Fire & Smoke Detection

Overview

Fire & Smoke Detection node identifies fire and smoke regions in an input frame using a built-in detection model and returns bounding boxes with class labels.

Use this node to trigger safety alerts, activate suppression systems, or log events in surveillance and industrial monitoring flows.

Input

Input Image

image required

The image frame to analyze. Connect this to a camera or upstream image output.

Confidence Threshold

number required

Minimum confidence score to keep a detection. See Confidence Threshold for tuning guidance.

Default: 0.5

NMS Threshold

number required

Non-Maximum Suppression threshold for overlapping detections. See NMS Threshold for tuning guidance.

Default: 0.3

Overlay Results

boolean required

Whether to draw detection bounding boxes and labels on the output frame. See Overlay Results.

Model Directory Path

string

Optional path to a custom model directory. When left empty, the node uses the bundled fire and smoke detection model. See Model Directory Path for details.

Output

Overlay Image

image

Output frame from the node. If overlays are enabled, detected fire and smoke regions are annotated with bounding boxes and labels.

Detected Count

integer

Number of fire or smoke regions detected in the current frame.

Detected Objects

array

Array of detection objects. Each object contains:

• bbox array: Bounding box [x, y, width, height] in image pixel coordinates.
• label string: Detected class — fire or smoke.
• confidence number: Detection confidence score.