Facial recognition is becoming more popular in newer systems for access control — a shift that began before the pandemic and has intensified with a market shift toward “touchless” systems. A new facial recognition platform is emerging that responds to the access control industry’s increased interest in facial recognition by expanding the concept with a new higher level of technology.

At the core of the new system is high-performance, true-3D sensing with facial depth map processing at low power consumption, which enriches the capabilities of small-footprint access control devices.

New proficiencies include anti-spoofing (preventing the use of a 2D photo of an authorized user to gain entry) and anti-tailgating (preventing an unauthorized person from gaining entry by following an authorized user) in real time and in challenging lighting conditions. The system uses “true 3D sensing,” which incorporates single-camera structured-light 3D sensing—as opposed to dual-camera depth sensing or IR video imaging-based approaches.

AI vision processing and 3D sensing technologies

The new “Janus reference design” incorporates AI vision processing, 3D sensing technologies, and RGB-IR CMOS image sensor technologies from Ambarella, Lumentum and ON Semiconductor. Specifically, Lumentum’s high-reliability, high-density VCSEL projector for 3D sensing combines with ON Semiconductor’s RGB-IR CMOS image sensor and Ambarella’s powerful AI vision system on chip (SoC).

The Ambarella, Lumentum, and ON Semiconductor engineering teams worked together to incorporate their complementary technologies into the reference design.

Janus reference design

A reference design offers OEM product and engineering teams a fully functional engineering reference implementation that they can use as the basis for their own product. Teams will often customize a reference design with their choice of various third-party hardware components to fit their product specifications and positioning. They might also integrate their own software, algorithms, and back-end system integrations. The advantage to this approach is that the manufacturer can get to market quickly with a next-generation product that emphasises their core strengths.

3D depth information for facial recognition

Generally, it takes between nine months and a year for a manufacturer to get to market using a fully functional reference design, such as the one developed jointly by Ambarella, Lumentum and ON Semiconductor.

The Janus platform leverages 3D depth information generated via structured light for facial recognition with a >99% recognition accuracy rate. Traditional 2D-based solutions are prone to false acceptance and presentation attacks, whereas 3D sensing delivers advanced security—just as mobile phones use true-depth cameras for facial recognition. 3D facial recognition also significantly reduces the gender and ethnic biases demonstrated by some 2D facial recognition solutions.

The Janus reference design is also aimed at future smart locks for enterprise and residential use: its unique single-camera 3D sensing solution will help OEMs overcome cost and manufacturability barriers, while the ultra-low power edge AI capability can effectively extend the battery life, which in turn reduces maintenance cost.

Video security and access control

Ambarella sees touchless access control, as well as the convergence of video security and access control, as the mega-trends driving industry innovation and growth—using video, computer vision, and 3D sensing to not only address safety and security, but also to improve the user experience and public health, says William Xu, director of marketing for Ambarella.

The convergence of video security cameras and access control readers has been widely discussed by leading access control OEMs. In many cases, they already integrate video security cameras, readers, door controllers, cloud-access, and the like. In most enterprise installations, one would typically find security cameras installed where there are access control readers. Combining the two devices significantly reduces the maintenance cost and system complexity.

“In comparison to fingerprint or other contact-based approaches, Janus-based access control is touchless—requiring no physical contact with authentication hardware such as fingerprint sensors or keypads—reducing infection risk while enabling a seamless experience,” says Mr. Xu. “The Janus platform provides true 3D depth information to prevent unauthorized individuals from mimicking legitimate users, and the advanced embedded AI processor enables tracking and anti-tailgating algorithms. Janus-based devices perform well in challenging lighting conditions and they are capable of authenticating multiple users simultaneously, with imperceptible latency.”

Lumentum’s high-reliability, high-density VCSEL projector for 3D sensing combines with ON Semiconductor’s RGB-IR CMOS image sensor and Ambarella’s powerful AI vision system

Access Control and public health

What was once purely a security challenge—namely, how to prevent unauthorized entry into a restricted area—has evolved into a public health challenge as well. Many traditional access control methods, from number pads to fingerprint readers, require touch in order to function, and if the current global pandemic has made one thing evident, it’s that minimizing physical contact between users and surfaces is vital to community well-being.

