EPRI Distribution Inspection Imagery: Drone-Collected Utility Asset Photos

Drone-collected imagery of electrical distribution infrastructure has been labeled to support public benefit and research and development. The dataset is hosted on Kaggle, but specific details on volume, collection dates, and contributors are not provided. Its primary purpose is to facilitate machine learning applications for utility asset inspection.

Use Cases

• Train object detection models to identify utility poles and other assets based on drone imagery.
• Develop semantic segmentation models for classifying different components of electrical distribution systems.
• Benchmark anomaly detection algorithms for identifying damaged or faulty equipment in inspection photos.
• Support research into automated geospatial mapping and inventory of utility assets from aerial imagery.

Strengths

• Data is specifically labeled for machine learning tasks, indicating a prepared training resource.
• Imagery is collected via drones, suggesting a consistent aerial perspective suitable for automated analysis.

Limitations

• Row count, file formats, and dataset size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics and label categories must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

Kaggle

Collection Method

Drone-collected imagery

Time Range

null

Freshness

Last updated date is unknown.

Geography

null