# A Modelling and Evaluation Framework for EuroCrops-Driven Sentinel-2 Crop Segmentation Source: [A Modelling and Evaluation Framework for EuroCrops-Driven Sentinel-2 Crop Segmentation]() Collection index: [llms.txt]() Data updated: 2026-06-02T05:00:51.557Z Authors: Alexandra Nicoleta Scarlat, Ioana Cristina Plajer, Alexandra Baicoianu Published: 2026-05-30T11:20:29.000Z ## Abstract This work presents a configurable pipeline for generating semantic-segmentation-ready agricultural datasets from Sentinel-2 imagery and EuroCrops parcel-level annotations. The workflow transforms heterogeneous vector crop annotations into aligned multispectral image--mask pairs through label harmonization, Sentinel-2 product selection, spatial alignment, rasterization, patch extraction, quality filtering, and class-aware sample selection. The generated dataset contains 67,337 patches from five European countries and uses a reduced taxonomy of ten crop classes plus background. A four-level U-Net with Group Normalization was trained using 10 Sentinel-2 spectral bands and a composite loss combining class-weighted cross-entropy and Dice loss. On the internal EuroCrops-based test split, the model achieved a mean Intersection over Union (mIoU) of 0.7665, a pixel accuracy of 0.8693, and a mean class accuracy of 0.9072. Compared with spectral and spatial-context Random Forest baselines, the U-Net showed the importance of learned multi-scale spatial representations for crop segmentation. External evaluation was performed on unseen Belgian EuroCrops subsets, DACIA5, and PASTIS. The results show a clear performance gap under external and cross-dataset evaluation, especially for benchmarks with different taxonomies, annotation protocols, spatial coverage, or temporal organization. The model transfers more reliably to dominant and taxonomically aligned classes such as maize and wheat, while performance remains limited for several minority classes and for the adapted single-date PASTIS setting. These findings highlight both the potential and the limitations of using EuroCrops-derived supervision for Sentinel-2 crop segmentation under realistic domain shifts. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Semantic Segmentation on EuroCrops BE\_VLG 2021 (unseen Belgian subsets)](): Semantic segmentation; EuroCrops BE\_VLG 2021 (unseen Belgian subsets); mIoU (All-pixel): 36.73. - [Semantic Segmentation on EuroCrops BE\_WAL 2021 (unseen Belgian)](): Semantic segmentation; EuroCrops BE\_WAL 2021 (unseen Belgian); mIoU (All-pixel): 38.36. - [Semantic Segmentation on DACIA5 All valid pixels (April-September)](): Semantic segmentation; DACIA5 All valid pixels (April-September); mIoU (All Classes): 15.05. - [Semantic Segmentation on DACIA5 Labeled crop pixels only (April-September)](): Semantic segmentation; DACIA5 Labeled crop pixels only (April-September); mIoU (All Classes): 23.83. - [Semantic Segmentation on PASTIS Labeled crop pixels only (April-September)](): Semantic segmentation; PASTIS Labeled crop pixels only (April-September); mIoU (All Classes): 11.89. ## Original sources and implementation [Original paper]() [PDF]()