There are two types of monitoring solutions: Rules based monitoring and Self-learning behavioural analytics monitoring.

Rules based monitoring:

  • Requires specific rules to be set to predetermine what events you want a controller to view and when. E.g. you set half the camera over a perimeter fence to alert on movement and the other half not to. Only when the movement occurs in the predefined no-go area does the controller get alerts, but in reality you would want to be able to pick up on abnormalities as early in the frame as is possible i.e. as the human enters the frame, what if the threat doesn’t move into the predetermined no go zone for 5-10 minutes but loiters in the other half unnoticed?? Without predefining this no-go zone you open the camera up to flooding of unnecessary false alarms where every movement in the frame is flagged.
  • Further to the above, if a cat, bird or leaves of a tree continuously moves in this predefined no-go zones rules based system does not distinguish that these are not a threat and will alert on this “abnormal behavior continuously” with no way of telling the system that these alerts are not necessary or a threat.

Self-learning behavioural analytics monitoring:

  • Self-learning behavioural analytics software differs greatly in that it utilises a self-learning algorithm which requires no human intervention or rules to be set to ensure alerts that are a threat are alerted in the control room. Rather this type of software learns camera’s specific environment and flags behavioural abnormalities based on the specific environment, this allows our control room to pick up security incidents as they happen making our solution truly proactive when dealing with suspicious and/or criminal activity and ensuring false alarms are greatly minimized. This type of software is always learning and adapting to changing environments where it has been deployed without needing to set any rules. This ensures that the site is ALWAYS being monitored effectively.

System functionality FAQ's

Question 1: How does the camera on site communicate with the control room?

Answer: The camera onsite communicates with the control room by plugging the onsite NVR into our Smart analytics server via an RTSP stream (real time streaming protocol). This in turn is plugged into a router utilising the onsite existing internet connectivity, and feeding through a VPN (virtual Private network) into the control room, all alerts are presented through to the controllers within milliseconds of it being flagged as abnormal on-site.


Question 2: How does the software work?

Answer: The software utilises a self-learning behavioural analytics platform that learns the environment and flags abnormalities, this allows our control room to pick security incidents as they happen making our solution truly proactive when dealing with suspicious and/or criminal activity. The solution acts as a cost effective guard replacement alternative. Our detection technology integrates into most existing CCTV cameras for minimal hardware requirement and capital outlay ensuring maximum return on your investment.


Question 3: Who responds when there is an incident?

Answer: If the client has an active armed response service provider Urban Secure will alert the clients armed response service provider. If the client does not have an armed response service provider Urban Secure has a network of 150-armed response service providers (best companies in all areas) that we will recommend based on your location and budget.


Question 4: How does the incident reporting work?

Answer: The second and positive incident is picked up by a controller, it is then escalated to a high-level supervisor where the situation is dealt with immediately according to the site specific SOP (standard operation procedure). Once the incident is closed the client receives all video as well as a detailed report of exactly what occurred and how the incident was handled.

Leave a Comment

You must be logged in to post a comment.