Smoke and Fire Detection

Smoke and Fire Detection

Nodeflux Smoke and Fire Detection

Nodeflux Smoke and Fire Detection addresses the incidents of smoke and fire event, promoting improved safety and security within monitored environments even in the case wildfire event. This solution integrates advanced Large Vision Models and Visual Transformers directly into the Nodeflux Visionaire platform. The analytics system is capable of automatically detecting smoke and fire separately in real-time scenarios to mitigate the early fire event as soon as the visual inference result identified.

Designed Applications

Nodeflux Smoke and Fire Detection is primarily intended for use in public and environment safety as well as surveillance applications. It helps prevent disasters, protect lives, and reduce property damage as well as aids in mitigating or even preventing potential risks or threats associated with other unforeseen events that may transpire.

Disclaimer: In a sense of large vision models and visual transformers technology, the performance of this analytic might slightly differ in your environment, depending on several variables such as camera specs, camera height, camera angle, weather conditions, etc. We highly recommend you to test our analytics and run benchmarks on your own images, with ground truth or your quality expectations prepared beforehand. Please contact us for more info.

Architecture

Visionaire Smoke and Fire Detection utilizes a combination of other services:

1. Postgres - For database,

2. Docker Snapshot - For object detection related,

3. Visionaire Docker Stream (must be v4.57.11 and above) - For video stream processing,

4. VisionaireDashboard (Optional) - Built-in dashboard for visualization.

Additional Necessary Deployment (OVOD)

Since Nodeflux Smoke and Fire Detection utilizes the snapshot platform, an additional docker-compose.yml file needs to be deployed alongside the other aforementioned services. The following directive link outlines the installation process for deploying the docker-compose.yml file specific to the Open Vocabulary Object Detection (OVOD) pipeline of snapshot platform:

OVOD Installation

Analytics Configuration

When you are assigning this analytics into the stream through the dashboard, you should configure some parameters setting below:

{
  "dump_interval": 3,      
  "address": "http://192.168.101.105:4006",
  "dump_confidence" : 0.3
}

Parameter	Explanation
dump_interval	How frequent you want the stream as a frame snapshot to get the inference result. Unit in second
address	The IP address and its port where the OVOD snapshot service is deployed
dump_confidence	The value of minimum confidence threshold so that the inference result should be dumped as the event. Unit ranged between 0.1 to 1.