Documentation for Axxon One 3.0/3.1 version. Documentation for Axxon One 2.0 version is also available.


Video stream and scene requirements for the Crowd estimation VA

Image requirements for the Crowd estimation VA

Hardware requirements for neural analytics operation

To configure the Crowd estimation VA, do the following:

  1. Go to the Detectors tab.

  2. Below the required camera, click Create… → Category: Retail → Crowd estimation VA.

By default, the detector is enabled and set to count crowds of people.

If necessary, you can change the settings of the detector parameters given in the table:

ParameterValueDescription
Object features
Video streamMain streamIf the camera supports multistreaming, select the stream for which detection is needed
Record mask to archiveYesBy default, the recording of the mask to the archive is disabled. To record the sensitivity scale of the detector to the archive (see Display of the information from a detector (mask)), select the Yes value
No
Object identification
NameCrowd estimation VAEnter the detector name or leave the default name
EnableYesBy default, the detector is enabled. To disable, select the No value
No
TypeCrowd estimation VAName of the detector type (non-editable field)
Number of frames processed per second0.017

Enter the number of frames for the detector to process per second. The value must be in the range [0.016, 100]

Decoder mode

Select a processor for decoding video. CPU is selected by default


Auto

GPU takes priority (decoding on NVIDIA NVDEC chips). If a suitable GPU isn't available, decoding is performed using Intel Quick Sync Video technology. Otherwise, decoding is performed on the CPU

CPU

Decoding is performed on the CPU

GPU

Decoding is performed on the GPU (decoding on NVIDIA NVDEC chips)

HuaweiNPU

Decoding is performed on the HuaweiNPU

Delayed start (sec)0

Enter the time in seconds to delay the start of the detector after receiving the first frame. The delay is applied randomly, ranging from 0 to the specified value.

The default value is 0, meaning the detector starts without delay.

The optimal value of delay depends on the configuration and the number of detectors. We recommend entering a value of delay that is no less than the number of detectors. The value must be in the range [0, 3600] 

Detection mode
CPU

Select a processor for the operation (see Selecting Nvidia GPU when configuring detectors)

Attention!

If you specify other processing resource than the CPU, this device will carry the most of the computing load. However, the CPU will also be used to run the detector.

Nvidia GPU 0
Nvidia GPU 1
Nvidia GPU 2
Nvidia GPU 3
Trigger to
Greater than or equal to threshold value

Select when you want to generate events. The Crowd estimation VA will generate events from the threshold value set in the Number of alarm objects field

Less than or equal to threshold value
Change in readings
Number of alarm objects1

Enter the number of objects at which an event occurs. The value must be in the range [0, 10000]

Visualization settings
Detected objectsNo

The default value is No. If you want to view detected objects, select the Yes value

Attention!

To ensure correct display of detected objects (to eliminate flickering), enter the value of at least 4 in Number of frames processed per second. After you complete configuration, we recommend entering the original value of 0.017 to reduce the load on the server.

Yes

By default, the entire frame is the detection area. In the preview window, you can specify the detection areas using the anchor points  (see Configuration of a detection area). To delete the specified area, click the Delete button. 

To save the parameters of the detector, click the Apply button. To cancel the changes, click the Cancel button.

It is possible to display the sensor and the number of objects in the monitored area in the surveillance window on the layout (see Display the number of detected objects).

  • No labels