Video surveillance is commonly associated with security. But in most cases, it's used to record incidents and assist in investigations after the fact rather than prevent undesirable events.

Artificial intelligence–powered video analytics is a highly promising trend that fundamentally changes the way things work. Extracting manageable data from a video stream can help recognize risky situations early on, minimizing damage and, ideally, completely avoid emergencies. At the same time, AI significantly expands the areas of application of video surveillance beyond security systems.

AI significantly expands the areas of application of video surveillance beyond security system

 

However, the hype around this new, trendy technology prevents the potential user from choosing quality solutions in a wide variety of products. This often leads to over-expectation, followed by a complete let-down. Can AI-powered video analytics really be the key to a technological breakthrough in video surveillance? We'll take a look at what the technology can do, what it can't, and where it can go from here.

Technological Breakthrough Or Just Another Bubble?

It's often said that the video management software (VMS) market is becoming increasingly commoditized and widely available. A lot of products with similar features (or, at least, similar promises from the manufacturer) make it hard to choose. As a result, vendor names and reputations are turning into one of their primary selling points. Manufacturers have two choices available: get wrapped up in a price war and rely on cutting expenses, or offer a product that's truly innovative and revolutionary.

Manufacturers have two choices available: get wrapped up in a price war, or offer a product that's truly innovative and revolutionary

VMS developers who choose the second route are gravitating towards creating products that use artificial intelligence based on neural networks and deep learning. Emerging two or three years ago, the AI video analytics market is experiencing a boom in growth. This new tech wave has stirred the still, stagnant backwaters of the VMS world and gave small, ambitious developers something to be optimistic about. It seems they now have a chance to emerge as market leaders in the next few years.

However, the hype around this popular trend is raising reasonable concerns among experienced security industry professionals. These concerns come from clients looking for a solution to their problems, and from suppliers building a long-term development strategy.

This largely resembles another tech bubble, like the one built up around pre-AI video analytics and burst when it became clear that the promises around it were pure marketing hype. However, there are factors that indicate that AI-powered video surveillance systems aren't another bubble.

The Three Factors

The first — and the main one — comes from systems already in place on customers' sites. They fulfill the same promises made during the previous bubble by hotheads in a rush to teach the computer to analyze events in real time using a classical algorithmic approach.

The second is the fact that this new technology has seen investment from not only software and cloud startups, but also established VMS developers. Even giants like Intel, which has presented a full line of neural network accelerator hardware and a set of software tools that streamlines working with them, specifically in the field of computer vision.

This new technology has seen investment from not only software and cloud startups, but also established VMS developers

The third factor lies in artificial intelligence's abilities. AI plays chess, drives cars, and works wonders in many other fields. Why shouldn't it be applied to video monitoring and analysis?

What AI Can Do

Just what can artificial intelligence do in video surveillance systems at this stage of development? It can't quite analyze a sequence of events and understand the "logic" of what's happening in the cameras' field of view. At least not yet. But it's probable that AI will learn to do this in the next few years. But neural network analytics can already detect, classify, and track objects very well, providing high accuracy even in busy scenes.

 

Artificial intelligence can be used in the real world to:

  • Detect smoke and flames for early fire warning at open areas (forest, open warehouse, parking lot, etc.);
  • Distinguish people/vehicles from animals and other moving objects, e.g. to protect the perimeter of a nature park from poachers;
  • Distinguish a person in a helmet and protective clothing from a person without them to prevent accidents at a dangerous production facility or construction site;
  • Count objects of a specific type, e.g. cars in a parking lot, people in the sales floor, wares moving on a conveyor belt, etc. in non-security-related solutions.

Those are just a few examples. After training a neural network, it can tackle other, similar tasks, too. Generally, a neural network trained in specific conditions isn't replicable. In other words, it won't work as well under different conditions. On the other hand, developers have learned how to quickly train AI for the needs of a specific project. The most important requirement is having enough video footage.

Somewhat apart from that is the use of neural networks in facial and automatic number-plate recognition. This is an example of reproducible neural networks (train once, deploy everywhere), which makes them more appealing commercially. If non-reproducible neural networks have only recently become economically feasible due to the rapid evolution of specialized hardware (aforementioned Intel's product, for example), then the use of AI in facial recognition and ANPR has been well established for a long time.

The use of AI in facial recognition and ANPR has been well established for a long time

Another kind of AI analytics that we'll explore is behavior analytics. This function, probably more than any other, is bringing video surveillance systems closer to understanding what's happening on camera. Its potential is vast.