Janus was originally designed to facilitate the next generation of facial-recognition-based access control readers—enabling 3D sensing and high recognition speed for seamless authentication. COVID-19 has accelerated industry-wide research, development, and timelines for Janus-based solutions, says Mr. Xu.

Deep learning and artificial intelligence drive all the new capabilities offered in Janus—capabilities that are only possible due to the platform’s high computational horsepower. The core deep learning and AI capabilities of Janus enable a wide range of advanced features only possible with an embedded vision SoC, says Mr. Xu. All are performed in real time, even when multiple users are being processed simultaneously. These include

  • the extraction and comparison of facial depth maps with those registered in the system;
  • 3D liveness detection, ensuring that the system can distinguish between real users and photo or video playback attacks;
  • anti-tailgating, which relies on computer vision algorithms to detect and track when an unauthorized person follows a legitimate user inside;
  • face mask detection; and
  • people counting.

VCSEL technology

According to Ken Huang, Director of Product Line Management, 3D Sensing, Lumentum: “Lumentum’s VCSEL technology is one of the Janus design’s core strengths and differentiators. The process begins when Lumentum’s high-resolution dot projector projects thousands of dots onto the scene to create a unique 3D depth pattern of a user’s face. Most traditional biometric facial security systems rely on 2D images of users—simple photographs—which reduces authentication accuracy. In contrast, the 3D depth map generated by Lumentum’s technology provides the foundation of a more accurate, more secure, and more intelligent system overall. In addition, Lumentum’s VCSEL solutions incorporate a Class 1, eye-safe laser with zero field failures to date.”

Adds Paige Peng, Product Marketing Manager, Commercial Sensing Division, ON Semiconductor: “If we think of Ambarella’s CV25 as the brain of the Janus design, the AR0237IR from ON Semiconductor is the eye. The AR0237IR image sensor captures the information, and the CV25 processes it. Other face recognition systems use two “eyes” – one to recognize RGB patterns to generate the viewing image stream, and another IR module to detect liveliness in motion. The Janus solution leverages a single “eye”—the AR0237IR—to obtain both visible and infrared images for depth sensing and advanced algorithms such as anti-spoofing and 3D recognition. AR0237IR also provides good sensitivity in various lighting conditions and supports high-dynamic-range (HDR) functions.”

The single-camera 3D sensing solution for access control operates in three seamless steps:

  • Step 1: Lumentum’s high-resolution dot projector creates a unique 3D depth map of a user’s face;
  • Step 2: ON Semiconductor’s RGB-IR image sensor captures the high-resolution images from Step 1, even in low-light or high dynamic range conditions;
  • Step 3: Ambarella’s advanced vision SoC takes the high-resolution images captured in Step 2 and uses deep neural networks (DNNs) for depth processing, facial recognition, anti-tailgating, and anti-spoofing while video encoding and network software run simultaneously.
Share with LinkedIn Share with Twitter Share with Facebook Share with Facebook
Download PDF version Download PDF version

Author profile

Larry Anderson Editor, SecurityInformed.com & SourceSecurity.com

An experienced journalist and long-time presence in the US security industry, Larry is SecurityInformed.com's eyes and ears in the fast-changing security marketplace, attending industry and corporate events, interviewing security leaders and contributing original editorial content to the site. He leads SecurityInformed's team of dedicated editorial and content professionals, guiding the "editorial roadmap" to ensure the site provides the most relevant content for security professionals.

In case you missed it

The Importance of a Secure Perimeter in Safeguarding our Schools
The Importance of a Secure Perimeter in Safeguarding our Schools

