
Integrate, as the dictionary puts it, is “to bring together or incorporate parts into a whole”. An integrated security system is therefore a whole system composed of several sub-systems, each providing a certain function to fulfill the operational needs of the entire security system.
The 4 D’s of Security Operational Readiness
An efficient integrated security system must satisfy 4 basic functions that constitute what is called “Security Operational Readiness”:
Deter
As we all know, it always makes sense to keep an intruder out rather than to confront him when he has already set foot into the premises. Sight of obstacles such as a 3m high fence with concertina wire and of CCTV monitoring equipment are likely to discourage intruders and keep them away.
Detect
Nevertheless, attempts to scale the fence or gain access to the sterile zone, must be detected by intrusion sensors that could be of various technologies such as fence disturbance sensors, microwave barriers, infra-red beams or underground electromagnetic field cables. Several sensor sub-systems are normally deployed to work together for two reasons: firstly to allow mutual back up in case one fails, and secondly, to overcome the usual problem related to nuisance alarms caused by extraneous elements such as wind, trash, animals, etc. This is why the Integrated Platform in the Control Center will be configured so as to sound a major alarm only when a pre-determined sequence of alarms is received from these sub-systems. Once a major alarm is sounded in the CC, videos from the alarm zone will be displayed on monitors, recorded, and logged on printers for future analysis.
Delay
Attempting to scale the perimeter and engaging the concertina wire would impose on the intruder a delay dictated only by his skill by the means he has at his disposal. However, detection of his intrusion attempt would have already taken place by the sensor sub-systems and the security forces will have been already dispatched to the scene.
Deny
Ultimately, the security forces would confront the intruder and deny his access to the premises.
Clearly, early detection is paramount to the efficiency of Operation Readiness. However, there are three issues of concern with intrusion detection. Firstly there is the problem of nuisance alarms aforementioned. Secondly, the deployment of several detection sub-systems complicates the integration design and adds to the overall system cost. Thirdly, the intrusion detection sub-systems normally work over specific zone lengths, typically 100m-300m (although there are sub-systems that claim to use localization techniques with accuracies of 3m-30m). The longer the zone lengths, the more difficult it is to manipulate the CCTV cameras to handle and record the alarm scene. Several cameras would have to be deployed simultaneously since no one camera could adequately cover the entire zone.
The Solution: Advent of the Information Technology
Throughout civilization, there have been specific milestones of progress that had profound influence on the course of human life. Examples include the invention of the wheel, discovery of penicillin and invention of gun powder etc. Today we are living in an era marked by the Information Technology with ever far reaching effects. Almost every analogue value in life is now being digitized and transformed into mathematical combinations. It is now possible to transform video images into trillions of digits, called “pixels” every group of which is processed and analyzed separately, then configured to pre-determined patterns. This has lead to the advent of Video Content Analysis technology.
Video Content Analysis
Also called Object Video Analysis or Video Object Detection or simply Video Analytics, this technology transforms the video image to a number of megapixels, typically 2, 4, 8, 16 and very soon 32. Each specific group of pixels is programmed to represent a certain aspect of the image and is given a meaning. Sets of rules are entered to dictate what those aspects represent. For example, a group of 50 pixels can be simulated to approximate the shape of a man standing. Then rules are set to check his number, speed, direction, presence, etc. we train the system to respond to each and every aspect differently. Consequently, we end up with a very powerful tool that not only recognizes the presence of the target but also his behavior. For example, a man could be seen walking along a fence, but as soon as he starts to climb it or crosses a virtual line an alarm will trigger. Alarms could be set to sound if a piece of luggage was left behind or an object is removed from its place, or when the number of cars or people exceeds a certain value at a certain time and date. The applications are only limited by the imagination.
Video Analytics as Applied to Intrusion Detection
When we apply this technology to the perimeter intrusion detection, we see immediately the advantages:
1- Simpler design as there will be no other sub-systems to integrate.
2- Substantial saving on the cost of the other sub-systems and their related infra structure.
3- Far less nuisance alarms as the system becomes relative to the behavior of the target rather than to the target itself.
Already designers have recognized this advantage and are now pushing the technology still further…right to the camera at the edge where video analytics and recording are now carried out in the camera itself with high definition. Furthermore, some cameras are now equipped with GPS targeting techniques. They do not only detect intruders but also pinpoint their locations to a remarkable accuracy.
Conclusion
In my opinion, as video analytics becomes more and more intelligent, it will eventually replace all other detector sub-systems as means of intrusion detection.
Biography
Ousama S Kabbani is an experienced veteran of the security world. He is an electronic engineer and has for many years been working extensively throughout the Middle East. He currently holds the position of General Manager of the Industrial Security & Safety Division (ISSD) of Abdulla Fouad Holding Co. in Saudi Arabia.