Skip to main content
Skip table of contents

The Face recognition analytics type. View and edit

View case

When viewing an analytics case of the Face recognition analytics type, you can choose between table and tile display modes. The tile display mode is set as the default.

In table mode, additional possible matches for the recognized object are displayed. In the tile mode, only the match with the highest confidence percentage is shown.

Both the tile and table display modes allow you to set the number of items displayed per page (15, 25, 50, and 100).

To switch display modes, select the corresponding button.

image 40 (1).png

View modes

Table display mode

In table mode, the following columns are displayed

  • Event image. Two images are displayed: the image obtained from the camera and the reference image. Select an image to enlarge it.

  • Event name. Displays one of the following values:

    • Face recognized

    • Face unrecognized

  • Resource name/group. The name of the resource that serves as the reference image and the name of the group in which the resource is located.

  • Camera. The name of the camera that captured the event. Clicking on the name displays a video player for viewing the archive at the time of the event.

  • Similarity %. The percentage of confidence in recognizing the reference and obtained images.

  • Created. The date and time of the event.

  • The Plus icon to add the resulting image to the resource group as a reference image.

Group 22 (10).png

Table mode

Tile display mode

In the tile display mode, each event contains the following information:

Tile display mode

  • Two images: the image obtained from the camera and the reference image. Click either image to enlarge it.

  • The name of the resource that serves as the reference image.

  • The name of the group in which the resource is located.

  • Event name.

  • Camera name: the name of the camera that captured the event. Clicking on the name displays a video player for viewing the archive at the time of the event.

  • Date and time of the event.

  • Similarity percentage in recognizing the reference and obtained images:

    • 0-20% – red

    • 21-99% – orange

    • 100% – green

When you hover over an event, a button for adding the resulting image as a reference image is displayed.

Adding an image as a reference

You can add an image taken from a camera to a resource group as a reference image.

To do this, select the Add button.

image 40 (2).png

The Add button

211.png

The Add button in tile mode

As a result, the addition form will be displayed on the screen.

212.png

Adding an image as a reference

Fill the form. Specify which resource group the image should be placed in. To do this, click Choose group and select a group from the list.

If necessary, you can create a new resource group. To do this, click Create group and enter the name of the group in the field below.

After filling out the form, click Save resource.

Filters

A set of filters is available:

Filters

  • Cameras – only events which were recorded by the selected cameras are displayed.

  • Period – only events that occurred during the specified period are displayed.

  • Groups – displays only events in which the selected resource groups were used for facial recognition.

  • Resources – only displays events in which the selected resources were used for comparison.

  • Events – only certain events are displayed (person identified, person not identified).

  • Reliability – only events with a percentage of reliability that fall within the specified limit are displayed.

To enlarge an image for detailed viewing, click on the image. Both the camera image and the analytics resource image will be displayed.

The Similarity percentage filter is used when there are multiple possible matches for the recognized object. You can fill in only one filter field (with a value not exceeding 100%) to set the lower or upper limit for filtering.

By clicking on the camera name in any display mode, you can access the camera archive.

Editing the case

To edit the case, click the Settings button on the analytics case viewing page.

To edit a case, select the step and make changes.

You can edit all previously specified settings, except for changing the type of analytics in step 1.

After finishing editing the case, click Save.

Editing the case with the Face recognition analytics type

If, when editing a case, a camera with existing events is deleted, then you must select a method for deleting the camera:

Deleting the camera

  • Delete with saving events.

  • Delete camera and its events.

When you select the Delete with saving events option, all events that were recorded by this camera will be displayed in the analytics case. In this case, events will be taken into account when generating a report, when viewing the list of events, a transition to viewing the archive at the time of the event will be available, etc.

If you select the Delete camera and its events option, all events that were recorded by this camera will be completely deleted.

Generate a report

To generate a report, select Generate report.

As a result, the report setup form will be displayed:

196.png

Setting up the report

Specify the period for which the report will be generated.

Specify for which cameras the report will be generated.

Click Submit.

Once the report is generated, a message with a link to the report will be displayed. Click to download the report in XLXS format.

image 40 (3).png

A generated report view

Stream settings

Selecting the Stream settings button opens the Analytics case status panel that contains additional information about the current case.

image 40 (4).png

The Analytics case status panel

The following details for each added to the case cameras are available:

  • A status of an analytics case.

  • UUID of an analytics stream

At the left bottom, you can change the status of the analytics case. If changed, save the changes by selecting Save.

JavaScript errors detected

Please note, these errors can depend on your browser setup.

If this problem persists, please contact our support.