Voice recognition systems

Facial recognition tech for APAC airports

In today’s fast-evolving aviation landscape, innovation isn’t optional — it’s essential. With passenger traffic in the Asia-Pacific continuing to surge, the region is facing a pivotal moment: adapt with scalable tech, or fall behind. In January 2025 alone, APAC carriers accounted for 56.6% of global passenger traffic growth. Airports are operating at near capacity, with a record-high Passenger Load Factor (PLF) of 82.1%, leading to pronounced congestion at check-in count...

Camden EXIT WATCH: Enhanced delayed egress locks

Camden Door Controls has announced a number of enhancements to its popular CX-DE1200 Series EXIT WATCH™ delayed egress magnetic locks. These include a louder 75 dB voice annunciation, packaging the lock with a 90 dB remote sounder, and adding the option for a cost-effective kit that includes the maglock, remote sounder, and a remote reset key switch. Camden EXIT WATCH™ series Camden EXIT WATCH™ series delayed egress magnetic locks feature a 1200 lb. holding force, rugged al...

HID expands Cardiff office: 40 new jobs in tech hub

HID, the company behind the security card readers and access systems used in office buildings, universities, and government facilities worldwide – including several UK landmarks – has undergone a major expansion of its Cardiff office. The expansion creates 35-40 high-skilled jobs, reinforcing Cardiff’s status as a growing hub for technology and innovation. HID’s Cardiff office The investment has transformed HID’s Cardiff office into a state-of-the-art technology...

HID report: Unified solutions & biometric tech rise

As organisations navigate an increasingly complex threat landscape, security pioneers are making strategic shifts toward unified platforms and emerging technologies, according to the newly released 2025 State of Security and Identity Report from HID. The comprehensive study gathered responses from 1,800 partners, end users, and security and IT personnel worldwide, and reveals a significant transformation in how businesses are approaching security, with mobile credentials and artificial intellig...

Latest aviation innovations at IASEA 2025 Singapore

The 8th edition of Inter Airport Southeast Asia (IASEA) was held in Singapore from 25-27 March 2025. Spanning across 6,259 sqm of exhibition space, the airport exhibition and conference brought together 3,621 industry trade professionals from 53 countries and regions, and featured 134 exhibiting companies, including pavilions representing Germany, Italy, Japan, and Singapore. The event saw a 43% increase in exhibitor size and a 15.6% rise in attendee numbers, underscoring the rising significanc...

RecFaces: The future of APAC airport security

As the world emerges from the pandemic's shadow five years after it spread globally, everything seems to be back on track. Air travel volumes have matched and even surpassed the pre-COVID level. But are they truly “back”? The pandemic did more than just bring masks and border closures. It fundamentally reshaped the world, accelerating digitalisation beyond recognition and irrevocably redefining security paradigms and human behaviour. Heightened security concerns Standard solutions...

