SecurityWorldMarket

10-08-2021

Perimeter security – Part 2 of 6

Perimeter security: current and future trends

Some of the future trends in perimeter security will be more advanced video analytics, drone detection solutions and more collaboration between technologies.

“A lot of emphasis is being placed upon cameras and utilising sensing technologies – developments in combination sensors are going to be key”, says Mark Cosgrave, Division Manager for Western Europe at Optex.

According to Marketsandmarkets, traditional perimeter security platforms have increasingly been replaced by more advanced perimeter security platforms using artificial intelligence, machine learning and facial recognition. Video analytics solutions combined with facial recognition, object tracking and license plate recognition have become critical for national border and facility protection. Integrating these technologies can sometimes be a challenge but will also increase the perimeter security. For example, if the access control and intrusion alarm systems are linked, the access control system can be programmed to lock down a facility based on the type of alarm that sounds once the system determines an intruder.

Improved video analytics

Andries Nouwens, Director Sales EMEA, Business Unit Video Systems and Solu-tions, at Bosch Security Systems.
Andries Nouwens, Director Sales EMEA, Business Unit Video Systems and Solu-tions, at Bosch Security Systems.

Andries Nouwens stresses that video analytics enables cameras to understand what they are seeing and create an alert if there are threats the moment they happen. He says: “It also makes it possible to retrieve the right footage from hours of stored video instantly and analyse the scene by providing different statistics so an operator can act faster and more efficiently to potential threats.”

Andries Nouwens also points out that video analytics solutions today have the ability to perform video content analysis over large distances and are able to differentiate between genuine security events and false triggers such as snow, trees moving in the wind, rain, hail and water reflections that can make video data more difficult to interpret.

He exemplifies with Camera Trainer, a machine learning technology from Bosch: “The camera retains information on new user-defined objects and situations, or any subsequent changes. It then refers to this new knowledge when processing scenes and can combine it with pre-determined alarm rules and object filters for even more accuracy and flexibility” he says.

Cameras can be taught to recognise and detect stationary objects or certain situations instead of being triggered by motion alone.

“With this, video metadata is further enriched since the technology can deliver data when objects are present, but also once they have been removed, therefore providing more informative data”, he says.

Mark Cosgrave, Division Manager for Western Europe at Optex.
Mark Cosgrave, Division Manager for Western Europe at Optex.

Mark Cosgrave also believes video analytics strengthened by machine learning has become more suitable for perimeter security: “It has gone through evolution, there are improvements in reliability and performance”, he says.

Mark Cosgrave thinks video analytics will continue to grow and Optex has now moved into what the company calls sensor driven verification.

“There is no one solution, when we walk and move around in our environment, we are using not just our sight, we are also using our balance, our taste, our hearing and speech. It is necessary to deliver high reliability and that comes with an input of sensors. It can be camera sensors, thermal sensors, laser or radar. You need those sensor inputs to deliver a high-quality reliable solution”, he says.

Drone detection

Both Andries Nouwens and Mark Cosgrave mention drone detection as an upcoming trend within perimeter security. They can minimise costs of disruptions.

“Drones are a big problem for a lot of people at high end utilities, prisons and airports”, says Mark Cosgrave.

Andries Nouwens highlights that it is possible to seamlessly track persons and objects across multiple cameras visualised on a geographical map. For example, with the Genetec Restricted Security Area Solution, cameras featuring built-in AI, can generate information on an object’s location, speed and trajectory. Moving targets will be classified as a person, car, or truck and intuitively displayed on a map so users can assess and respond to threats in less time.

“A moving camera can be triggered at the same time and steer towards the moving object and follow the person with intelligent tracking. Users can follow the object over multiple cameras in real time. When the object moves to other cameras, the RSA solution does an automatic handover and shows the video footage of the next camera,” Andries Nouwens says.

Future prospects

Mark Cosgrave states that people increasingly have started to take their outside security need more seriously. He says: “It makes sense to have security outside because you are able to hopefully prevent somebody getting into the premises. If you have external sensors, you have that ability to be able to get early warnings”, he says.

Andries Nouwens thinks artificial intelligence, like video analytics, will gain importance to further improve security and uncover business opportunity beyond.

He says: “This follows the overall trend of moving towards datadriven solutions. Future solutions will increasingly possess learning capabilities to more easily cater to specific customer requirements or changing situations.”

Note: This editorial article has primarily been produced for the security trade magazine Detektor in collaboration with Securityworldmarket.com.



Leverandører
Tilbage til toppen