Schools play a key role in shaping our future. Following the reopening of classrooms up and down the country, young minds are returning to some normality. Once again they're being inspired, learning useful skills, and forging new interests to ensure our country's continued prosperity. Schools need a comprehensive security infrastructure to protect the children who attend them. Most notably, secure perimeters that keep unwanted people out, but also ensure visitors, parents, and students alike can access their shared community space without feeling segregated or unwelcome. Robust boundary fencing  However, although safety is often the primary concern of parents, with tighter budgetary constraints and funds prioritised to make schools COVID safe, it can be all too easy to let important perimeter replacement or improvement programmes slip. The purpose of boundary fencing is to restrict unauthorized entry and exit The purpose of boundary fencing is to restrict unauthorized entry and exit from school grounds, and should be specifically designed to be fit for purpose. Opting for fencing with a welded pale-through-rail construction and concealed anti-tamper connectors between  fence panels and posts delivers a robust boundary that's virtually impossible to break through. This style of fencing also gives a better finish with no visible joints or unsightly bolts. Attractive and practical solutions Unlike generic riveted palisade fencing, this solution is both attractive and practical, more so now that LPS 1175 SR1 certified and Secured by Design versions are available. The style of fencing should meet a school's desired security and safety requirements, simultaneously, it should not compromise on aesthetic considerations. As part of the public face of the school, it should be attractive as well as functional, helping overcome any concerns of creating a prison-like environment and promote a sense of well-being. It's recommended that perimeter fencing should be: a minimum of two metres high, vandal-resistant and sturdy, grounded on a hard surface, challenging to scale, and have an anti-climb topping, much the same as a high-security option. Access all areas Each educational site must consider the number of necessary entrances A perimeter fence requires secure access points and gates. Each educational site must consider the number of necessary entrances. These should be kept to a minimum, to make it easier to maintain control of visitor movement. However, in larger schools this is not always possible and additional entrances may be required to prevent potentially dangerous congestion at the start and end of the school day. Furthermore, separate gates must be installed for vehicles and pedestrians to ensure they are kept at a safe distance, and avoid unnecessary openings of large, double leaf gates. All access points should be locked during the day to keep students on-site and prevent intruders from gaining access to school grounds. Gates should ideally be matched in  design, height and construction to the fencing,  to prevent creating vulnerable areas and compromising security. Automatic vs manual While automatic gates offer more control, manual gates shouldn't be overlooked. Not only are they easier to install and usually cheaper than automated gates, but they also don't rely on power, so if your site's supply is cut off, they provide a hassle-free exit. Furthermore, gates that are only used at the start and end of the day can be easily locked manually by staff. However, automated gates do offer welcome flexibility, as they include access control devices such as remote controls, keypads and card readers, which will also increase the school's security. They're also robust and heavy, meaning it's incredibly difficult to force them open. Electric gates offer additional versatility with a choice of either full automation, or a hybrid of manned and automatic security, with staff able to allow visitors access via intercom or video system. Securing outdoor facilities It’s also essential to consider outdoor areas when it comes to specifying security options for educational environments. Specialist security fencing should be specified where recreational areas double up as the school’s boundary fencing. The security of the site's sports facilities will also need to be considered. Commonly known as MUGAs (Multi-Use Games Areas), enclosures can be designed with specialist mesh systems to allow multiple sports to be played in the same location while providing safety to participants, spectators and buildings. When it comes to play areas in nurseries and junior schools, installing RoSPA approved and BS EN 1176 compliant fencing and gates is recommended. These are available in both timber and steel options and tested for their ability to provide a safe fencing and gate solution - designed to reduce the risk of limb entrapment. Acoustic fencing is also worth considering for these environments, particularly in urban areas or where housing is close to school play areas. It can help reduce incoming ambient noise from neighbouring busy roads, railway lines, or construction sites, and contain the school noise within its boundary. Offering sufficient protection Focus on learning unimpeded by threat The current generation of children deserve an environment where they can focus on learning unimpeded by threat. Schools need robust perimeter solutions that welcome pupils, offer peace of mind to parents, and provide them with sufficient protection against intruders. Ultimately, it's the responsibility of the head teachers to engage in dialogue with knowledgeable security professionals to get the most appropriate and effective security solutions for their school, staff and students.

Can 5G Boost Innovation in the Smart Home Market?
Can 5G Boost Innovation in the Smart Home Market?

