View Source

Configuring the detection tool

To configure the Equipment detector, do the following:

Go to the Detection Tools tab.
Below the required camera, click Create… → Category: Production Safety → Equipment detector.

By default, the detection tool is enabled and set to detect people who enter or stay in the protected area without the necessary equipment and personal protective equipment (PPE).

If necessary, you can change the detection tool parameters. The list of parameters is given in the table:

Parameter

Value

Description

Object features

Record mask to archive

Yes

By default, mask is recorded to the archive (human body segmentation) (see Displaying information from a detector (mask)). To disable the parameter, select the No value

No

Record objects tracking

Yes

The metadata is recorded to the database by default. To disable the parameter, select the No value

No

Video stream

Main stream

If the camera supports multistreaming, select the stream for which detection is needed

Other

Enable

Yes

The detection tool is enabled by default. To disable the detection tool, select the No value

No

Name

Equipment detector

Enter the detection tool name or leave the default name

Decoder mode

Auto

Select a processing resource for decoding video streams. 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

CPU

GPU

HuaweiNPU

Number of frames processed per second

1

Specify the number of frames that the detection tool will process per second. The value must be in the range [0.016; 100].

To operate the detection in a "gateway" (see Examples of configuring the Equipment detector for solving typical tasks), we recommend using the standard settings of the detection tool: 1 FPS and 3 FPS for output.

In conditions of dynamically moving people, we recommend setting the delay to at least 4 FPS, and the number of frames for output to at least 6 FPS.

Type

Equipment detector

Name of the detection tool type (non-editable field)

Advanced settings

Сlassification network 1 file

By default, the following neural networks are initialized: Classification neural network (equipment and PPE on the head) and Classification neural network (equipment and PPE on the body) according to the selected processing device in the Neural network mode parameter. To initialize only one item of equipment, select the required classification neural network file.
There must be a separate classification neural network to recognize equipment on each body segment. The standard classification neural networks for different processor types are located in the C:\Program Files\Common Files\AxxonSoft\DetectorPack\NeuroSDK directory. You don't need to select the standard neural networks in this field, the system will automatically select the required one. If you use a custom neural network, specify the path to the file.

To ensure the correct operation of the neural network on Linux OS, the corresponding file must be located in the /opt/AxxonSoft/DetectorPack/NeuroSDK directory.

If you use standard classification neural network (training wasn't performed in operating conditions), we guarantee the overall accuracy of 80-95% and the percentage of false positives of 5-20%.

Сlassification network 2 file

Сlassification network 3 file

Сlassification network 4 file

Сlassification network 5 file

Mask

Yes

The parameter is disabled by default. If you want to display the human body segmentation in the preview window, select the Yes value

No

Number of measurements in a row to trigger detection

3

Specify the minimum number of frames with a violation for the detection tool to generate an event. The value must be in the range [1, 20].

To operate the detection in a "gateway", we recommend using the standard settings of the detection tool: 1 FPS and 3 FPS for output.

In conditions of dynamically moving people, we recommend setting the delay to at least 4 FPS, and the number of frames for output to at least 6 FPS.

One event per PPE element

Yes

By default, the detection tool event is generated once for each element of equipment within a track of a person. If you want the detection tool to generate an event for each equipment violation, select the No value.

An individual not wearing a helmet appears in the frame, puts on a helmet, then puts it off. If the One event per PPE element parameter is enabled, you will have one alarm event, otherwise—two.

No

Segmenting network file

By default, the Segmenting neural network is initialized according to the selected processing device in the Neural network mode parameter. The standard segmenting neural networks for different processor types are located in the C:\Program Files\Common Files\AxxonSoft\DetectorPack\NeuroSDK directory. You don't need to select the standard neural networks in this field, the system will automatically select the required one. If you use a custom neural network, specify the path to the file.

To ensure the correct operation of the neural network on Linux OS, the corresponding file must be located in the /opt/AxxonSoft/DetectorPack/NeuroSDK directory.

If you use standard segmenting neural network (training wasn't performed in operating conditions), we guarantee the overall accuracy of 80-95% and the percentage of false positives of 5-20%.

Basic settings

Min person height

0.01

Specify the minimum height and width of a person in the frame as a percentage of the frame height/width (0.15 = 15%). Objects that are smaller than the specified size won't be detected. The value must be in the range [0, 1]

Min person width

0.01

Neural network mode

CPU

Select a processor for the neural network operation (see Hardware requirements for neural analytics operation, Selecting Nvidia GPU when configuring detectors).

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 Optimizing the operation of neural analytics on GPU in Windows OS, Optimizing the operation of neural analytics on GPU in Linux OS).
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 the detection tool.
If you select Intel HDDL, due to the features of the device, only the segmenting neural network will be processed on it, the CPU will be used to run the classification neural networks.

Nvidia GPU 0

Nvidia GPU 1

Nvidia GPU 2

Nvidia GPU 3

Intel GPU

Invert Results

Invert results of default network

No

The parameter is disabled by default. To receive events about the detection of equipment that are initialized in the Classification network file parameter by default (Classification neural network (equipment and PPE on the head) and Classification neural network (equipment and PPE on the body)), select the required value from the list

OnlyHelmet

OnlySafetyVest

AllNet

Invert results of network 0

Yes

The parameter is disabled by default. To receive events about the detection of equipment specified in the Classification network file parameters, set the Yes value in the corresponding parameters Invert results of network

No

Invert results of network 1

Yes

No

Invert results of network 2

Yes

No

Invert results of network 3

Yes

No

Invert results of network 4

Yes

No

By default, the entire frame is the detection area. In the preview window, you can specify the detection areas using the anchor points (see Configuring a detection area).

For convenience of configuration, you can "freeze" the frame. Click the button. To cancel the action, click this button again.
To hide detection area, click the button. To cancel the action, click this button again.

To save the parameters of the detection tool, click the Apply button. To discard the changes, click the Cancel button.

The Equipment detector is now configured.

The Equipment detector generates an event when a person is in the frame without the necessary equipment and personal protective equipment on the specified parts of the body or when equipment and personal protective equipment is improperly applied.

Axxon One 2.0 (english) > Configuring the Equipment detector > 6.jpg
The Equipment detector recognizes equipment of the following colors:

Helmets:
1. Yellow.
2. White.
3. Blue.
4. Green.
5. Orange.
6. Black.
7. Red.
Vests:
1. Yellow.
2. Orange.

To ensure the correct reception of the E-mail notifications (see E-mail notification) after the detection tool event is generated, you must configure a separate macro with an E-mail message for each item of equipment.

Examples of configuring the Equipment detector for solving typical tasks

Typical tasks for detecting equipment and personal protective equipment:

Detection in gateway conditions. A gateway is a border, which can either be a virtual line or a door, a barrier, a turnstile. The algorithm for working in gateway conditions is as follows:
1. A person stops in front of an area where equipment or PPE is required.
2. A person poses in a way that it is possible to check the presence of all elements of equipment (the element must not be overlapped by the person, other elements of equipment, or items of clothing).
3. Screening of a person. There must be no obstacles between the person and the camera, blocking the person for screening.
Detection in a production conditions: persons move freely in the detection area.

By default, the Equipment detector is configured for detection in gateway conditions.

The recommended settings for solving typical tasks are as follows:

Parameters	Equipment detection in gateway conditions	Equipment detection in production conditions
Basic
Number of frames processed per second	1	7
Min person height	0.01	0.09
Min person width	0.01	0.03
Advanced
Number of measurements in a row to trigger detection	3	7
Mask	No	No
One event per PPE element	Yes	Yes