Video-based fire and smoke detection

The problem

How it works

1. 1Camera streams are analysed continuously on-premise.
2. 2A model trained on flame and smoke appearance flags candidate regions.
3. 3A short temporal check reduces false alarms from reflections, lights or steam.
4. 4An alert with a snapshot is sent to the control room and configured channels.

What can be detected

• Visible flame
• Visible smoke
• Event location and evidence clip for review

Typical scenarios

• High-bay warehouses and production halls
• Outdoor storage yards and waste areas
• Kitchens and technical rooms
• Battery charging areas

Working with existing cameras

• The camera must have a view of the area where fire or smoke could appear; detection range depends on resolution and distance.
• Strong backlight, steam and dust are reviewed during camera assessment.
• Most IP cameras and NVRs that provide RTSP or ONVIF streams can be connected, regardless of brand.
• Before a pilot, sample footage from each candidate camera is reviewed for resolution, angle, distance and lighting.
• Where a view is not suitable, the recommendation is usually to reposition the camera or add one — not to replace the whole system.

Integration

• Control room dashboard and mobile alerts
• REST API / webhook into existing alarm workflows
• Relay outputs for local signalling

Privacy and security

• No face recognition is used in safety and operations analytics; events describe what happened, not who.
• On-premise processing keeps video inside your network.
• Evidence clips, retention periods and user access are configurable and should follow your GDPR/KVKK policies.
• Workplace monitoring should be introduced with clear employee information and, where applicable, consultation with employee representatives.

Limitations

• Video fire detection is a supplementary layer. It does not replace certified fire detection and alarm systems required by regulation.
• Fire that is not visible to a camera (inside equipment, behind walls) cannot be detected.