We still have some way to go before we see 5G service rolled out as a UK-wide service, but we can discuss future implications of 5G, and how it can improve device capabilities once its widely accessible.  The impact of 5G lies within the increase in the amount of data that can be transferred between smart home devices through a cloud-based system. By utilising the cloud’s mass computing power and its ability to process data in larger capacities, we are able to receive more in-depth analytics that can help improve smart home devices by making them faster, more informed and intelligent.  Will  5G  boost innovation in the smart home market, if so, how?  Innovation using 5G can be shown with current smart home CCTV systemsAn example of innovation using 5G can be shown with current smart home CCTV systems. Products at present allow you to use functionalities such as motion detection. As it stands, this is a basic form of monitoring that monitors changes in picture and notifies the owner when something unusual has happened.   Once 5G comes into play, the video data captured can be sent off to the cloud, interpreted in more detail, and can allow the system to conclude whether the movement is from a human, object or animal. Facial recognition could also come into play here, providing a more seamless service when reporting incidents to the police.  Similarly, you can use data from various different devices within your home to toughen security measures. A video camera used in conjunction with a presence or heat detector can eliminate small errors by providing the ‘bigger picture’ with more data points to work with - an amalgamation of all smart devices used in tandem.  We can’t solely rely on smart devices to make decisions for us, but what we can do is improve device processing so that by the time we step in, we already have all the necessary information to assess the appropriate call to action.  What will 5G enable homeowners to do within their homes?  Smart homes using 4G currently operate in a fragmented fashion, incorporating Wi-Fi, Bluetooth, and other network protocols. Unlike 4G, 5G will work with low-power devices, making it useful for a broader array of connected products. This means that all devices we will be able to connect any internet connected product in order to allow integrated communication between all devices.   Can be connected to work together in conjunction with your home system Your fridge and other kitchen appliances, for example, can be connected to work together in conjunction with your home system to create an entirely automated home. If your freezer is internet connected, you may get a notification if it loses power, but as everything else is also connected within your home, you will be able to determine whether it is a power fault or product fault straight away. Faster connectivity means that users can quickly take advantage of the data their smart devices provide, such as water use sensors which can monitor levels and allow for behavior changes to curb water usage The same can be applied with vehicles. Cars in future will be autonomous and include an integrated dashcam which can then be connected to your security system to provide added security on-site in the peripheral of your home, alerting you to potential intruders before they reach your front door. 

AI Deployed To Safeguard Cities: How, Where And What
AI Deployed To Safeguard Cities: How, Where And What

