CAMBER Dataset – Road Survey Video #1

This dataset contains road survey video recordings collected within the CAMBER - European Road Infrastructure Monitoring Project. The videos capture forward-facing visual information on road infrastructure condition, including pavement defects, surface degradation, and road furniture elements.

**Data Generation Method:**
The video was recorded using mobile devices during road monitoring campaigns. Video frames were processed using YOLO-based computer vision models to detect road malfunctions including potholes, manholes, and speed breakers. Detections were geo-referenced using synchronized GPS track data.

**Type of Data:** Video / Image sequences

**Data Format(s):** MP4 (video), CSV (detection results), GPX (GPS track)

**Spatial Coverage:**
- Route distance: 6.48 km
- Geographic coordinates: WGS84 (EPSG:4326)

**Temporal Coverage:**
- Recording date: 2026-01-29 14:49:43 UTC
- Duration: 0.61 hours

**Processing Information:**
- Video ID: 1
- Processing status: Pending
- Safety assessment: Green (Totally Safe)
- Associated Route ID: 1

**Machine Learning Model:**
- Model: models/road_malfunction_3class.pt
- Detected classes: manhole,pothole,speedbreaker
- Confidence threshold: 0.45
- Minimum track age: 5 frames

**Data Access:**
The dataset files are stored on the CAMBER project infrastructure with S3-compatible storage. Zenodo serves as a persistent metadata record and DOI landing page. Access to the full dataset is provided via the related identifiers or upon request.

**Funding:**
This work was funded by the European Union under the Horizon Europe programme, Grant Agreement No. 101156387.