How Behavior Analytics Works

From a technical point of view, behavior analytics combines artificial intelligence with a classic algorithmic approach. A neural network trained on a multitude of scenarios can determine the position of the bodies, heads, and limbs of humans in the camera's field of view. The algorithm outputs an array of data containing descriptions of their poses.

Conditions can be set for data to detect a specific pose, such as raised hands, prostrated or crouching persons. Developers can use this to quickly create new detection tools to identify potentially dangerous behavior specified by a government or business client. There's no need for additional training of the neural network.

How Behavior Analytics Can Be Deployed

Someone crouched down next to an ATM could be a technician, CIT guard, or burglar. Bank security should be notified in any of the cases.

A person in shooter position, together with a bank employee or cashier with their hands raised could indicate a robbery. The system can be configured to automatically send alerts with a surveillance snapshot to the police so they can assess the threat and take action if needed. It's vital that the police receive the alert, even if the employee is unable to activate the alarm.

In many cases, attention should be directed to a prostrate individual. This could be somebody who needs immediate help, or it could be someone sleeping in an inappropriate public place, for example, a 24/7 ATM space.

Behavioral analytics can also be used to ensure workplace safety. For example, tracking whether employees are holding the handrails when using the stairs at a manufacturing facility or a construction site.

What Now?

Behavior analytics can be deployed wherever your clients' imagination takes them. With this feature, practically any pose that indicates potentially dangerous behavior can be detected. Timely response to an alarm helps avoid material damages or, in other situations, casualties.

Practically any pose that indicates potentially dangerous behavior can be detected

An area of potential development for behavior analytics is the ability to analyze a sequence of poses by the same person or a combination of poses and relative positions of several individuals. That will be the next level of evolution in AI's use in video surveillance: moving from "detecting" to "understanding" behavior in real time.

In its most basic form, this type of analytics can be deployed to detect deviations from the search procedure in correctional facilities when a person being inspected must assume a pre-defined sequence of poses. A more advanced form allows it to detect any kind of abnormal behavior, such as a brawl breaking out in a public space. Ideally, behavior analytics can predict dangerous situations based on nearly imperceptible cues gleaned from collected statistics and a Big Data analysis.

At the moment, this sounds like pure fantasy, but what seemed like whimsy not too long ago is now a reality with AI. It's already beaten humans in chess and the game of Go (Weiqi). Will artificial intelligence be able to outplay humans at charades one day? It's entirely possible that we'll soon see for ourselves.

Share with LinkedIn Share with Twitter Share with Facebook Share with Facebook
Download PDF version Download PDF version

Author profile

Alan Ataev Chief Sales and Marketing Officer, AxxonSoft

In case you missed it

Be Our Guest: How to Manage Visitors With Both Safety and Service
Be Our Guest: How to Manage Visitors With Both Safety and Service