The threat landscape we operate in today is changing all the time. Around the world, pressures on law enforcement bodies remain incredibly high as they face the challenge of rising international threat levels and a backdrop of intense political, social and economic uncertainty. It is a challenge that demands a considered, proactive and dynamic response. It’s clear that new technologies, such as Artificial Intelligence (AI), can dramatically improve the effectiveness of today’s physical and cyber security systems and help us to better defend against a wide-spectrum of threats. Finding the balance Specifically speaking, for physical security systems to be effective, they must have the full support of the public. Airport-style environments where security checkpoint processes are implemented are both time consuming and obstructive, and feels, at times, they are in no one’s best interest. Oppressive, fortress-like environments are likely to quickly lose the backing of the public, who want to be able to go about their daily lives without being delayed or obstructed by cumbersome security checkpoints and procedures. For physical security systems to be effective, they must have the full support of the public However, after a large-scale security threat or attack occurs, it is often these more overt systems that we gravitate towards, often fueled by a proven track record of both deterrence and detection. It’s the antithesis of ‘out of site, out of mind’ security. Having these large, bulky overt security systems offer reassurance to people and create a greater sense of security. But what if we could instill this sense of security without monstrous overt systems? What if today’s physical security systems could allow for seamless people flow while creating safe environments, all done in a covert manner without interrupting peoples’ way of life? This is exactly what can be achieved with the some of the new physical security applications that incorporate AI. Security solutions with AI: how, what, where? Today, security solutions driven by AI technologies are being developed and can be covertly deployed across a range of physical environments to protect our global citizenry. These new AI-driven technologies are taking multiple different forms, depending on the locations they are designed to protect. Video management surveillance systems (commonly referred to as VMS systems) are being enhanced by AI/computer vision technology to identify objects. These enhanced VMS systems can be deployed both inside and outside of buildings to identify and flag forbidden objects, such as visible guns, knives or aggressive people, in a wide range of public spaces, such as schools, hospitals, sport stadiums, event venues and transportation hubs. Recognized threat objects in hand or suspicious behaviours can be identified and flagged instantaneously for onsite security to further investigate. In addition, targeted magnetic and radar sensor technologies, concealed in everyday objects like planter boxes or inside walls, can now scan individuals and bags entering a building for concealed threat objects. Using AI/machine learning, these two sensor solutions combined can identify metal content on body and bag and match the item to a catalog of threat items, such as guns, rifles, knives and bombs with metal shrapnel. Without this advanced multi-sensor solution, it becomes nearly impossible to discover a weapon on a person's body before it appears in an assailant’s hands. This multi-sensor solution allows for a touchless, unobtrusive access to a building, but allows for immediate notification to onsite security when a concealed threat is detected. The hidden technology thus empowers security staff to intercept threats before they evolve into a wider scale attack, while also maintaining the privacy and civil liberties of the public, unless, of course, they are carrying a concealed weapon or pose a physical threat. AI-powered solutions proactively help onsite security to effectively safeguard the public Unlike many large, fixed detection security systems, AI-powered solutions proactively help onsite security to effectively safeguard the public without causing mass obstruction and disruption. Soft target hardening        Hardening a facility against physical attacks and threats can be expensive, as well as maintaining and running large fixed detection equipment. It may also result in the threat shifting to ‘softer’, less secure targets, for example schools, music venues and places of worship, all locations we’ve see active attacks in the past decade. Around the world, we have seen the devastation to communities and the aftermath impact of these attacks. In the USA, for example, in February 2018, a 19-year-old gunman walked into Marjory Stoneman Douglas High School in Florida, and opened fire, killing 17 students and faculty members. And in the UK in May 2017, a suicide bomber attacked a concert venue in Manchester, tragically killed 22 people.  It is a global priority to make these soft target public gathering places more secure. But in doing so we cannot turn them into fortresses. The security industry, public sector and national and local government must collaborate to deploy intelligent systems with technology at their core to not only protect lives, but also preserve a way of life. Integrated systems and behavioral detection One of the biggest advantages of using AI technology is that it’s possible to integrate this intelligent software into building smarter, safer communities and cities. Essentially, this means developing a layered system that connects multiple sensors for the detection of visible and invisible threats. Integrated systems mean that threats can be detected and tracked, with onsite and law enforcement notified faster, and possibly before an assault begins to take place. In many ways, it’s the equivalent of a neighborhood watch program made far more intelligent through the use of AI. Using technology in this way means that thousands of people can be screened seamlessly and quickly, without invading their civil liberties or privacy. It’s not only knives, guns and explosives that intelligent systems can detect. They can also be trained to detect behavior and potential invisible biological threats, such as viral threats currently facing our world today. This does not mean profiling individuals.    Instead, using AI that is deployed on existing CCTV or thermal camera systems, it looks for indicators that may identify a physical altercation and disturbance, an elevated body temperature, indicative of viral fever, or lack of a face mask for health safety compliance. When integrated, these solutions can provide onsite security with up-to-the-minute information to allow greater protection of the properties they serve. By using these intelligent, non-intrusive technologies, today’s security personnel are now more capable of detecting a wide range of threats. This is the future of public safety and security, and we should expect to see these new technologies becoming more common over the coming years, as cities around the world strive to create smart, safer communities. The human element While technology can make a significant impact to existing security systems, it would be wrong to position it as the end-all, be all to preventing future attacks. Technology is only part of the solution. Well trained security personnel are also required; individuals who know how to use new technologies and the data they provide, and then make informed decisions about how to engage a potential bad actor or threat. Not only will a properly trained security staff member help to prevent an attack from happening, but the extra insight provides by these intelligent systems can potentially interrupt an attacker in planning and walk through stages, or even before a weapon is drawn. This alone has many benefits beyond just preventing an attack. It means that authorities can help these individuals, some of whom may be suffering from mental health issues, to get the help they need from professional healthcare workers. By security personnel working with local authorities and healthcare professionals, potential attackers can get the support they need, from de-radicalisation programmes to specialized counseling, helping them return to being a healthy, productive member of society. These intelligent systems can potentially interrupt an attacker in planning AI for safer communities AI’s ability to detect visible or invisible threats or behavioral anomalies will prove enormously valuable to many sectors across our global economy. Perhaps none more so, though, than to institutes of education, where we have seen many violent attacks over the course of the last few years. Specifically, the application of AI for detecting odd behavioral activity could be used to identify potential active shooter attacks, or even students who may be depressed and prone to committing suicide. Both tragedies we see weekly around the world. One thing is clear, cross-sector collaboration and the application of integrated, intelligent AI technology that puts data and ultimate control into a human’s hands can be key to making our communities safer places to live.