What are the latest trends in perimeter security technology?

With an estimated million drones entering the world's airspace each month, drone detection has become a hot topic in the discussion around perimeter protection. Some are used as flying security cameras, as part of a coordinated security installation, but others might be used for malicious intent. So it’s important to know why a drone is there, and be able to trace where it’s going, to ensure its purpose is legal and coordinated. Perimeter protection typically relies on a multi-layered approach that involves a variety of different technologies, including cameras, drones, radars, LiDARs, facial recognition technology and other standalone systems. To ensure that these systems do not work in unconnected siloes, it’s important to put in place a unified approach to cross qualify incidents and intrusions alerts. Connecting these systems into a VMS system also ensures that alarms and incident are immediately correlated to a video feed to avoid false alarms.

What’s really important in perimeter security is the minimisation of false alarms, not simply the potential detection of what might be an unauthorised person or object. In light of that, many systems now include alarm validation that can confirm an alarm event using a camera. The utilisation of AI-based technologies can further validate the accuracy of the alarm, making it as accurate and precise as possible. I anticipate seeing more cross-technological integrations to reduce false alarms, so that personnel in an alarm centre spend as little time as possible in validating an alarm.

The adoption of accelerometer-based technology on fence monitoring systems is really gaining traction, due to its ability to overcome the age-old false alarm issues of other fence monitoring technologies. Detection accuracy is now down to 3 metres and sensitivity settings are quite broad to provide better false alarm discrimination. The beauty of this type of system is its ability to be deployed on all kinds of fence systems and wall supports. More important, though, is the move to ensure that the systems are IP compatible and provide connectivity options to the larger security network via API’s, Modbus, or IP Alarm transmission. Equally important is the need to integrate with VMS systems, such as Milestone, Genetec, Qognify, Advancvis etc. One such system on the market is the G-Fence line from Sorhea and Protec - the G-Fence 2400 being the newest addition to the range.

In the case of vehicle-ramming attacks, there are straightforward security solutions to protect patrons. Physical architecture such as bollards, barriers and barricades can stop a vehicle from entering a high-traffic area. There are also trends to stop vehicle attacks beyond main entrance gates. The first is in a location where the primary use is pedestrian but, frequently, vehicles need to pass through. An example could be a square in which the maintenance truck comes through to clean the square periodically. Such access points are well served by bollards, some moveable that go up and down to let vehicles through and others that are fixed. The other access locations are those that are temporary. Sections normally open to traffic will be closed to create pedestrian paths and gathering points. These locations are best protected with crash-rated portable barriers that erect in 15 minutes and then removed after the event.

As the pace of physical security innovation increases, the cost of perimeter security is coming down quickly. Some technologies such as Thermal Detection were previously unaffordable for the average business owner, but have been positively impacted by these innovations. Integration of these sensors with a verified intrusion system, low-light capable cameras, and a full-featured video management system, ensure that detection and notification of intrusions are reliably reported with very low false alarm rates. Another interesting angle to these innovations is the proactive automated response to intrusions. One way of deterring intrusion is to provide an automated audio response to an intrusion event. Further, some business estates choose to establish multiple perimeters with escalating response scenarios. An auto dealership may warn people with an audio response to leave the premises after hours (without offending potential shoppers!), if intruders advance to a second perimeter, a sterner message can also be added.

Like many elements of security, perimeter technology can be enhanced with a multi-layered approach. Video analytics is one such tool that can be used as a standalone solution, or in conjunction with other technologies. With the help of AI and computer vision, security teams can not only predict, prevent, detect and respond, but they will also have access to deeper analysis for what is happening, enabling a more proactive and appropriate response. Detecting the intrusion of an individual is one thing, but the response might be escalated if that individual has a weapon in hand. Computer vision simply adds another layer of information that can be analysed for more detailed situational awareness, ultimately resulting in a more targeted approach to perimeter security.

For many years, intercoms were used for room-to-room communication, in place of a doorbell or at a gate system. However, the capabilities of an IP video intercom system far exceed past technological functions. An IP system can easily integrate with cameras and a network video recorder, allowing the intercom to do a larger job, including offering more features and third-party streaming to an NVR. What customers are requesting these days, in terms of enhanced perimeter security, is the ability to have remote monitoring, mobile access and the ability to audit when an access code was used, where it was used at any time. The ability to access historical video footage is vital to many businesses in ensuring holistic security is being maintained. Today, this can be achieved by a video intercom.

We are seeing a steady increase in the deployment of face recognition at perimeters. The advancements recently made in AI and algorithm efficiency, like those introduced by SAFR in 2020 and 2021, have delivered amazing new levels of accuracy, speed, and stability across skin tones. Lightweight facial recognition software can now be embedded in low-cost edge hardware with minimal memory and power demands. By pushing more processing to the edge, such as performing video processing and detection tasks inside a camera directly, the total cost of ownership (TCO) of surveillance hardware is substantially lowered. Users have the option of using local servers for operations, or only performing the initial processing function onsite, while benefiting from cloud hosting for all other processes. There is new demand for touchless access control with face recognition as an option, which delivers a much friendlier and less costly alternative to key cards or fingerprints.