View all

Expert commentary

Privacy measures in modern facial recognition tech

In today's rapidly evolving security landscape, facial recognition has become a buzzword that often triggers privacy and data protection concerns. However, many security professionals may not realise that "facial recognition" is an umbrella term encompassing various technologies, each with distinct applications and privacy implications. Let's dive into what these technologies really mean for security system resellers, integrators, and end-users. Facial recognition tech At its core, facial recognition technology re-identifies or verifies individuals based on their facial features, which are used as biometrics. However, not all biometric systems are based on unique identifying features. Some analyse general characteristics like facial hair style or other distinctive marks. Such traits, known as soft biometrics, can aid in identification but aren't unique enough to verify someone's identity uniquely. Personally identifiable information Modern systems incorporate robust safeguards, including data encryption and strict retention policies While facial recognition technology has applications ranging from access control to crime prevention to investigation, its implementation varies widely depending on specific needs. Modern systems incorporate robust safeguards, including data encryption and strict retention policies, to ensure the responsible handling of any personally identifiable information (PII). Understanding key technologies and applications Facial recognition encompasses several distinct technologies, each serving specific purposes. Here's a comprehensive breakdown of these technologies and their real-world applications. Key technologies: Face Verification (1:1): A one-to-one comparison where a person claims an identity (e.g., by showing an ID card), and the system verifies whether the face matches the provided identity. Example: In airports, face verification is used for automated passport control. When a traveler approaches a gate, their face is scanned and compared to the photo stored in the government database. If the face matches, the traveler is allowed through the gate without manual checks. Face Identification (1 to many): A one-to-many comparison, where a face captured by a system is compared to a database of multiple faces and facial features to identify the person. This process is often used in security or surveillance contexts. Example: In the case of a missing child at an airport, a system could scan the faces of all passengers passing through checkpoints and compare them to a photo of the child in a database. If a match is found, it triggers an alert. Face Re-identification (Many to Many): Many-to-many comparisons where multiple faces are compared to multiple other faces. This is typically used to track a person’s movement anonymously across different areas by matching their facial images at different checkpoints, without knowing their identity. Example: In a retail environment, facial re-identification might be used to track how long an anonymous person spends moving from one section of a store to another by re-identifying their face as they enter and leave different camera views. Facial recognition can be used both for real-time and offline applications. Real-Time Facial Recognition: Real-time facial recognition refers to the immediate processing of a live video feed, comparing faces to a database to generate instant alerts when a match is found. Example: At large public events like sports stadiums, real-time facial recognition might be used to detect banned individuals (e.g., known hooligans) as they attempt to enter. Post-Event (Recorded) Facial Recognition: This refers to analysing video recordings after the event has occurred, rather than in real-time. Facial recognition is applied to recorded data to identify or track individuals. Example: After a crime, investigators could use facial recognition software on recorded video from security cameras to identify suspects by matching their faces to known databases. These definitions cover various aspects of facial recognition technology, its different applications, and how biometrics are used for identification and tracking purposes. Biometrics: Biometric technologies use a person’s distinguishing physical characteristics, such as their face, fingerprint, or iris, to identify them. Example: Fingerprint or face scanning for unlocking a phone or using iris recognition for secure entry at high-security buildings like data centers. Hard Biometrics: Hard biometrics refer to physical characteristics that are sufficiently unique enough to be used for identifying a specific individual, such as a face, fingerprint, or iris. Example: Using iris recognition at airport security checkpoints to confirm the identity of a traveler. Soft Biometrics: Soft biometrics (personal features) include general attributes like height or body shape, which are not unique enough to identify a person on their own but can help narrow down re-identification when combined with other information. Example: Using height and body shape to help identify a suspect in a camera scene when facial features alone are unreliable. Appearance Similarity: This refers to distinguishing between people based on their appearance (e.g., clothing, accessories) rather than biometric features. It’s often used for accelerated investigation and statistical analysis rather than identification. Example: A retail store may track customers based on the clothes they are wearing to monitor how long they stay in the store, without tracking their faces or personal details. Liveness Detection: A method used to determine whether the subject in front of a facial recognition system is a live human being and not a photo or a video recording. Example: In some mobile payment systems, facial recognition requires users to blink or move their head slightly to ensure they are a live person and not someone trying to use a photo for authentication. Mathematical Representation: Non-reversible mathematical representations are lists of numbers based on a person's facial image or appearance based on clothing. These numbers represent characteristics but cannot be easily used to recreate the face. Example: When an organisation stores only the mathematical representations from a face rather than an actual image, even if the data is stolen, it is nearly impossible to recreate the person’s face or use the data with another system. Privacy and security considerations Modern facial recognition systems prioritise privacy through various protective measures, moving far beyond the basic security protocols of the past. Solutions integrate multiple layers of protection designed to safeguard personal data while maintaining system effectiveness. These sophisticated privacy controls work in concert to ensure responsible data handling and comply with evolving security standards. Key protective measures include: Biometric template isolation that keeps facial recognition templates separate from other personal data, with dedicated secure storage environments. Template encryption frameworks specifically designed for biometric data, using industry-standard protocols that protect facial features during both processing and storage. Biometric data anonymisation that converts facial features into non-reversible mathematical representations – into numbers - prevents the reconstruction of original face images. Cascading deletion protocols automatically remove both raw facial data and derived biometric templates after their authorised use period. Segmented access controls that separate facial recognition administrative functions (like enrollment and template management) from regular system operation. Privacy standards The key is selecting the right tool for each application and ensuring that personal data is collected The security industry continues to evolve, finding innovative ways to balance effective surveillance with privacy protection. By understanding this comprehensive range of technologies, security professionals can better serve their clients with solutions that address specific needs while maintaining appropriate privacy standards. The key is selecting the right tool for each application and ensuring that personal data is collected only when necessary and protected when it is not. Statistical analysis and pattern recognition The variety of facial recognition applications demonstrates that not all systems require storing personal information. Many modern solutions focus on statistical analysis and pattern recognition rather than individual identification, offering powerful security benefits while respecting privacy concerns. This balance of capability and responsibility represents the future of video security technology.

