Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

High-Resolution Semantic Segmentation Dataset ofAeolianDunes in Five Mars-Analog Regions on Earth

Domain:

geospatial

Record type:

dataset
Creator:
Kai
Host:avatar

Abstract: This dataset provides high-resolution semantic segmentation maps ofaeoliansand dunes across five major Mars-analog regions on Earth: the Namib Desert (Africa), Atacama Desert (South America), Mojave Desert (North America), Qaidam Basin (Asia), and Yilgarn Craton (Australia). These regions are critical terrestrial analogs for studying Martianaeolianprocesses.

The classification results were generated using a novel Vision-Language Model (SegFormer with Text-Aware Fusion) trained on Google Satellite Embeddings (64-dimensional feature vectors) and expert geomorphological knowledge. The model achieves an Overall Accuracy (OA) of 96.11% and a Kappa coefficient of 0.9418. This dataset offers a detailed inventory of dune types, supporting comparative planetary studies and Earth system science research.

Data Content: The dataset consists of 5 GeoTIFF (.tif) raster files and 1 Excel file:

  1. Namib.tif: Classification result for the Namib Desert.
  2. Atacama.tif: Classification result for the Atacama Desert.
  3. Mojave.tif: Classification result for the Mojave Desert.
  4. Qaidam.tif: Classification result for the Qaidam Basin.
  5. Yilgran.tif: Classification result for the Yilgarn Craton.
  6. ClassificationScheme.xlsx: Detailed description of the classification taxonomy and criteria.

Pixel Value Definitions (Classification Scheme): The raster values correspond to the following dune categories:

  • 0: Crescentic Dunes (Cre)
  • 1: Dome Dunes (Dom)
  • 2: Linear Dunes (Lin)
  • 3: Nebkha (Neb)
  • 4: Network Dunes (Net)
  • 5: Obstacle Dunes (Obs)
  • 6: Others (Oth)
  • 7: Parabolic Dunes (Par)
  • 8: Sand Sheets (San)
  • 9: Star Dunes (Sta)
  • 255: NoData / Background

Technical Details:

  • Format: GeoTIFF (LZW Compression)
  • Coordinate System: WGS 84 (EPSG:4326)
  • Source Data: Google Satellite Embeddings

Visit

figshare.com

Tasks

computer visionimage classification

Tags

Earth and space science informaticsPlanetary geologySand dunesAeolian landformsMars analogsSemantic segmentationDeep learningGoogle Satellite EmbeddingsGeomorphology

Licenses

CC BY 4.0

Similar

Semantic Segmentation of Crop Type in Ghana DatasetSemantic Segmentation-Based Building Footprint Extraction Using Very High-Resolution Satellite Images and Multi-Source GIS DataHigh-Resolution Semantic Segmentation of Woodland Fires Using Residual Attention UNet and Time Series of Sentinel-2Semantic Segmentation of Crop Type in South Sudan DatasetUsing image segmentation models to analyse high-resolution earth observation data: new tools to monitor disease risks in changing environmentsHigh-Resolution Load Dataset in Morocco

Semantic Segmentation of Crop Type in Ghana Dataset

Semantic Segmentation-Based Building Footprint Extraction Using Very High-Resolution Satellite Images and Multi-Source GIS Data

Automatic extraction of building footprints from high-resolution satellite imagery has become an imp

High-Resolution Semantic Segmentation of Woodland Fires Using Residual Attention UNet and Time Series of Sentinel-2

Southern Africa experiences a great number of wildfires, but the dependence on low-resolution produc

Semantic Segmentation of Crop Type in South Sudan Dataset

Using image segmentation models to analyse high-resolution earth observation data: new tools to monitor disease risks in changing environments

Abstract Background In the near future, the incidence of mosquito-borne diseases may expand to new s

High-Resolution Load Dataset in Morocco

High-Resolution Load Dataset from Smart Meters Across Various Cities in Morocco