To configure the queue detection tool, do the following:

1. Enable the detection tool — select Yes for the corresponding parameter.
2. Enter the name of the detection tool or leave the default name.
3. To record mask (highlighting of queues/lines) to archive (see Extra information overlay (Masks)), select Yes for the Record mask to archive parameter (1).
   
4. If a camera supports multistreaming, select the stream for which detection is needed (2). Selecting a low-quality video stream allows reducing the load on the Server.
   

5. Set an event transmission interval in seconds for sending data to the AxxonData report subsystem (see Queue length report) if the queue length exceeds the set limit (3). Zero value means no transmission of events.

6. Select a processing resource for decoding video streams (4). When you select a GPU, a stand-alone graphics card takes priority (when decoding with Nvidia NVDEC chips). If there's no appropriate GPU, the decoding will use the Intel Quick Sync Video technology. Otherwise, CPU resources will be used for decoding.
7. By default, the frame is compressed during the analysis to the specified size (5, default size is 1920 pixels on the larger side). In this case, the following algorithm is used:
   1. if the original resolution on the larger side of the frame is greater than the one specified in the Frame size change field, then it is divided in half;
   2. if the resulting resolution is less than the specified one, then the algorithm stops and this resolution is used;

   3. if the resulting resolution is still greater than the specified one, then it is divided in half until it becomes less.

      For example, the original video resolution is 2048*1536, the specified value is 1000. In this case, the original resolution will be halved 2 times (512*384), because after the first division, the value on the larger side of the frame will be greater than the specified value (1024 > 1000).

      
      

      To avoid detection errors on streams with a higher resolution, it is recommended that compression be reduced.

      
      

8. Set the interval in seconds for sending events to the AxxonData reporting subsystem (see Queue length report) if the queue size exceeds the specified value (6). If the value is set to 0, then events will not be sent.

9. Set the frame rate value for the detection tool to process (6). This value should be in the range [0.016, 100].  The higher the value of this parameter, the greater the CPU load.

10. Specify the number of people in the queue to trigger the alarm when exceeded (7). The value should be within the range of [2; 20]. 
11. Specify the detection tool sensitivity in standard units from 0 to 1 (8). The higher the sensitivity, the smaller the disturbances analyzed by the queue detection algorithm. Alternatively, the lower the sensitivity, the greater changes in scene are processed by the queue detection tool.
    You should set the sensitivity value empirically based on the Motion Mask data displayed in the Preview window.
    
12. In the Preview window, you can set the detection zones with the help of anchor points much like privacy masks in Scene Analytics.

13. Click  and set the approximate size of a human. You can do so by dragging the corners of the rectangular area.

    It is recommended to set the minimum size of a human graphically, the numbers in the Human size field are conditional values.

    
    

14. Click Apply.

The queue detection tool is now configured. When the detection tool is triggered, the following events are generated:

Camera "6. Camera". Detection "Queue detection" triggered, quaue (min.: 6, max.: 6)

where min. and max. is estimated queue length.