In today’s fraught times, business continuity and success hinges on how you manage the visitors to your company. By prioritizing safety and security, and coupling them with top-notch attention and customer service, you win loyalty and gain a reputation that will serve you in years to come. An excellent way to accomplish this is by identifying and implementing the best visitor management system for your company. And visitor management systems go beyond ensuring the safety of your visitors and staff safety from your visitors. A feature-rich VMS will track your guests' activities, so you can better understand their preferences for future visits. That way, you can manage visitor experience and tailor amenities and preferences. Both customer loyalty and brand reputation benefit. Visitor management systems: who uses it, and why is it used? Visitor management refers to all the processes put together by an organization to welcome, process, and keep track Visitor management refers to all the processes put together by an organization to welcome, process, and keep track of all the guests daily. A visitor management system (VMS) is the technology used to manage guests for their convenience, safety, and security. Several features are typical in today’s applications. They include preregistration tools,  video intercoms, self-check-in stations, and health screening. In visitor management, the term "visitor" doesn't only refer to guests but also anyone without an authorized access credential. For instance, an employee without their access credential logs in as a visitor. The same applies to a delivery man or a technician carrying out routine maintenance. A VMS helps to account for everyone within the organization at any given time. Who uses visitor management systems? You need a visitor management system to manage a school or hospital, an office, or even a residential building. Here's why: Visitor management system for schools: schools are among society’s most vulnerable facilities. A VMS is almost mandatory in this setting. It helps to identify visitors, detect intruders, and alert security of any unauthorized access. Visitor management system for offices: A VMS accounts for guests at all times. They include clients, maintenance contractors, delivery men, employees without credentials, friends, and family, Visitor management system for hospitals: access control is essential in hospitals, and managing visitors plays a major role. Hospitals offer access to pharmaceuticals, medical records, newborns, and expensive equipment. It is crucial to monitor restricted hallways and sections with video intercoms and track unauthorized persons' movements. Residential visitor management system: tracking people's movement is a key VMS component. In case of a crime, knowing who had access to the building within a specific time frame can help in the investigation. Plus, tracking the activities of visitors can deter future crime. Why is the visitor management system important? A video intercom makes it much more difficult for a visitor to impersonate a known guest. VMS accounts for everyone within the organization in cases of emergency. VMSs can prevent intruders and alert the security department of a breach. A VMS creates a positive visitor experience, which shapes perception of the organization. With a trusted VMS in place, employees can focus on being productive. Health screening gives staff peace of mind. It increases employees' willingness to return to work in the midst of the COVID 19 pandemic. How does a good VMS address occupant and visitor safety? The necessary technology to ensure building safety The best visitor management systems contain the necessary technology to ensure building safety. To maximize occupant and visitor safety, a VMS should have the following features: Job one of a VMS is visitor identification. It also helps deter potential criminals. Some VMSs go beyond identification by running a quick check on the visitor's ID and alerting security of any discrepancies. By identifying and proving a visitor's identity, the VMS ensures the safety of employees and other visitors. VMS helps with compliance A good visitor management system helps the organization follow regulations, such as for occupancy. In the COVID era, some states may require health screening for guests. Health screening helps protect the building's occupants from exposure to health hazards. Information security VMSs also aid in information protection. It takes mere seconds for a rogue visitor to download files into a jump drive, photograph exposed blueprints, or copy customer lists. Visitor management systems restrict visitor access to parts of the building and track the whereabouts of guests. Visitor privacy With pen and paper systems, walking up to the receptionist often gives visitors full view of the visitors list. Visitor management systems seal that vulnerability. Visitors can check in without fear that anyone nearby can see their information. Emergency evacuation With a good VMS, the exact number of people within the building is always known. In the case of an emergency, first responders can use VMS data to identify everyone on site. This is a safety net for both the occupants and visitors to the organization. How to manage building visitors System features depend on the purpose and setting of the VMS. Yet certain features and processes are essential. Preauthorization and health screening The first step is knowing the visitors upfront. Preauthorisation allows everyone to know who is coming and when. Guests specify the time and purpose of their visits. You get to welcome and accommodate your visitors accordingly. Some systems may also be able to upload documents of interest, such as proposals, contracts, presentations, or agendas. Health screening is critical today. It signals that the organization cares about its guests. A visitor is more likely to visit an organization that prioritises health and safety. Health screening is a way to protect your staff and send the right message. Video intercom Along with health screening, video intercom is a key element of VMSs. It enables secure video identification with remote, touchless, and COVID-safe access into buildings. Intercoms are a safe and secure way to communicate with audio and video without physical contact. Video allows you to visually verify the visitor. The audio component enables spoken communication. Some systems even use facial recognition technology and mobile app unlock. When integrated with access control, visitor arrival is seamless. Upgrade to touchless access Touchless access is the safest and most secure VMS option Touchless access is the safest and most secure VMS option. It is more sophisticated because it receives visitors without them having to lift a finger. It's also convenient and effective. In this time of the novel coronavirus, the demand for hands-free systems is surging. VMS has pivoted to met this demand. Many organizations are finding how touchless systems increase safety in the workplace. Visitor logging is essential for managing guests to your building. Besides being a source for verification and data tracing, it also helps in real-time to know who signed into the building and who hasn't signed out yet. Tracking the movement of visitors within the facility makes it clear where they are at all times. This way, there can be an effective emergency action plan for visitors and other occupants. This feature has use in contact tracing, health investigations, and other investigations, such as for theft.

