Papers › Enhancing crop segmentation in satellite image time-series with transformer networks

Enhancing crop segmentation in satellite image time-series with transformer networks

3 Apr 2024ACNP 2024 4archive 2025-07-28

Ignazio Gallo, Mattia Gatti, Nicola Landro, C. Loschiavo, M. Boschetti, Riccardo La Grassa, A. U. Rehman

Recent studies have shown that Convolutional Neural Networks (CNNs) achieve impressive results in crop segmentation of Satellite Image Time-Series (SITS). However, the emergence of transformer networks in various vision tasks raises the question of whether they can outperform CNNs in crop segmentation of SITS. This paper presents a revised version of the Transformer-based Swin UNETR model adapted specifically for crop segmentation of SITS. The proposed model demonstrates significant advancements, achieving a validation accuracy of 96.14% and a test accuracy of 95.26% on the Munich dataset, surpassing the previous best results of 93.55% for validation and 92.94% for the test. Additionally, the model’s performance on the Lombardia dataset is comparable to UNet3D and superior to FPN and DeepLabV3. Experiments of this study indicate that the model will likely achieve comparable or superior accuracy to CNNs while requiring significantly less training time. These findings highlight the potential of transformer-based architectures for crop segmentation in SITS, opening new avenues for remote sensing applications.

Paper PDF Code

Code

gitlab.com/mattiagatti/sentinel2-crop-mapping-models officialpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Segmentation Semantic Segmentation Time Series UNET Segmentation

Results from the paper archive 2025-07-28

Task	Dataset	Model	Metric	Value	Rank at snapshot	Leaderboard	Report
Semantic Segmentation	Lombardia Sentinel-2 Image Time Series for Crop Mapping	UNet3D	Overall Accuracy	80.77	#1 of 4	Archive leaderboard	report
Semantic Segmentation	Lombardia Sentinel-2 Image Time Series for Crop Mapping	Swin UNETR	Overall Accuracy	79.64	#2 of 4	Archive leaderboard	report
Semantic Segmentation	Lombardia Sentinel-2 Image Time Series for Crop Mapping	3D FPN with NDVI Loss	Overall Accuracy	77.23	#3 of 4	Archive leaderboard	report
Semantic Segmentation	Lombardia Sentinel-2 Image Time Series for Crop Mapping	DeepLabv3 3D	Overall Accuracy	74.51	#4 of 4	Archive leaderboard	report
UNET Segmentation	Munich Sentinel2 Crop Segmentation	Swin UNETR	Overall Accuracy	95.26	#1 of 5	Archive leaderboard	report
UNET Segmentation	Munich Sentinel2 Crop Segmentation	UNet3D	Overall Accuracy	94.73	#2 of 5	Archive leaderboard	report
UNET Segmentation	Munich Sentinel2 Crop Segmentation	DeepLabv3 3D	Overall Accuracy	85.98	#5 of 5	Archive leaderboard	report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

1x1 Convolution Attention Batch Normalization Concatenated Skip Connection Convolution Dense Connections FPN Linear Layer Max Pooling Multi-Head Attention Position-Wise Feed-Forward Layer ReLU Residual Connection Softmax U-Net UNETR

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections