Page History

Configuration of the Neurocounter module includes: configuring the detector and selecting the area of interest. You can configure the Neurocounter module on The Neurocounter module can be configured on the settings panel of the Neurocounter object created on the basis of the Camera object on the Hardware tab of the System settings dialog window.

Image ModifiedThe Neurocounter module is configured as follows:

Configuring the detector

1. Go to the the settings panel of the Neurocounter object settings panel.
   
2. Set the the Show objects on image  checkbox (1), if it is necessary to frame the detected objects object on the video image in the debug window (see Start the debug window).
3. From the Camera position drop-down list, select:
   1. Wall—objects are detected only if their lower part gets into the area of interest specified in the detector settings.
   2. Ceiling—objects are detected even if their lower part doesn't get into the area of interest specified in the detector settings.
4. In the Number of frames for analysis and output
 field (2)
5. field, specify the number of frames
to
6. that must be processed to determine the number of objects on them.
7. In the Frames processed per second [0
,
8. .016, 100] field
(3)
9. ,
set
10. specify the number of frames processed per second by
the detection tool
11. the neural network in the range from 0.016 to 100. For all other frames interpolation is performed—finding intermediate values by the available discrete set of its known values. The greater the value of the parameter, the more accurate the detector operation, but the higher the load on the processor.
12. From the Send event drop-down list
(4)
13. , select the condition by which an event with the number of detected objects
will be
14. is generated:
    • If threshold exceeded is
triggered
15. • generated if the number of detected objects in the image is greater than the value specified in the Alarm objects count field
.
16. • ;
    • If threshold not reached is
triggered
17. • generated if the number of detected objects in the image is less than the value specified in the Alarm objects count field
.
18. • ;
    • On count change is
triggered
19. • generated every time the number of detected objects changes
.
20. • ;
    • By period is
triggered
21. • generated by
a
22. • the time period:
      1. In the Event periodicity
 field (5), set
23. • 1. field, specify the time after which the event with the number of detected objects
will be
24. • 1. is generated. The range of values: from 1 to 100—for seconds, minutes, hours; from 1 to 20—for days.
      2. From the Time interval
 drop
25. • 1. drop-down list
(6)
26. • 1. , select the time unit of the counter period: seconds, minutes, hours, days.
         

         Info
         title Note
         If the entered value exceeds the allowable range, then after you click the Apply button, the maximum value is set automatically.

27. In
the 
28. the Alarm objects count field
(7)
29. ,
set
30. specify the threshold number of detected objects in the area of interest. It is used in the If threshold exceeded and If threshold not reached conditions. The default value is 5.

31. In the Recognition threshold [0, 100]

 field (8)

32. field, enter the

neural counter sensitivity—

33. neurocounter sensitivity—an integer value in the range from 0 to 100. The default value is 30.

    Info
    title Note
    The

neural counter

34. neurocounter sensitivity is determined experimentally. The lower the sensitivity, the higher the

more false triggerings there might be

35. probability of false alarms. The higher the sensitivity, the lower the

fewer false triggerings there might be

36. probability of false alarms, however, some useful tracks

might

37. can be skipped. See Example of configuring neurocounter for solving typical tasks.

38. Set the Scanning mode checkbox to detect small objects. If you enable this mode, the load on the system increases. So in step 5 we recommend specifying a small number of frames processed per second. By default, the checkbox is cleared. For more information on the scanning mode, see Configuring the Scanning mode.
39. By default, the standard (default) neural network is initialized according to the object type selected in step 14 and the device type in step 13. The standard neural networks for different processor types are selected automatically; you must not do it manually. If you use a custom neural network, then click the Image Added button to the right of a unique neural network is prepared for use, in the Tracking model field, click the Image Removed button (9), and select the file in the standard Windows Explorer Explorer window that opens. If the field is left blank, the default neural networks will be used for detection. They are selected automatically depending on the selected object type (11) and device (10)., specify its file.
    

    Note
    title Attention!
    To train a neural network, contact the AxxonSoft technical support (see Data collection requirements for neural network training). A neural network trained for a specific scene allows detecting objects of a certain type only (for example, a person, cyclist, motorcyclist, and so on).

40. Set the Model quantization checkbox to enable themodel quantization. By default, the checkbox is cleared. This parameter allows reducing the consumption of the GPU's computational power.
    

    Info
    title Note
    AxxonSoft conducted a study in which a neural network model was trained to identify the characteristics of the detected object. The following results of the study were obtained: model quantization can lead to both an increase in the percentage of recognition and a decrease. This is due to the generalization of the mathematical model. The difference in detection percentage ranges within ±1.5%, and the difference in object identification ranges within ±2%. Model quantization is only applicable for NVIDIA GPUs. The first launch of the detector with the activated quantization feature can take longer than a standard launch. If you use the GPU caching, next time the detector with quantization runs without delays.

41. If necessary, specify the class of the detected object in the Target classes field. If you want to count and display tracks of several classes, specify them separated by a comma with a space. For example, 1, 10.
    The numerical values of classes for the embedded neural networks: 1—Human/Human (top view), 10—Vehicle.
    

    Info
    title Note
    If you specify a class/classes from the neural network and a class/classes missing from the neural network, the tracks of a class/classes from the neural network are counted and displayed (Object type, Neural network file). If you specify a class/classes missing from the neural network, tracks aren't counted and displayed.

42. From the Device drop-down list, select the device on which the neural network operates: CPU, one of NVIDIA GPUs, or one of Intel GPUs. Auto (default value)If the path to the neural network was not specified at step 7, from the Device drop-down list (10), select the device on which the neural network will operate. Auto—the device is selected automatically: NVIDIA GPU gets the highest priority, followed by Intel GPU, then CPU.
    

    Note
    title Attention!
    We recommend using the GPU. It can take several minutes to launch the algorithm on the NVIDIA GPU after you apply the settings. You can use caching to speed up future launches (see Optimizing the operation of neural analytics on GPU).

43. From the Object type drop-down list
(11)
44. , select the object type
if the path to the neural network was not specified at step 7
45. :
    • Human—the camera is directed at a person at
the
46. • an angle of 100-160°
.
47. • ;
    • Human (top-down view)—the camera is directed at a person from above at a
sight
48. • slight angle
.
49. • ;
    • Vehicle—the camera is directed at a vehicle at
the
50. • an angle of 100-160°;
    • Person and vehicle (Nano)—person and vehicle recognition, small neural network size;
    • Person and vehicle (Medium)—person and vehicle recognition, medium neural network size;
    • Person and vehicle (Large)—person and vehicle recognition, large neural network size.
      

      Info
      title Note
      Neural networks are named taking into account the objects they detect. The names can include the size of the neural network (Nano, Medium, Large), which indicates the amount of consumed resources.

Specify the detection surveillance area on the video image:

51. • The larger the neural network, the higher the accuracy of object recognition.

Selecting the area of interest

1. Click the Settings button (12). The Detection settings window will open. As a result, the detection settings window opens.
   Image Modified
2. Click the Stop video button (1) in the Detection settings window to pause the playback and capture the frame of the video image.
3. Click the the Area of interest button (2) to specify the area of interest. The button is highlighted in blue.
   Image Added
   
4. On the captured
video image
5. frame, sequentially set the anchor points of the area
, the situation
6. in which
you want to analyze, by sequentially clicking the left mouse button (3). Only one area can be added
7. the objects are detected by using the mouse (3). The rest of the frame is faded. If you don't specify the area of interest, the entire frame is analyzed.
   

   Info
   title Note
   You can add only one area of interest. If you try to add a second area, the first

area will be

8. one is deleted.

After adding

9. To delete an area,

the rest of the video image will be darkened.
Image Removed


10. click the Image Added button to the right of the Area of interest button.

11. Click the OK
 button
12. button (4) to save the detector settings and return to the settings panel of the Neurocounter object.
13. Click the
Apply button (13)
14. Apply button to save the changes.

Configuring the Neurocounter module is complete.