A lot of money is spent on buying cameras, recorders, installation and so on in the hope that it will improve security, and it does. However, the use of intelligent systems can reduce the capital layout of systems and, at the same time, improve the security measures.
This is achieved by systems that introduce intelligence, emulating human intelligence, into the mix. CCTV applications can be divided into a number of main classes and these require different approaches to achieving the required goals. In street surveillance the objective would be to isolate potential or active incidents amongst a very busy background. Site surveillance often have fairly quiet backgrounds whereas office security is more about access control than observing behaviour. When designing a security system it is imperative that the responsible security officer is part of the design and that the site’s own peculiar characteristics are taken into account.
Most security operations can be broken down into monitoring, incident identification, incident management, post-event analysis and prosecution actions. For each of the above classes the emphasis will be different on the different actions.
What is an intelligent system? Intelligent systems automate tasks, they operate without or with minimal human intervention and they can perform intelligent interpretation of events automatically and in context with the present scenario. Furthermore, intelligent systems can perform mundane tasks without supervision such as searching intelligently through video material for objects, incidents or even actions.
Different aspects of intelligent systems
Monitoring: Obviously monitoring can be achieved with a variety of sensors and these could be intelligent as well. From a CCTV perspective intelligent systems would employ video analytics as a first port-of-call. These systems would strive to improve detection and in particular automated detection. This will assist human operators by taking over the mundane and boring tasks of continuous monitoring.
Modern image processing techniques can support this function successfully by offering both behaviour detection (motion and tracking analysis) as well as specific object detection (facial, ANPR, shape recognition). In addition, analysis of detected events can, by using sophisticated analysis and cross-correlation techniques, significantly reduce false alarms. Analysis of historic events and time-line analysis can create pro-active trends and contextual analysis can be used to create priorities in detected events.
Intelligent analysis however is not restricted to video events but can also take into account any event detected by other detection systems such as perimeter detection, access control and so on.
Identification: The identification of an event can be achieved by specific recognition tools to recognise faces or vehicles at a specific location but can also be the result of behavioural analysis such a loitering, running and so on.
Incident management: This task is often seen as exclusively a human task. However, intelligent systems can perform a supportive role in this by assisting in classifying objects (face, ANPR identification), by enhancing visual views such as image sharpening, noise reduction, equalisation and so on and by offering, at their fingertips, access to relevant historical information such as recent recordings, prior analysis of events and so on.
Post-event analysis: While it is the security officer’s duty to execute this function, intelligent systems can assist greatly by providing rapid historical analysis of data such as retrieving specific faces out of video footage, analysing behavioural trends in video footage and so on. Indeed any action performed on live video can be performed on recorded video with the same success ratio provided recording was done at the same performance level as the source camera can provide.
Intelligent system requirements: In order to perform intelligent analysis a number of features need to exist in a system. These include the basic sensors (cameras) but with sufficient resolution and picture quality to extract information. In addition identification tools such as ANPR, facial recognition and specific shape identification tools greatly assist in creating objects of interest (OOI).
Intelligent automated detection and intelligent analysis: Automated detection is, of course, available in just about any detection device from PIRs to IR beams etc. Video analytics can offer simplistic motion detection but intelligence comes when detected events are filtered to reduce false/nuisance alarms and events are classified according to specific needs. Thus intelligent analysis will take the output of a video analytic detector and the OOIs and process them in context with the environment. In this way slow moving objects in the context of fast moving objects will be detected and vice versa. The period that an object has remained stationary can be detected as well as the speed at which it is moving. Similarly, behavioural aspects such as loitering, running and so on can be determined. What all of these have in common is that they are executed using time-line analysis of events and not just a single occurrence. Smart systems will do this even with multiple sources and perform combined analysis to corroborate a single sensor’s findings by other sensors’ outputs.
Unfortunately intelligent systems cannot easily come in a shrink-wrap format. Just like one has to train staff at a particular site for the particular requirements, intelligent systems have to be applied with consultation with the security needs to provide optimum solutions. While moves are afoot to automate even this process by the introduction of learning functionality, this is still not at a very advanced stage.
In summary we believe that intelligent systems will be the future of security and smart users will start early to understand their own needs and employ such systems to assist with this process to not only reduce crime but to eradicate it totally. Learning of your own requirements, your own needs and system configuration will take time, but in the long run will be worth it both in improved security provision and cost saving.
Dr Coetzer can be contacted at Protoclea Advanced Image Engineering ( benniec@protoclea.com or www.facebook.com/protoclea).
© Technews Publishing (Pty) Ltd. | All Rights Reserved.
printer friendly version