dormakaba News, Innovation and Digitization, Trade Fairs
dormakaba presents technology innovation of an AI based anti-tailgating solution at BAU 2023
Tuesday, 18 April 2023
Interview with Marc Faresse: Marc is an expert in AI and Video Analytics within the Group Innovation Management Team at dormakaba. With his many years of experience, he gained within and outside the company, he is helping dormakaba to introduce AI-based technologies like face recognition into its offering.
What is tailgating and what challenges does it present?
Tailgating, the passage of an unauthorized person behind authorized personnel, is one of the most common physical security breaches. Also known as tailgating often results from a random act of kindness such as holding the door to a stranger. It can be seamless and a lot less suspicious to follow an authorized person rather than breaking into a building. Those with criminal intentions are well aware of this.
In high-traffic settings like large facilities, companies, or residential complexes, the risk of tailgating is higher. Tailgating can expose people and companies to perils like loss of revenue or goods, reputation, and even physical danger. Hence, it’s not a surprise that tailgating is among the top three security concerns of Fortune 1000 companies.
How is it possible to mitigate the risks of tailgating?
One of the most efficient ways of mitigating tailgating risks is to install appropriate access control systems and manage them methodically — a suitable solution as such is turnstiles.
As the entrance control method of choice for bustling facilities, turnstiles allow only one person at a time, and only after the visitors present the appropriate entrance credentials.
Depending on the needs of a building, it’s possible to operate the turnstiles either with or without the assistance of the front-desk or security staff.
Integration of video surveillance in key security spots on a building, and particularly the main entrance, not only deters criminals but also helps law enforcement authorities to identify the tailgaters in case of a crime.
Thanks to technological advancements in biometrics and machine learning, some modern video security systems can even differentiate between people passing in the foreground and tailgaters.
Wearable identification, such as badges, ensures that anyone who carries it is authorized to be in the building. This includes all permanent staff and visitors, as well as temporary workers.
While visitor badges are inexpensive and ubiquitous, other tools such as a QR code generated from a mobile app or biometric credentials can also help to prevent the risk of tailgating.
While appropriate measures such as turnstiles, credentials, or surveillance can reduce the risk of tailgating, security is a collective effort.
Has dormakaba offered an innovative concept?
The idea is to cover all hiding areas and dead angles in airlock, room, corridors, or a gate with an AI system that could recognize a person and a part of body detection to determine if there is a tailgating situation in airlock or any space that has restricted access. In addition, this AI based system could recognize more objects and detects more situation like counting the number of people crossing the gate or counting the number of luggage passing by a self-boarding gate or a mom with her baby in a trolly, reacting according to each particular situation.
The innovation here is a supplement camera to count a lower body part in combination with a camera that could detect the person from the top.
What kind of anti-tailgating development is behind the concept of dormakaba?
The concept is based on AI. The camera sees and recognizes a person and other objects that could cross a gate or a door. Our patent application is currently being processed.
- We are currently working on the proof of concept.
- The camera only captures the lower part of the body to ensure privacy.
- The sensors are small and compact, allowing for very short gates.
- We will receive additional information about objects passing through the gates and respond accordingly, such as detecting a person in a wheelchair or a person with their dog.
- This data can provide more detailed information for better decision-making.
- Our goal is to improve safety and enhance the user experience and to avoid false alarms.
Using cameras and sensors to detect people as they move through a space. The AI algorithm can analyze the data from the cameras and sensors to understand the situation and react as needed, while dealing with data privacy concerns.
This involves training machine learning algorithms to recognize patterns and can be particularly useful in situations where there are limited sensors or cameras available, or where the environment is more challenging.
This is continuing the AI development we have done with existing sensors we have in the ARGUS gates launched in 2016.
Is there a product that will incorporate this innovation?
At the BAU 2023 trade fair, we will present the new compact Argus V60 sensor gate. It offers safety, elegance, and efficiency in a very small space. Thanks to modern sensor technology, this compact sensor barrier, which has already won the German Design Award Special 2023, gives architects and users more freedom without compromising on personal protection and separation detection. Particularly in areas such as foyers with elevator systems or within office buildings.
This communication contains certain forward-looking statements including, but not limited to, those using the words “believes”, “assumes”, “expects” or formulations of a similar kind. Such forward-looking statements reflect the current judgement of the company, involve risks and uncertainties and are made on the basis of assumptions and expectations that the company believes to be reasonable at this time but may prove to be erroneous. Undue reliance should not be placed on such statements because, by their nature, they are subject to known and unknown risks, uncertainties and other factors outside of the company's and the Group's control which could lead to substantial differences between the actual future results, the financial situation, the development or performance of the company or the Group and those either expressed or implied by such statements. Except as required by applicable law or regulation, the company accepts no obligation to continue to report, update or otherwise review such forward-looking statements or adjust them to new information, or future events or developments.
This communication does not constitute an offer or an invitation for the sale or purchase of securities in any jurisdiction.
dormakaba®, dorma+kaba®, Kaba®, Dorma®, Ilco®, LEGIC®, Silca®, BEST® etc. are registered trademarks of the dormakaba Group. Due to country-specific constraints or marketing considerations, some of the dormakaba Group products and systems may not be available in every market.