Documentation for Axxon One 2.0. Documentation for other versions of Axxon One is available too.

Previous page Configuring perspective  Displaying the number of detected objects Next page

On the page:


General information

The scanning mode in Axxon One can improve detection of small objects or objects in areas far away from the camera. For example, when the camera is placed at a long distance and objects (such as people, smoke, fire) are small in the frame, scanning mode can improve detection.

In the scanning mode, the frame is divided into an arbitrary number of scanning windows for detection accuracy.

Scanning windows are parts of the frame that are analyzed separately and all together at the same time. Scanning windows allow the detector to receive more events, which improves the accuracy of detecting distant or small objects that otherwise can be missed.

Attention!

The scanning mode doesn’t provide absolute detection accuracy, regardless of the number of windows.

The scanning mode is available for:

  1. Fire detector.
  2. Smoke detector.
  3. Human pose detector.
  4. Neural counter.
  5. Neural tracker.

Configuring the scanning mode

You can configure the scanning mode using the parameters of the corresponding detector. The list of required parameters and their values are presented in the table below.

ParameterValueDescription
Scanning windowYesSelect the Yes value to enable the scanning mode
No
Scanning window heightNumeric value (in pixels)

The height and width of the scanning window are determined according to the actual size of the frame and the required number of windows.

For example, the real frame size is 1920×1080 pixels. To divide the frame into four equal windows, set the width of the scanning window to 960 pixels and the height to 540 pixels

Scanning window width
Scanning window step height

The scanning step determines the relative offset of the windows. If the step is equal to the height and width of the scanning window respectively, the segments will line up one after another. 

Reducing the height or width of the scanning step increases the number of windows due to their overlapping each other with an offset. This increases the detection accuracy, but also increases the CPU load

Attention!

The height and width of the scanning step must not be greater than the height and width of the scanning window—detector doesn't operate with such settings.

Scanning window step width

Recommendations for configuring the scanning mode

  1. Set the same values for the window height and scanning step, as well as the window width and scanning step.
  2. Set the optimal number of scanning windows—from four to eight. The number of windows depends on specific purposes, quality of the original frame, capabilities of devices. If the original frame is of low quality or small size, a large number of windows won’t increase the detection accuracy, but can lead to false events from a detector or no detection. You must select the number of scanning windows experimentally.
  3. If you don't get events from a detector at the same values of window and step, you must experimentally reduce the scanning steps relative to the height or width.

Attention!

Increasing the number of windows increases the load on the device.

Example of configuring the scanning mode

It is necessary to configure the scanning mode for the smoke detector to detect smoke in the distance. If the actual frame size is 1920×1080 pixels, set the following values in the detector settings to divide the frame into four equal windows:

  1. Scanning window—Yes.
  2. Scanning window height—540.
  3. Scanning window width—960.
  4. Scanning window step height—540.
  5. Scanning window step width—960.

As a result, the frame is conditionally divided into four equal windows that are analyzed separately and all together at the same time. If smoke is detected in any of these windows, the detector can generate an event.

  • No labels