Download PDF version Contact company

A tipping point is defined as: “The point at which a series of changes becomes significant enough to cause a larger, more important change”. In the same way that IP video changed surveillance a decade ago, our industry is now feeling the impact of recent developments in Artificial Intelligence, Machine Learning, Deep Learning, Big Data, and Intelligent Video Analysis.

Keyword definitions

Let’s start with a few more definitions. Artificial Intelligence (AI) deals with the simulation of intelligent behaviour in computers. Machine Learning (ML) deals with developing computer algorithms that access data and use it to learn for themselves. Neural networks are computer systems that loosely mimic human brain operation.

Deep Learning is a subset of ML based on neural networks that has been proven to provide breakthrough capabilities in many problems that were previously unsolvable, and Big Data, or metadata, refers to huge amounts of structured and/or unstructured data -- in our case, the immense quantities of video information being generated daily by security cameras deployed in cities around the world. Deep Learning is tipped to change Intelligent Video Analysis (IVA), the digital video technology integrated with analytical software that is a basic tool for our industry.

AI in surveillance

Traditionally, the main benefit of surveillance cameras is the ability to collect evidence for debriefing or investigation, as well as the ability to view events remotely in real-time. A decade ago, video analytics technologies were introduced to solve the problem of human inattention -- computers don’t get tired, bored or distracted, and can monitor a camera continuously.

And then, camera costs dropped, deployment skyrocketed, and video management systems began collecting reams of useless, costly unstructured data. AI technology seemed to answer the pressing new industry needs of how to use this Big Data effectively, make a return on the investment in expensive storage, while maintaining (or even lowering) human capital costs

Three limiting factors

All this was theoretical, however, as multiple technological barriers prevented AI solutions from real-world utilisation. Despite decades of research on how to cause a computer to accurately recognise different objects in a video stream, the quality of the results, especially in urban environments, was, to put it mildly, underwhelming.

Deep Learning software must be able to differentiate between different objects and under various circumstances
Deep Learning has matured to the point where it can accurately detect and classify objects both in still images and in video

AI was limited primarily by these three factors: 

  • Lack of understanding -- The software must be able to differentiate between different objects (person, vehicle, animal, etc.), and under various circumstances (day, night, seasonal weather conditions, etc.).

  • Inability to learn -- Traditional IVA applications relied on a rule-based approach that required software configuration -- by a human operator -- for each monitoring camera and each type of alert. Although effective in some scenarios, the exponential growth in camera counts rendered this approach impractical, given the amount of manual labor required to configure, reconfigure, and maintain rules.

  • High cost -- The hard truth is that budgets for security and safety will always be constrained. Until recently, implementing real-time AI was extremely cost-prohibitive, sometimes requiring a 1:1 server to camera ratio.

Meeting challenges

That was yesterday. Today, the application of AI in security applications has reached its tipping point, meeting the above-mentioned challenges. 

  • Understanding -- Deep Learning has matured to the point where it can accurately detect and classify objects both in still images and in video. DL technology is fast becoming the basic building block for IVA.

  • Ability to Learn -- As an AI solution collects and analyses data over time, it creates metadata that describes all objects in each video stream. Machine Learning techniques process this metadata to generate models for “normally observed” behaviour. These models are applied in real-time to detect behaviours deviating from the norm. Only those flagged as suspicious events require review by a human operator. This technique allows the solution to scale to an unlimited number of cameras, with no need for a human to configure each new device.

  • Lower cost -- The rapid increase in GPU computational capacity, coupled with mass market adoption, has lowered server costs to a reasonable level. Today, with the correct implementation, a single server can be deployed across hundreds and even thousands of cameras.

The convergence of Deep Learning for video analysis, advances in AI for fully automated event detection, plus the significant reduction in cost to implement these techniques – including cloud-based software as a service (SaaS) models -- means that the fully automated video surveillance solution for cities is fast becoming a reality. We’ll see more of this type of solution being deployed over the coming months, and within the next few years, it will be standard in any Smart City deployment.

Share with LinkedIn Share with Twitter Share with Facebook Share with What's App Share with Facebook
Download PDF version Download PDF version

Author profile

In case you missed it

Which security markets are embracing touchless and contactless systems?
Which security markets are embracing touchless and contactless systems?

The idea of touchless systems has gained new levels of prominence during the last year, driven by the global COVID-19 pandemic. Contactless systems have been part of the industry’s toolbox for decades, while technologies like facial and iris recognition are finding new uses every day. We asked this week’s Expert Panel Roundtable: Which security markets are embracing touchless, contactless systems and why? 

