Low Altitude Disaster Imagery (LADI): Civil Air Patrol Aerial Photos, 2015-2023

The Low Altitude Disaster Imagery (LADI) Dataset consists of human and machine annotated airborne images. The imagery was collected by the Civil Air Patrol in support of various disaster responses from 2015 to 2023. It was created by the MIT Lincoln Laboratory Humanitarian Assistance and Disaster Relief group.

Use Cases

• Training object detection models for disaster-related infrastructure damage based on annotated airborne imagery.
• Developing segmentation models for post-disaster landscape analysis based on low-altitude, oblique perspective photos.
• Benchmarking computer vision algorithms on a domain-specific dataset featuring disaster-related features rarely found in standard benchmarks.
• Conducting geospatial analysis of disaster impact over time based on imagery collected from 2015 to 2023.

Strengths

• Imagery has a distinctive low-altitude, oblique perspective, which is rarely featured in standard computer vision benchmarks.
• Data is annotated by both humans and machines, potentially offering varied label quality.
• Covers a multi-year time range from 2015 to 2023, allowing for temporal analysis.
• Released under a permissive CC-BY-4.0 license for broad reuse.

Limitations

• Row count, file formats, and column-level documentation are unknown, which limits suitability assessment.
• Last update date is unknown; freshness unverified.
• Description metadata is limited; actual data quality and geographic coverage require manual inspection after download.

Provenance

Source

MIT Lincoln Laboratory Humanitarian Assistance and Disaster Relief group.

Collection Method

Airborne images collected by the Civil Air Patrol.

Time Range

2015-2023

Freshness

Last updated date is unknown.

Geography

null