Four years of UAV-LiDAR data (2022–2025) were used to evaluate four individual tree segmentation algorithms at a coastal Douglas-fir genetic trial site near Jordan River, Vancouver Island. The dataset includes normalized point clouds with a density of approximately 100 points/m² and performance metrics assessed against a field-referenced dataset of 1,526 live trees. The study was authored by Gayatri Deepak Kulkarni and harvested from Borealis Dataverse.
Use Cases
- Benchmarking tree segmentation algorithm performance based on precision, recall, and F-score metrics described in the study.
- Analyzing the impact of stand structure and tree mortality on segmentation accuracy based on the described conditions of the trial plot.
- Developing methods for repeated forest surveys using UAV-LiDAR data based on the multi-temporal (four-year) data collection.
- Studying coastal Douglas-fir growth and development in a genetic trial context using the individual tree structural data.
Strengths
- Multi-temporal data collected over four years (2022–2025), allowing for analysis of consistency over time.
- Algorithm performance was assessed against a field-referenced dataset of 1,526 live trees, providing a concrete validation baseline.
- Point cloud data has a specified density of approximately 100 points/m², indicating a known resolution.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count and specific file formats are unknown, which may limit suitability assessment.
- The dataset focuses on a single species and location, which may limit generalizability to other forest types.
Provenance
- Source
- Borealis Harvested Dataverse
- Collection Method
- Data collected via UAV equipped with LiDAR over a genetic trial site, processed using the lidR package in R.
- Time Range
- 2022–2025
- Freshness
- Last updated 2026-05-02 04:11:14; freshness should be verified.
- Geography
- Coastal Douglas-fir genetic trial site near Jordan River, Vancouver Island, British Columbia, Canada.