Galaxy Segmentation Benchmark from Pan-STARRS Imaging and Galaxy Zoo Labels

Pan-STARRS imaging data provides the basis for a galaxy segmentation benchmark. Labels for galactic centers, bars, spiral arms, and foreground stars are derived from citizen scientist contributions to the Galaxy Zoo: 3D project. The dataset is hosted on AWS Open Data and originates from NASA.

Use Cases

• Train semantic segmentation models to identify galactic centers based on labeled imaging data.
• Develop algorithms for detecting galactic bars and spiral arms in telescope images.
• Benchmark model performance on separating foreground stars from extended galaxy structures.
• Study the reliability of citizen science labels for complex astronomical classification tasks.

Strengths

• Labels are derived from the established Galaxy Zoo: 3D citizen science project.
• Imaging data comes from the Pan-STARRS astronomical survey.
• The dataset has no restrictions on use.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

NASA

Collection Method

Derived from citizen scientist labels from the Galaxy Zoo: 3D project applied to Pan-STARRS imaging data.