Optical Flow Wind Velocity Estimates from Tree Motion Video

Supplementary files for a proof-of-concept study correlating Optical Flow Tracking Velocimetry (OFTV) derived wind estimates with peak 3-second gust speeds. Results from a pear tree in a domestic setting show correlations between OFTV-𝑠3 and cup-anemometer data (R2 = 0.65, p < 0.05). The data was authored by Sai Kulkarni and last updated on 2026-04-17.

Use Cases

• Calibrating visual anemometry models based on optical flow tracking of tree motion.
• Validating machine learning frameworks for wind speed estimation using video data.
• Assessing site-specific wind hazards for the (re)insurance sector based on the described methodology.
• Exploring crowd-sourced wind mapping using existing CCTV infrastructure as suggested in the description.

Strengths

• Demonstrates correlation between OFTV-derived wind estimates and cup-anemometer data (R2 = 0.65, p < 0.05).
• Shows strong correlation (R2 ≥ 0.92) across a range of reference objects like flags and grass.
• Files are provided under a permissive CC-BY 4.0 license.

Limitations

• Dataset is very small (341.5 KB), indicating limited scope.
• Row count and column-level documentation are unknown; data structure must be inferred from supplementary PDFs.
• Data is from a single proof-of-concept study in a domestic setting, which may limit generalizability.

Provenance

Source

figshare, authored by Sai Kulkarni.

Collection Method

Optical Flow Tracking Velocimetry (OFTV) applied to video footage of trees and other objects.

Freshness

Last updated 2026-04-17 11:17:22.

Geography

Likely a domestic setting, as referenced by a pear tree in the description.