ELDOR: Drone Imagery of Illegal Gold Mining in the Amazon Rainforest

ELDOR is a benchmark dataset for detecting illegal gold mining activities in the Amazon Rainforest. The dataset consists of drone-captured orthomosaic imagery and corresponding labels, organized into train, validation, and test splits. It was created by IRSC and documented in a 2026 research paper.

Use Cases

• Training object detection models to identify mining sites based on labeled drone imagery.
• Benchmarking remote sensing algorithms for environmental conservation tasks.
• Analyzing patterns of illegal land use in the Amazon Rainforest based on geospatial image data.

Strengths

• Dataset is associated with a published research paper (ELDOR: A Dataset and Benchmark for Illegal Gold Mining in the Amazon Rainforest).
• Data is pre-split into train, validation, and test sets for machine learning workflows.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and total dataset size are unknown, which may limit suitability assessment.

Provenance

Source

IRSC, as indicated by the author field.

Collection Method

Likely gathered via drone surveys, as indicated by the source folder names 'Drone_Orthomosaic_V1'.

Freshness

Last updated 2026-05-18 15:17:30; freshness should be verified.

Geography

Amazon Rainforest, as stated in the paper title.