To configure Neurocounter, do the following:

1. To record mask (highlighting of recognized objects) to archive, select Yes for the corresponding parameter (1).
   
2. If the camera supports multistreaming, select the stream for which detection is needed (2). 
   
3. Select a processing resource for decoding video streams (3). When you select a GPU, a stand-alone graphics card takes priority (when decoding with NVIDIA NVDEC chips). If there is no appropriate GPU, the decoding will use the Intel Quick Sync Video technology. Otherwise, CPU resources will be used for decoding.
4. If you need to outline objects in the preview window, select Yes for the Detected objects parameter (4).
5. Set the recognition threshold for objects in percent (5). If the recognition probability falls below the specified value, the data will be ignored. The higher the value, the higher the accuracy, but some triggers may not be considered.

6. Set the frame rate value for the detection tool to process per second (6). This value should be in the range [0.016, 100]. 

   The default values (3 output frames and 1 FPS) indicate that Neurocounter will analyze one frame every second. If Neurocounter detects more objects than the specified threshold value on 3 frames, then it triggers.

   
   

7. Select the processor for the neural network - CPU, one of GPUs, or Intel NCS (7, see Hardware requirements for neural analytics operation, General Information on Configuring Detection). 

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

   
   

   It may take several minutes to launch the algorithm on NVIDIA GPU after you apply the settings. You can use caching to speed up future launches (see Configuring the acceleration of GPU-based neuroanalytics).

   
   

8. In the Object type field (11), select the object type for counting, or in the Neural network file field (8), select the neural network file.
   

   To train your neural network, contact AxxonSoft (see Data collection requirements for neural network training). A trained neural network for a particular scene allows you to detect only objects of a certain type (e.g. person, cyclist, motorcyclist, etc.).  If the neural network file is not specified, the default file will be used, which is selected depending on the selected object type (11) and the selected processor for the neural network operation (7).

   
   

   For correct neural network operation on Linux, place the corresponding file in the /opt/AxxonSoft/AxxonNext/ directory.

   
   

9. Set the triggering condition for Neurocounter:

   1. In the Number of alarm objects field, set the threshold value for the number of objects in FOV (9). 

   2. In the Trigger upon count field, select when you want to generate the trigger − when the number of objects in the detection zone is greater or less than the threshold value (12). 

10. Set the minimum number of frames with the excessive numbers of objects for Neurocounter to trigger (10). The value should be within the range of [2; 20].

11. In the preview window, you can set the detection zones with the help of anchor points much like privacy masks in Scene Analytics (see Setting General Zones for Scene Analytics). By default, the entire FOV is a detection zone.
    

12. Click the Apply button.

After the neural tracker is created, it is possible to display the sensor and the number of objects in the controlled zone in the video surveillance window on the layout. To configure this option, do the following:

1. Switch to the Layout Editing mode (see Switching to layout editing mode).
2. Place the sensor anywhere in FOV.
   
3. Customize the font. To do this, press the button.
4. Save the layout (see Exiting layout editing mode). As a result, the sensor and the number of objects will be displayed in the selected spot: