The idea of perimeter protection is nothing new. Even in ancient times, civilizations erected walls and other obstacles to withstand invaders at their borders. What is new is how modern perimeter defense systems are using cutting edge, sensor-based technology to warn of a threat or intrusion and provide data that can help organizations mount an appropriate response before harm can be done.

A multi-technology approach to perimeter protection 

Many successful perimeter security solutions integrate a combination of physical barriers, advanced detection devices, video surveillance technology, and analytics. This strategy allows each layer to bring unique protections to the solution.

• Physical barriers: These barriers might be fencing or cement panels. You can augment them with anti-climbing devices like barbed wire or high-voltage current. While they might not stop an intrusion, they can certainly delay it.
• Cable-based sensors: These sensors are usually buried, or fence mounted. Because you don’t have to deploy them in a straight line, they can cover around corners and dead ground areas. While they can alert you to a possible intrusion, they can’t provide actionable information as to the number of intruders or other details you might need to prepare a suitable response. In addition, animals, moving plants and trees, and severe weather conditions tend to trigger false alarms.
• Other intrusion detection sensors: These would be devices such as radar. Not only can they capture the GPS location of an object or person, they can also determine the rate of speed and the direction of movement, which makes it possible to accurately calculate the point of contact. They operate 24/7 and are virtually unaffected by common triggers like moving shadows or light beams, small animals or insects and adverse weather conditions. To reduce interference with other electronic devices in the environment, you might need to change the radar’s frequency or lower its power, which would hamper the effective range of the device.
• Video-based technology: Adding video cameras to perimeter defense allows you to visually verify alerts in real time. When integrated with strategically placed sensors, cameras with pan/tilt/zoom capabilities can automatically pivot to an incident location, track the target, and capture forensic evidence for post-incident investigation. Depending on local laws, you can use video cameras to monitor beyond your physical perimeter, providing an additional surveillance buffer and more time for your security team to mount a response.

Visible light and thermal cameras: Not an either/or solution

Selecting the right visual camera is critical to 24/7 perimeter defense. While traditional visible light surveillance cameras operate in natural or augmented lighting conditions, their effectiveness drops precipitously in inclement weather and is virtually nil in complete darkness. That’s why it’s important to also integrate thermal cameras into your perimeter protection solution. In optimal light, visible light cameras do well in capturing forensic detail and tracking targets. What thermal cameras bring to the table is the ability to detect heat signatures of people, animals, and objects in total darkness, hidden in the shadows, or wearing camouflage to conceal their approach. Unlike visible light cameras, they can even detect intruders in extreme weather conditions. With the addition of analytics, some thermal cameras can also distinguish between types of intrusion targets and alert security based on a preset list of conditions such as the direction and speed of a person or vehicle.

The advantage of edge-based video analytics 

Adding video analytics to perimeter security further enhances intruder detection. In addition to basic motion detection, there are analytics that can discern moving objects, count people, read license plates, expose loitering, reveal people and objects crossing into specific zones, and much more.

Besides heightened awareness, video analytics can benefit your organization in other ways. For instance, analytics can direct the camera to only record video that contains activity. By allowing the camera to process the video — known as intelligence at the edge — you can significantly reduce the load on your network because the cameras will only stream relevant video. You’ll also preserve critical forensic details because the analysis is performed on uncompressed video.

Relying on the camera to extract relevant data also increases security operator efficiency since it decreases the amount of video they need to review when they receive an alert. It also eliminates the need for dedicated analytic servers, significantly lowering your storage costs. And it optimizes server resources by reducing the processing load on each server so they can handle more video streams.

Today’s smaller and faster in-camera processors are helping to drive this switch from server-based analytics to edge-based versions. Software developers can harness this power to provide new levels of intelligence and analytic opportunities never seen before on the edge.

The impact of deep learning and artificial intelligence (AI)

That leap in analytical performance at the edge is being accelerated further by Deep Learning Processing Units (DLPU) embedded in new network cameras. We’re already seeing developers building on AI and machine learning to train deep learning algorithms to accurately detect, recognize, and classify objects and people.

What does this mean for perimeter protection solutions? Once the technology is perfected, you’ll be able to deploy camera-based analytics capable of distinguishing differences in clothing colors, objects an intruder might be carrying, and other granular details with a high degree of accuracy. Advanced deep learning analytics could potentially lead to highly targeted detection systems capable of identifying and differentiating between employees, customers, members of the public, or potential threats. You’ll be able to set up intrusion detection solutions on the perimeter that only alert on very specific objects and conditions — an advanced version of If-This-Then-This (ITYTT) decision trees.

From a security perspective, applying these advanced deep learning analytics to your multi-layered perimeter protection solution will inevitably result in a more efficient and accurate system of detection and deterrence.

Automating soft responses to intruders

In defending one’s perimeter, deterrence is always preferable to confrontation. Often making the intruder aware that they’ve been detected is sufficient to cause them to flee. Integrating automated soft response solutions can, in many cases, forestall the need for human intervention in an intrusion alert. For example, the alert might trigger floodlights, an illuminated sign stating that the area is under surveillance, or a speaker system delivering a warning to leave the vicinity. Only if any of those actions failed to achieve the desired response would security need to intervene in person. This strategy reduces the burden on security staff, freeing them to focus their resources on more urgent events.

The best defense is an integrated defense

As threats continue to evolve, so must the countermeasures organizations put in place to mitigate them. Deploying an integrated, multi-technology approach to perimeter defense gives organizations the agility to optimize protection for their business, people and property.