011 551 1765 info@activeeye.co.za
Select Page


Active Eye and IRIS AI are Revolutionising the Security Industry

AI Learning

Identification of over 9000 common objects and still learning

ANPR Verification

Automatic number-plate recognition with verification against police database of suspicious vehicles

AI Tracking

Automouos heat mapping of suspicious persons and tracking

Flagging Unusual Events

Automatic identification of unusual events and flagging thereof

Facial Recogition

Biometric identification by scanning a person’s face and matching it against a library of known faces

People Counting

Accurately counts in real time the number of people passing under a camera and in which direction

Gender and Ethnicity

Facial recognition software providing tracking of gender and ethnicity

Traits and Emotional States

Identifying and tracking of human-related traits and emotional states

Age Approximation

Determination of age approximation and tracking multiple groupings

About Active Eye & Why We are
The Best For the Job

The advances in Artificial Intelligence (AI) has allowed Active Track to combine the monitoring of CCTV cameras with AI to provide a far more powerful security surveillance model based on the concept of self-learning, not a rule-based environment which builds a base of “normal” versus “abnormal” behaviours – be it a vehicle driving the wrong way down a one-way street, or a person entering premises outside of normal hours. The Active Eye solution has a significantly lower false alarm rate, having identified the behaviours of the environment it is protecting.

Benefits of
Active Eye

  • On-site tools for officers on-site to react to an event informed with objective facts
  • The officer on-site can give real-time feedback on an event
  • Feedback sent in real time to the monitoring station in order to close an event or escalate an event
  • Make officers on site more effective by sending them to the exact locations where the anomalous behaviour has occurred and give the responding officer an image of the anomalous behaviour
  • Transforms each CCTV camera on site into a security officer, allowing for live information movement to the relevant people when anomalous behaviour occurs on a site
  • Escalates to the control room if a feed of a camera is lost for any reason
  • Significant cost saving for the end user as fewer officers are required to cover the site


CCTV is often one of the mainstays of a modern security system. Its primary focus is to act as a detection and verification system for other security measures. CCTV can be a single or combination of systems and technologies to form the overall security solution, some of these may include:

  • Visible band or infrared CCTV
  • Thermal Imaging
  • Video Analytics

CCTV is normally used to achieve one or more of the following:

  • Detect an intruder within a reasonable time frame
  • Verify an alarm from a Perimeter Intruder Detection System (PIDS)
  • Provide support to a guard or security forces
  • Provide evidence suitable for use in court

Global standards operate on the “five-minute rule”.

  • This assumes that each part of a perimeter or sensitive asset is viewed by either a guard or CCTV once every five minutes. This limits the potential time for an attack and forces an attacker to carry out a rapid attack, therefore being more likely to trigger an electronic detection system.
  • Within the South African environment, the five-minute rule is insufficient to meet our risk profile.

Efficient collection, analysis, and application of data and the intelligent use of it are becoming ever more critical in this industry. Thus, improving video intelligence appears to be an inevitable, industry-wide goal.

Detecting anomalous data, and finding ever more efficient ways to allow surveillance to shift from post-incident tracing to alerts during incidents – or even pre-incident alerts

Rationale for
Offsite CCTV Monitoring

  • Dual verification of an event
  • Informs the officer of the anticipated possible risk he/she is investigating
  • Allows for a focused response to an alarm condition
    • Allows the officer to identify a specific person or event
    • Gives the ability to determine the efficiency and effectiveness of a guard on site
  • Conventional tools available to a security company allow for limited information such as
    • The frequency of patrols
    • Areas visited and length of patrol
    • Subjective hourly radio reports
  • CCTV provides an objective view of events on site

Human Limitations in
Monitoring CCTV

  • External factors reduce the likelihood of an operator spotting suspicious behaviour:
    • Number of screens monitored
      • the more camera views are displayed, the less they are looked at
    • Method of monitoring (split screens vs single screens views)
    • Level of alertness
    • Room environment (too hot, too cold, too bright, too dim)
    • Field of view of cameras
    • Refresh rate/bandwidth issues on cameras
  • Internal factors minimising the detection of suspicious behaviour, or potential problems:
    • Limitations in observation due to divided attention and distraction
    • Limitations in recall for an event
    • Personal beliefs and biases
    • Previous experience and knowledge
    • Human recall is heavily susceptible to misinformation.
  • Staff welfare
    • Proactive CCTV viewing is a difficult cognitive task for humans to undertake for any length of time. To help ensure effective staff performance
    • Individual issues
  • Cognitive lapses/boredom
  • Fatigue
    • Directly affects alertness, involvement and motivation in the job, and the capacity to pick and respond to warning signs either within a CCTV environment or with alarms or access violations. While many common security violations may be minor, there are situations in which life-threatening conditions or those that involve substantial loss can be incurred.
  • Poor control room design, broken furniture, ineffective ventilation systems and poor supervisory oversight are all potential issues that increase the potential for fatigue.
  • Longer shifts make it more difficult for people to maintain concentration

Types of
CCTV Monitoring Systems

Live Feed Monitoring

Monitoring agents watch live CCTV feeds

  • Inherent problems with humans manning monitoring stations
  • Up to 80% of movement missed depending on the number of cameras being analysed
  • Monitoring agents must make a decision if the scenes are an anomalous – inherent weakness in the system due to human limitations
  • High Data Costs
  • Expensive hardware required

Black Screen Monitoring

Monitoring agents watch a “black screen” which changes to live feed on motion

  • Subject to many false alarms due to weather conditions and animals in the CCTV picture
  • Monitoring agents must make a decision if the scenes are an anomalous – inherent weakness in the system due to human limitations
  • Trigger events – such as passing a threshold is
  • Monitoring agents must contact the officers on site to direct them to the problem area – great potential for delay
  • Expensive hardware required




  • Analyses existing video content, which is constant
  • Detection and classification methodology of items and events
  • Rule-Based
  • Identifies and describes an object or event
  • Requires human intervention to action data received from the detection of object and event.
  • AI is far more powerful and bases itself on the concept of self-learning, not a rule-based environment, and building a base of ‘normal’ versus ‘abnormal’ behaviours
  • Lower false alarm rate, having identified the behaviours of the environment
  • Incorporates analytics
  • Develops model of patterns and behavior on a site
  • Most appropriate in a real world environment as it achieves dynamic and self-adapting video analytics
  • Makes autonomous decisions without the impact of Human Limitations if movement and objects detected are anomalous
  • Carries out the escalation procedures and site instructions without human intervention




7 Drome Road
Lyndhurst, Formain


(011) 551-1687