.com/paper/2312.08384/) (2023)
- [Global Renewables Watch: A Temporal Dataset of Solar and Wind Energy Derived from Satellite Imagery](https://mlsanity.com/paper/2503.14860/) (2025)
- [The first global agricultural field boundary map at 10m resolution](https://mlsanity.com/paper/2605.11055/) (2026)
- [GuideMe: A Mobile Application based on Global Positioning System and Object Recognition Towards a Smart Tourist Guide](https://mlsanity.com/paper/2105.13426/) (2021)
- [Deep Learning and Earth Observation to Support the Sustainable Development Goals](https://mlsanity.com/paper/2112.11367/) (2021)
- [A knowledge-based data-driven (KBDD) framework for all-day identification of cloud types using satellite remote sensing](https://mlsanity.com/paper/2312.00308/) (2023)
- [Tackling fluffy clouds: robust agricultural field boundary delineation from Sentinel-1 and Sentinel-2 satellite image time series](https://mlsanity.com/paper/2409.13568/) (2024)
- [Segment Anything for Satellite Imagery: A Strong Baseline and a Regional Dataset for Automatic Field Delineation](https://mlsanity.com/paper/2506.16318/) (2025)
- [Segmentation-Aware Latent Diffusion for Satellite Image Super-Resolution: Enabling Smallholder Farm Boundary Delineation](https://mlsanity.com/paper/2511.14481/) (2025)
- [Fields of The World: A Field Guide for Extracting Agricultural Field Boundaries](https://mlsanity.com/paper/2602.08131/) (2026)
- [The role of spatial context and multitask learning in the detection of organic and conventional farming systems based on Sentinel-2 time series](https://mlsanity.com/paper/2603.24552/) (2026)
- [MAgSeg: Segmentation of Agricultural Landscapes in High-Resolution Satellite Imagery using Multimodal Large Language Models](https://mlsanity.com/paper/2605.16179/) (2026)
- [Remote sensing data imputation using deep learning for multispectral imagery](https://mlsanity.com/paper/2605.24003/) (2026)

## Abstract (arXiv)

Applications of digital agricultural services often require either farmers or their advisers to provide digital records of their field boundaries. Automatic extraction of field boundaries from satellite imagery would reduce the reliance on manual input of these records which is time consuming and error-prone, and would underpin the provision of remote products and services. The lack of current field boundary data sets seems to indicate low uptake of existing methods,presumably because of expensive image preprocessing requirements and local, often arbitrary, tuning. In this paper, we address the problem of field boundary extraction from satellite images as a multitask semantic segmentation problem. We used ResUNet-a, a deep convolutional neural network with a fully connected UNet backbone that features dilated convolutions and conditioned inference, to assign three labels to each pixel: 1) the probability of belonging to a field; 2) the probability of being part of a boundary; and 3) the distance to the closest boundary. These labels can then be combined to obtain closed field boundaries. Using a single composite image from Sentinel-2, the model was highly accurate in mapping field extent, field boundaries, and, consequently, individual fields. Replacing the monthly composite with a single-date image close to the compositing period only marginally decreased accuracy. We then showed in a series of experiments that our model generalised well across resolutions, sensors, space and time without recalibration. Building consensus by averaging model predictions from at least four images acquired across the season is the key to coping with the temporal variations of accuracy. By minimising image preprocessing requirements and replacing local arbitrary decisions by data-driven ones, our approach is expected to facilitate the extraction of individual crop fields at scale.