How body worn cameras and AI can curb the issue of abusive behaviour
How body worn cameras and AI can curb the issue of abusive behaviour

Amongst the many negative consequences of the pandemic is a rise in violent and abusive behaviour across society. Health workers have experienced it on a regular basis. So too have police officers and public transport workers. Unfortunately, violence and abuse towards shop workers is also endemic in British society. To address this problem which, in truth, has been on the rise since long before the emergence of COVID-19, we need better deterrents. The ability to prosecute these offences is one such deterrent, but just as important is the ability to deescalate situations before they spill over into unacceptable or unlawful behaviour. Major retail customers In both instances, organisations of all sizes are now recognising that the answer could involve greater use of rapidly advancing body worn camera technology. Andy Marsh, the Chief Constable of Avon and Somerset Police, is one of the police officers responsible for introducing body worn cameras to the UK police force, where they are now in widespread use. Andy Marsh is one of the police officers responsible for introducing body worn cameras to the UK police force He explains that “The reason the majority of people don’t speed or drink-drive is that rational human beings weigh up the risk and consequences of breaking the law and getting caught. Body worn cameras help provide appropriate ‘desistance’, especially where there are forward-facing screens so the person interacting with the wearer can see themselves and their behaviour.” Evidence shows that if a forward-facing camera is switched on before the intervention becomes hostile, it will generally lead to a de-escalation – as often as 90% of the time, according to one of our major retail customers. Digital evidence investigations Only a tiny handful of abusive incidents ever translate into arrests and prosecutions. A key issue is a lack of clear evidence – how to get past the usual impasse of one person’s word against the other. Body worn cameras break the deadlock and allow organisations to report incidents to the police with confidence, knowing that they will lead to action. Marsh suggests that “We usually see an earlier admission, an earlier guilty plea and a more appropriate sentence, where body worn camera footage is in play.” The technology has come on in leaps and bounds in recent years. For example, it’s now possible to record high-definition footage on a lightweight device that’s barely the size of a palm. And it’s not just about the evidence organisations gather themselves. Many police forces are looking at ways to make it easier for businesses and the public to collaborate on digital evidence investigations. Body worn cameras This is good for the victims of crime because it means we get the evidence more quickly" “We’ve created an online crime portal in Avon and Somerset which people can use to pass digital evidence and material to us without an officer having to attend their premises. This is good for the victims of crime because it means we get the evidence more quickly and can take action more swiftly to resolve that issue,” adds Marsh. Our body worn cameras can now even support facial recognition thanks to new, smart AI on the devices themselves, which can scan and process faces within a three-metre distance against a pre-defined database of people (which we call a watchlist). Any matches trigger alerts or additional camera activity such as recording and streaming, while the facial recognition data of people not on the watchlist itself is not recorded or saved to assuage privacy concerns. Similar criminal behaviour Where could this technology come in handy? Well, staff at gambling venues or in-store retail workers could undoubtedly benefit from the ability to quickly spot known fraudsters or addicts who have requested that venues refuse their custom. Stewards at mass sporting events could play a key role in helping to identify people who have been banned from attending. The primary reason for using body worn cameras is to increase the safety of frontline workers The primary reason for using body worn cameras is to increase the safety of frontline workers, deescalating confrontations and limiting the use of force. AI-powered facial recognition can also serve this purpose by helping them make better-informed choices about how to handle specific situations. For example, it is a massive advantage to police officers on the beat to understand that the person they are dealing with may have a history of similar criminal behaviour. Facial recognition technology But it’s also an advantage within retail, where aggressive incidents are on the rise and staff need all the help they can get to determine what an appropriate response should be to a particular customer incident. In fact, extensive consultation with our retail, police, transport and gambling customers indicates that introducing facial recognition technology to body worn cameras could be instrumental, not just in helping to prevent crime, but in tracking down vulnerable and missing people too. Of course, facial recognition technology has to be balanced against the need to protect the privacy of ordinary citizens. Video recording using body worn cameras has to be done proportionately – the same is true for the use of facial recognition technology. The technology also has to be compliant with GDPR, Data Protection, the Information Commissioners recommendations and so on. Positive working environment Violent and abusive incidents affect everyone in the immediate vicinity and create a culture of fear Importantly, it should be for a specific, proportionate and justifiable reason which, of course, means it should never be used for indiscriminate mass surveillance. Every organisation using this technology must remember that a facial recognition system match is not proof of someone’s identity, but rather, an indication of likelihood to help inform the user rather than dictate the course of action. Violent and abusive incidents affect everyone in the immediate vicinity and create a culture of fear and apprehension. This is why it’s so important to get on top of the problem – both on a societal and at an organisational level. Body worn cameras have a vital role to play, as an evidence-gathering tool and as a deterrent that empowers the wearer and creates a more positive working environment. Deterring unlawful behaviour One of the critical roles these cameras play is in staff training, providing real-world video evidence that can be used to educate and upskill workers across a variety of industries. Society’s problem with abusive and violent behaviour cannot be solved by technology alone. But with exceptional quality camera footage now a reality, and the possibility of AI technology at the device level in real-time, body worn cameras will only get better at deterring unlawful behaviour and helping to protect hardworking frontline staff. Alasdair Field is CEO of video technology provider Reveal, which works with UK police forces and major brands such as Matalan, JD Sports and Boots to help them improve staff safety, deescalate confrontations and reduce violent and abusive incidents.

