Behavioural Analytics

Recognize potentially dangerous situations by detecting specific human postures.

A behavioural video analytics system can generally be categorized into five functional categories: violence detection, suspicious activity, crowd behaviour, traffic monitoring, and environmental and personal safety, such as fire detection. A more advanced analytics can recognize the interactions and contexts relevant to diverse objects by building analytics this way.Analysing behaviour automatically detects the differences between the two situations – without any manual calibration required.

As part of any new or upgraded video system, intelligent behavioural analytics requires careful consideration to identify current and future needs, ensure that existing and future video system investments are leveraged, and ensure effectiveness of the solution.

Key Features

• New DL skeleton tracker provides 18 body points of interest.
• Hand object detects objects being held in hands
• Deep Learning People Tracker (DLPT)

• Aggressive behaviour detects both aggressive and Fighting.
• Fall detects people in a vertical position.

Other Features

• Repeatedly Rule: generates an event when an input rule is triggered a set number of times within a defined period.
  This allows for rule configurations such as, trigger event when a person enters a zone three times in four seconds.
• Counter Reset: allows a rule to trigger the counter to reset to 0.
• UI support for SSL and Certificates: allowing for a self managed end to end encrypted connection between your
  browser and the back end services.
• ONVIF Events: Events have now been added to our ONVIF integration allowing for events triggered by our rules to be consumed by external applications.

Behavioral Recognition Understanding Technology