TreeFinder is a large-scale, high-resolution benchmark dataset for mapping individual dead trees across the contiguous United States. The dataset, created by zhwang1 and last updated in May 2026, is built from 0.6-meter National Agriculture Imagery Program aerial imagery and provides pixel-level annotations of individual dead trees along with ecological metadata.
Use Cases
- Training and benchmarking computer vision models for dead tree detection based on high-resolution aerial imagery.
- Monitoring forest health and mortality at an individual tree scale across large geographic regions.
- Conducting ecological studies on tree mortality patterns based on the provided ecological metadata.
- Developing automated tools for forestry management and wildfire risk assessment based on dead tree locations.
Strengths
- Provides high-resolution 0.6-meter aerial imagery from the National Agriculture Imagery Program.
- Covers the contiguous United States, offering a large-scale geographic scope.
- Includes pixel-level annotations for individual dead trees, enabling precise model training.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count, file formats, and license information are unknown, which may limit suitability assessment.
Provenance
- Source
- National Agriculture Imagery Program (NAIP) aerial imagery.
- Collection Method
- Built from aerial imagery with pixel-level annotations.
- Time Range
- null
- Freshness
- Last updated 2026-05-08 20:26:06; freshness should be verified.
- Geography
- Contiguous United States (CONUS).