MCTED: Mars CTX Terrain-Elevation Dataset with Optical and DEM Pairs

MCTED is a machine-learning-ready dataset of optical images of the surface of Mars, paired with their corresponding digital elevation models. It was created by ESA-Datalabs using the NASA Ames Stereo Pipeline and Mars Reconnaissance Orbiter CTX instrument imagery, as referenced in Day et al. 2023. The dataset was last updated on the platform in November 2025.

Use Cases

• Train machine learning models for terrain classification based on paired optical and elevation data.
• Develop multimodal fusion models for Martian surface analysis based on image-DEM pairs.
• Conduct geomorphological studies of Mars based on high-resolution satellite imagery and elevation models.

Strengths

• Data is explicitly described as 'machine-learning-ready', suggesting preprocessing for direct use.
• Pairs optical images with corresponding digital elevation models, providing multimodal data.
• Created using the NASA Ames Stereo Pipeline and Mars Reconnaissance Orbiter CTX imagery, indicating a credible source methodology.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and dataset size are unknown, which may limit suitability assessment.
• License information is unavailable, which could restrict usage.

Provenance

Source

ESA-Datalabs, using NASA Mars Reconnaissance Orbiter CTX instrument imagery.

Collection Method

Created using the NASA Ames Stereo Pipeline to generate orthoimage-DEM pairs.

Time Range

null

Freshness

Last updated 2025-11-06 13:11:24; freshness should be verified.

Geography

Surface of Mars.