Fire and smoke detection
Traditional smoke detectors do not always work on time and accurately. Today, such detectors are being replaced by smart analytics systems.
Solution metrics
96%
Recognition accuracy
Accuracy measured on test videos in the daytime
- < 1
Number of false positives per day per camera
≥ 10 FPS
Frames per second
camera requirement
≥ HD
Recommended video stream quality
Technology Description
The Roadly video analytics system takes fire safety to a new level. Neural networks make it possible to detect fire and smoke with a high accuracy and a low number of false positives.
The main challenge for systems that do not use expensive thermal sensors is distinguishing fire and smoke from the headlight glare, fog, smoke from exhaust pipes, and similar visual effects. Roadly uses a set of algorithms designed to minimize the number of false positives and achieve high quality analysis.
Roadly offers a cost effective and efficient solution optimized for today's environments. Accurate recognition and real-time notifications allow you to quickly take action when fire and smoke are detected, and minimize damage.
The main challenge for systems that do not use expensive thermal sensors is distinguishing fire and smoke from the headlight glare, fog, smoke from exhaust pipes, and similar visual effects. Roadly uses a set of algorithms designed to minimize the number of false positives and achieve high quality analysis.
Roadly offers a cost effective and efficient solution optimized for today's environments. Accurate recognition and real-time notifications allow you to quickly take action when fire and smoke are detected, and minimize damage.
- Accurate differentiation with a minimum of false positives.
- Cutting down the use of the expensive thermal sensors and improving the efficiency of the entire alarm system.
Screenshots and demo videos
Features of the technology
Fire and smoke detection
Differentiation between the fire and smoke and visually similar phenomena
Instant alerts