ASSA ABLOY Opening Solutions embraces BIM to smooth specification and installation of door security solutions
ASSA ABLOY Opening Solutions embraces BIM to smooth specification and installation of door security solutions

BIM (building information modeling) provides a process for creating and managing information during the building lifecycle and beyond. BIM is often equated with 3D modeling of construction projects, but the visual component is just part of the value of BIM. Additional data, such as specifications and other documentation, is also part of the process, underlying the visual aspects, helping to drive decision making and providing immediate access to detailed information about all facets of the building process. Incorporating BIM systems For the last six years, ASSA ABLOY Opening Solutions has worked with specification writers and architects in Europe, the Middle East, and Africa (EMEA) to make it easy to incorporate ASSA ABLOY Opening Solutions doors, hardware, and security solutions into BIM systems. Everyone on a project can work together in the interactive and information-rich BIM environment. BIM tools are also used by contractors, distributors, facility owners, and security consultants. BIM software BIM information relating to doors, hardware, and security solutions is available in the cloud  BIM information relating to doors, hardware, and security solutions is available in the cloud with the company’s Openings Studio BIM software. This improves the process of door scheduling and visualisation and enables customers to focus on the design, installation, and management of openings. “If you have up-to-date information inside the BIM model, you can reduce mistakes and misunderstanding in the building industry,” says Marc Ameryckx, ASSA ABLOY Opening Solutions’ BIM Manager for the EMEIA region. “It helps to eliminate mistakes before they happen or as early as possible in the building process. The earlier, the less it costs. We provide data as soon as possible in the process.” (ASSA ABLOY Opening Solutions also has comparable systems available in other regions of the global company.) Centralised data in BIM 3D model Expanding the data available in BIM provides additional value compared to merely providing “BIM objects” that can be incorporated into a BIM 3D model. The combination of BIM modeling and the underlying specifications boosts the quality of the project and its key to success, says Marc Ameryckx. Even after the building is complete, the BIM model is still valuable, providing a repository of “as-built” information that can be used by building managers and security professionals tasked with operating and maintaining the building. For example, if a lock needs to be replaced, retrofitting is simpler because all the information about the lock and existing installation is available in a centralised data file. Revit and ArchiCAD A widely used BIM software is Revit from Autodesk, a program that brings architecture, engineering, and construction disciplines into a unified modeling environment to drive more efficient and cost-effective projects. Another BIM software program is ArchiCAD, developed by the Hungarian company Graphisoft. Openings Studio™ added a plugin for ArchiCAD this year, in addition to Revit. Tailor-made information security solutions We provide tailor-made information security solutions with various hardware on projects with more doors" “We can provide tailor-made information security solutions with various hardware on projects with more doors, adding more flexibility,” says Marc Ameryckx. “Customers do not need to be the experts on the products because we provide expertise as part of our specifications.” For example, how often do building mistakes occur because of a misunderstanding about the electrical needs of a lock and the wrong cabling is installed? The problem is especially expensive if it is discovered only after the walls are complete. Providing complete data about the electrical lock as part of a BIM system avoids the snafu. Another example is the specification of a deadbolt lock on a door that operates with an electric strike. The deadbolt undermines the intended operation of the electric strike and can interfere with escape routes in case of an emergency. The mistake becomes obvious in the BIM environment and can be rectified before consequences impact the real world. Data addition to Opening Suites site ASSA ABLOY Opening Solutions is continuously expanding the data it provides at the Opening Suites site, covering additional functionality and more components including the door, cabling, and electrical connections. Hardware sets are linked to specific doors in the BIM models, including all the details of various components, including article numbers, technical sheets, electrical requirements, all depending on customer expectations. Physical equipment includes QR codes that can be scanned by a smartphone to provide information on the door (A mobile app is in development). More details and more data Experienced BIM consultants work with the Openings Studio software on projects ranging from single doors to large buildings with many doors. Data will be more and more important, and there will be more data inside BIM models Adding more data and detail to the BIM process at the level of each door expands the usefulness of BIM, which has historically been focused on broader issues such as structural work and HVAC. “Openings Studio™ provides all the data to integrate doors and security in the BIM process,” says Marc Ameryckx. The higher level of detail may be a new aspect even for customers who already use BIM software. “Data will be more and more important, and there will be more data inside BIM models,” says Marc Ameryckx. In the future, the use of “digital twins” could expand the capabilities even further; for example, the software could simulate escape routes in case of fire. More data makes more things possible.