Cognitive

Advanced behaviour analytics for security companies

This artificial mind engine can tap into, mine through and understand large, complex data streams to extract and compile important insights that cannot immediately be detected by data analysts


Allows security companies to deliver real-time, location-aware processing, analysis and predictions on constantly changing big data streams in motion, irrespective of variety, velocity and volume


Cognitive is a data agnostic distributed artificial intelligence engine

Examples of behaviour analytics

  • Gives security companies the ability to analyse human behaviour from data
  • The data could be, as an example, video streams or sales transactions. Cognitive has the ability to detect trends and then make future predictions, or to detect unusual behaviours.
  • Monitoring and predicting student crowd behaviour during protest action to detect unruly behaviour and congestion issues.
  • Monitoring and predicting peculiarities in long queues, identifying tendencies to recognise unsocial behaviour and the overcrowding of buildings
  • Detecting, monitoring, and tracking individuals who are hanging around a neighbourhood road, and picked up by a number of cameras. If this is not a trend, Cognitive recognises this and alerts the security company.

Queueing: Dealing with increase in people - video analytics

Benefits of intelligent queue monitoring video analytics:

1.Continuous counting of the number of people arriving and queuing up

2. Cognitive monitors the normal establishment, build up, and ending of a queue of people/passengers, as well as the overall interpersonal dynamics within such a queue

3.Cognitive then captures and extracts any unusual behaviour in the queue such as pushing, breaking up of the queue or any sudden deviation of the usual dynamics within the queue

4. Continuously measures the length of the queue and reports back if the queue becomes too long

5. Cognitive keeps track of individuals as they approach and join the back of a queue, merge and move with the queue as it progresses forward

Tracking for security companies People tracking in a neighbourhood or shopping Centre - Video Analytics

  • The ability to isolate individuals from one another and from their surroundings and track/follow them across a single video screen and from one video screen to another. Cognitive can track multiple individuals as well
  • Observe and determine if people are loitering in a neighbourhood, outside a house, street corner or a particular shop in a shopping mall
  • Track individuals from entrance to a street or mall, and allocate the correct facial template to each individual that is being followed (if facial recognition cameras are used).
  • If any behaviour is different to the normal trend, it can be detected by Cognitive and reported as suspicious
  • If a person behaves normally, all recorded information on the individual will be deleted on exit. This is time saving and dramatically improves the ability to analyse only relevant aspects of video analytics. With this process the behaviour of every individual can be monitored and tracked as he/she moves through an area.

Diagnostics: Thermal Imaging for security companies

  • The ability to detect and identify people who are in good heath and those who might be suffering from an illness
  • Drug trafficking – research shows that body temperature rises with alien substances within the body either swallowed or within cavities.
  • Thermal imaging does not need light to operate. It produces a clear image in the darkest of nights and in most weather conditions.
  • Potential intruders that are hiding in the shadows are seamlessly detected by thermal imaging cameras.
  • Thermal contrast is almost impossible to mask.

Trafficking:

Child trafficking – Video Analytics

  • Cognitive has the ability to monitor the overall interpersonal dynamics between an adult and a child within a queue
  • It finds patterns in the behaviour of a number of individuals in the same timeframe to determine if any subversive trafficking group activities are taking place, or about to take place.
  • Whilst tracking an identified adult with a child within the environment, check if there is a sudden increase in the speed of movement of more than one individual and in which direction the movement is taking place.
  • Incorporate thermal image analysis to indicate excessive sweating of an adult with achild

Assisting security companies with vehicle theft from parking lots – human/vehicle video analytics and issuing tickets

Cognitive can classify the model of a vehicle, has number plate recognition as well as facial recognition when issuing ticket at the entrance to a parking lot. It creates a 3-tier package: 1.Vehicle, 2.Nr plate, 3.Person, 4. Ticket)

Vehicle classification, number plate recognition and facial recognition when presenting a ticket at the exit of a parking lot

Incorporate facial recognition at ticket vending machines (pay point) to highlight anomalies for same-person / not-same-person paying for issued ticket (parking entrance)

Initiate person tracking feature if not-same-person paying as well as behaviour analysis

Biometrics for security companies Facial Recognition specific solutions

Facial recognition, using static and video streams, linked to:

  • Cognitive incorporates a high risk person list into the system
  • Personnel database, SOP’s, access control and other sensors
  • Front desk – Person of Interest/ Customer information and profile
  • Continuous optimisation and improvement of models with new data

Cognitive’s benefits to security companies

Real time early warnings helping key role players take corrective measures at the right time

Continuous monitoring and integration of numerous streaming data

Management of high volume transactions

Automated actions on risks and opportunities to help reduce operational overload

When should your security company use Cognitive

  • Experiencing theft
  • False alarms or alerts
  • Inefficient tracking and monitoring of important assets
  • Difficulty in monitoring and identifying risks prior to them becoming threats
  • Leveraging information to gather intelligence on business operations, the market and competitors
  • Transforming data into usable intelligence which can be shared and leveraged across the organisation in real-time
  • Identifying and exposing abstract relationships