Why the Touchless Office is Another Argument for Going Passwordless
Why the Touchless Office is Another Argument for Going Passwordless

Security experts have discussed the demise of the passwords for years. As early as 2004, Bill Gates told the RSA Security Conference that passwords “just don’t meet the challenge for anything you really want to secure.” Change has been slow, but the sudden increase in remote working and the need for enterprises to become touchless as they try to encourage teams back to the office is increasing traction. Here we look at the future of passwordless authentication - using the example of trusted digital identities - and share tips on choosing a solution that works for your organisation. The move away from passwords was beginning to gain momentum pre-pandemic. Gartner reported an increase in clients asking for information on ‘passwordless’ solutions in 2019. Now Gartner predicts that 60% of large and global enterprises, and 90% of midsize enterprises, will put in place passwordless methods by 2022. This is up from 5% in 2018. The many limitations of passwords are well-documented, but the cost of data breaches may be the reason behind this sharp upswing. Stolen credentials – usually passwords – and phishing are the top two causes of data breaches according to the 2019 Verizon Data Breach Incident Report. Each breach costs businesses an average of anywhere between £4M to £8M depending on which studies you read. A catalyst for change As in so many other areas, the pandemic has been a catalyst for change. Newly remote workers using BYOD devices and home networks, sharing devices with other family members, and writing down passwords at home all make breaches more likely. And seasoned home workers represent a risk too.  It also means that enterprises are developing new procedures to mitigate the spread of disease. This includes a thorough examination of any activity that requires workers to touch surfaces. Entering passwords on shared keyboards or touchscreens falls squarely in this area of risk. As does handling physical smart cards or key fobs. Enterprises are expanding their searches from “passwordless” to “passwordless and touchless,” looking to replace physical authenticators. In the quest to go touchless these are items that can be easily eliminated. The future of passwordless authentication Using fingerprint or facial recognition often only provides a new front-end way to activate passwords Common alternatives to passwords are biometrics. But, using fingerprint or facial recognition often only provides a new front-end way to activate passwords. Passwords are still required for authentication after the biometric scan and these live in a central repository vulnerable to hackers. With one successful hack of the central repository, cyber-criminals can swipe thousands of details. In other words, biometrics on their own are not an improvement in security, only a better user experience. They need to be combined with a different approach that adds another layer of security. A more secure option is to move away from the centralised credential repository to a decentralised model. For example, one based on trusted digital identities. This is where digital certificates are stored on users’ phones. Think of encrypted digital certificates as virtual passports or ID cards that live on a worker’s device. Because they are stored on many separate phones, you are able to build a highly secure decentralised credential infrastructure. A solution that uses people’s phones is also compatible with touchless authentication systems. You can replace smart cards and key fobs with a phone-based security model and reduce the number of surfaces and items that people touch. This is especially beneficial for workplaces where people have to visit different sites, or for example in healthcare facilities. Replacing smartcards with a phone in a pocket reduces the number of items that clinicians need to take out and use a smartcard between and in different areas, which may have different contamination levels or disease control procedures. How do trusted digital identities work?   Workers unlock their mobile devices and access their trusted identity using fingerprint or facial recognition Here’s an example installation. You install a unique digital certificate on each user’s mobile device — this is their personal virtual ID card. Authorised users register themselves on their phones using automated onboarding tools. Workers unlock their mobile devices and access their trusted identity using fingerprint or facial recognition. Once they are authenticated, their device connects to their work computer via Bluetooth and automatically gives them access to the network and their applications with single sign on (SSO). This continues while their phone is in Bluetooth range of their workstation, a distance set by IT. When they leave their desk with their phone, they go out of range and they are automatically logged out of everything. Five tips on choosing a passwordless solution More automation means less disruption Consider how you can predict and eliminate unnecessary changeover disruptions. The task of onboarding large or widely dispersed employee populations can be a serious roadblock for many enterprises. Look for a solution that automates this process as much as possible. Scalability and your digital roadmap Will you maintain remote working? Having a high proportion of your team working remotely means that passwordless solutions will become more of a necessity. Are you expecting to grow or to add new cloud apps and broader connectivity with outside ecosystems? If so, you need password authentication that will scale easily. Encryption needs and regulatory requirements If your workers are accessing or sharing highly sensitive information or conducting high-value transactions, check that a solution meets all necessary regulatory requirements. The most secure passwordless platforms are from vendors whose solutions are approved for use by government authorities and are FIDO2-compliant. Prioritise decentralization Common hacker strategies like credential stuffing and exploitation of re-used credentials rely on stealing centralised repositories of password and log-in data. If you decentralise your credentials, then these strategies aren’t viable. Make sure that your passwordless solution goes beyond the front-end, or the initial user log-in and gets rid of your central password repository entirely. Make it about productivity too Look for a solution that offers single sign on to streamline login processes and simplify omnichannel workflows. For workers, this means less friction, for the enterprise, it means optimal productivity. Security improvements, productivity gains and user goodwill all combine to form a compelling case for going passwordless. The additional consideration of mitigating disease transmission and bringing peace of mind to employees only strengthens the passwordless argument. The new end goal is to do more than simply replace the passwords with another authenticator. Ideally, enterprises should aspire to touchless workplace experiences that create a safer, more secure and productive workforce.

