Hardware requirements for neural analytics operation

Requirements for the operation of the Smoke detector are given in the table:

Camera
  • We recommend using color cameras. With black and white cameras, the detection quality can be noticeably worse
  • Resolution must be at least 640x360
  • FPS directly affects detection efficiency. In most cases, the default value (1 detection per 10 seconds) is sufficient to detect smoke that lasts more than one minute

Correct operation of the detector isn't guaranteed if you use frames that were masked before they got into Axxon One by an IP camera or similar methods.

Scene and camera angle
  • The minimum smoke area in the frame depends on the neural network that you use. For a standard neural network (see Configuring the Smoke detector), the smoke area must occupy at least 10% of the frame

    In some cases, when smoke is clearly visible, the width/height of the smoke area can be sufficient at 1-3% of the frame width/height.

  • Smoke must be visually separated from the background

If you set a rectangular recognition area for the detector, the requirements apply to this area rather than to the entire frame (see Configuring the Smoke detector).