Capturing Inspection Data in a Structured Way: Markers, Metadata and Export

Camera inspections produce large amounts of material: long videos, numerous photos and measurements. How valuable this data is isn’t decided during recording, but by whether it can be found, analyzed and shared later.

From video to a usable record

A video on its own is hard to analyze. It becomes usable through metadata created automatically or with a few inputs during recording:

• the project and section the recording belongs to
• date, time and operator
• device and relevant settings
• position, for example from the system’s distance counter
• markers with finding type and image snapshot

Markers during recording

Findings are captured most efficiently during recording. A good interface offers predefined categories instead of free text, takes over position and time automatically and saves an image snapshot with each marker. Every finding becomes a structured record that can be filtered and analyzed later.

A sensible data model

A clear interface is backed by a clear data model. A hierarchical structure has proven itself: a project contains sections, a section contains recordings, a recording contains markers, and markers reference media files.

{
  "project": "Sample project",
  "section": "S-07",
  "recording": {
    "start": "2026-10-15T09:12:00",
    "markers": [
      { "position_m": 12.4, "type": "Crack", "image": "S07_0124.jpg" }
    ]
  }
}

Export and sharing

Structured data can be converted into different output formats: a report for the client, an industry-standard XML exchange format for the engineering firm or your own job management. What matters is consistent file naming and a completeness check before export. The article XML data exchange between drain cleaning companies and engineering firms describes the handover in detail.

Operation under real conditions

Inspections rarely take place at a desk. The software therefore has to work in a vehicle, in changing light and under time pressure:

• large, unambiguous buttons and few inputs
• keyboard shortcuts or hardware buttons for frequent actions
• clearly visible status for connection, storage and recording
• clear feedback that a marker has been saved

Conclusion

Structured inspection data isn’t created afterwards in the office, but during recording. A suitable interface, a clear data model and validated exports turn video footage into usable information – for reports, assessments and archives.