Deploying Video Analytics for Contact Tracing During COVID-19
Deploying Video Analytics for Contact Tracing During COVID-19

Developing an effective contact tracing system in the UK to monitor the spread of COVID-19 has proved to be problematic. The trials of the app developed by the government and its partners encountered numerous challenges, and despite the reopening of restaurants, pubs and shops, the current approach to contact tracing is inconsistent, with recent reports suggesting not all establishments are following the government guidance.  At the same time, businesses are being encouraged to ask employees to return to the workplace as lockdown restrictions ease, and the lack of an effective contact tracing system is only going to become more of an issue. Responsibility now lies with employers to ensure social distancing measures are adhered to in the workplace, trace any contact that a person infected with COVID-19 has had with others, and communicate consistent messaging across their organizations. Considering all of these challenges, it is not surprising that technology is being turned to for the answers.  Turning to technology  However, it is not just cutting-edge technology that can support measures to address health and safety issues related to COVID-19; the use of existing infrastructure is vital too. Consider the ubiquity of CCTV in workplaces and public spaces, especially in densely populated cities. Recent research has shown that London, for example, has 627,727 cameras for 9.3 million residents - the equivalent of 67.5 cameras per 1,000 people. The data collected from these feeds will play a key role in effectively tracing interactions and monitoring the adherence to social distancing measures. Tracing interactions and monitoring the adherence to social distancing measures As useful as this data is though, the sheer volume of it is enormous. Sifting through hundreds of hours of video footage collected from networks of thousands of cameras will be far too time-consuming and inaccurate to complete manually. This is where more advanced technology such as A.I.V.A. (Artificial Intelligence Video Analytics) is required. A.I.V.A. solutions use existing camera networks and geospatial algorithms to determine an individual’s location in the camera field of view in real-time, automatically learning the perspective of the scene and calculating the GPS coordinates of individuals in real-time based on their location in the camera field of view. Social distancing algorithms For example, with regards to social distancing, an algorithm can detect when two parties are within a meter proximity of each other. This will trigger an alert in the system and log the occurrence in a dashboard report. If government recommendations change and the suggested distance is 2 meters, the algorithm is easily adjusted. This approach will help to reinforce changes in behavior to encourage social distancing and, in the worst case, establish an effective contact tracing system for those who have been infected with COVID-19 and have come into contact with others. An effective contact tracing system for those who have been infected with COVID-19 Firstly, in terms of how this can be used for social distancing, AI powered video analytics can be used to identify particular hotspots where breaches occur. While isolated incidents of a breach in the 1m rule may not be particularly useful, when a series of occurrences is identified from thousands of hours of CCTV footage, vital insights can be gained into localised clusters of COVID-19 infections. The reports generated from this type of analysis can be extremely useful; for example, pinpointing particularly busy areas of a job site, identifying queues at a coffee station at a certain time of the day, or the most frequently used exit of an office building. Breaching the rules Armed with these insights, businesses can implement measures to try and alleviate such bottlenecks. In practice, this may involve implementing one-way systems or moving people from one congested area to a quieter one at particular times of the day, to help reduce the chances of breaching the 1m social distancing rule. There is even the capability of triggering an automatic alert when a breach is observed to remind employees of their responsibility to adhere to the guidelines. If a business is informed that one of its employees or a visitor has contracted COVID-19, the use of A.I.V.A can support in helping to identify areas that the person has visited and whether there were any other people in that area at the same time. Instead of informing others directly, the business can issue a notice to say exactly where that person was and advise employees who may have been present to be tested. This is particularly useful for businesses with large sites, who need to manage each building and facility on a case-by-case basis. Workforces can be protected by decreasing the chance of social distancing breaches Although simple to implement, AI powered video analytics can play a key role in helping businesses implement solutions that allow employees to safely return to work. With such technology in place, workforces can be protected by decreasing the chance of social distancing breaches occurring and effectively tracking those who have it. The technology does not rely on the identification of specific individuals nor their personal information or mobile phones, but rather recognises behavior patterns and uses this approach to provide accurate information to groups of people that need it. AI has long been touted as the technology set to revolutionise life as we know it, and now it has the chance to unlock its potential and protect people in a world significantly affected by COVID-19.