The intruder within-why humans are the weakest link in your CNI security chain

Security technology has witnessed huge advancements in recent years, particularly for those protecting critical assets or information. Facial and fingerprint recognition, ANPR and even 'mac addresses' or a person of interest's gait, now all make up the technology toolbox of forward-thinking organisations' security policy. However, humans remain the weakest link in any security chain, and the only way to correct this is to eliminate the human burden, argues Richard Hilson, head of sales for security access management specialist, Parking Facilities. Here, he examines the pitfalls in relying upon human intervention in critical national infrastructure (CNI) settings, and the future of biometrics to keep assets, information and personnel safe. Recent Verizon report 74 percent of corps are saying that insider threats are becoming more of a concern for them As with any workplace initiative, be it a simple recycling policy or a corporate password protection directive, technology is only as good as those who implement or operate it. Whilst technology can be fallible and gremlins do arise, it’s never as flawed as us mere humans with our unreliable ‘on/off’ switch. Likewise, we carry the ability to reason, to override procedures, or ignore policy, should we wish. Or as it happens, just make mistakes. In fact, according to a recent Verizon report, two out of three insider attacks happen as a result of negligence, and 74 percent of organisations are saying that insider threats are becoming more of a concern for them. Fundamental security practices Security breaches aren’t limited to external threats either, whether intentionally or not, they can come from within. Humans can, and do, ‘go rogue’, whether that’s pre-meditated criminal or malicious intent, or just by taking shortcuts. While data breaches court most news headlines in this digital era, some of the most significant security risks are those posed when employees neglect fundamental security practices such as sharing passwords or access cards. Employee negligence and insider threats Insider negligence remains one of the pioneering causes of security breaches Insider negligence remains one of the pioneering causes of security breaches. Employees who share passwords or access cards may do so out of convenience, ignorance, or a misplaced sense of trust. Unfortunately, this creates vulnerabilities that are left open to exploitation. When multiple employees share credentials, it becomes difficult to trace actions to a single individual. This lack of accountability can complicate incident investigations and allows malicious activities to go undetected. Emergency services resources Furthermore, the sharing of passwords or access cards means inaccurate accounting of personnel, and in the event of an evacuation or major incident, central IT systems will hold misleading information of employees’ locations which could have a huge impact upon safety and potentially emergency services resources. Employees with malicious intent can exploit shared credentials to carry out unauthorised activities while shifting blame to others, increasing the risk of deliberate sabotage or theft of sensitive data. Even when there is no malicious intent, employees who share access credentials risk unintentionally exposing them to unauthorised individuals, such as contractors, visitors, or external attackers. What happens when employees compromise security? Access cards are designed to limit entry to restricted physical locations. When shared, unauthorised personnel could enter secure areas such as control rooms, rail lines, large construction sites, data centres, power plants or indeed any site meant to be kept secure. This creates opportunities for sabotage, theft, or corporate espionage. Sensitive data held within critical sites, such as blueprints, system controls, and customer records Likewise, shared passwords can lead to unauthorised entry into IT systems, allowing hackers to install malware, ransomware, or spyware. For example, a cybercriminal gaining access to an energy grid system could shut down power to entire regions, causing chaos to millions of people, and disrupting essential services. And not all data breaches are caused by online hackers gaining entry through unsecure firewalls. Sensitive information held within critical sites, such as blueprints, system controls, and customer records, becomes vulnerable when access credentials are shared too, and the disclosure of such information can have a serious impact upon a company’s bottom line, operations, and ultimately its reputation. Eliminating the human burden One way to prevent human error, or to thwart malpractice is to reduce the burden upon employees to be compliant, and eliminate our flaws by using technology that requires no intervention, decision-making or reason. Facial recognition is widely used in the civil world now, despite the concerns of various lobbyists. Used correctly it is not a ‘catch all’, but an instant recognition of persons of interest cross referenced against a database of known suspects. Our car parks are governed by automatic number plate recognition (ANPR) to gain access in and out, while border controls are using advanced biometrics for everything from facial and fingerprint recognition, through to recognition of human characteristics and gait, for both entry and to apprehend. The UK Home Office is even accelerating its transition to digital border management, using biometric technology to improve efficiencies, safety, and to track and capture known or illegal persons. What of the CNI sites? But what of the corporate world? What of the CNI sites, our airports, our national construction developments such as HS2 or our high-rise office spaces? We always ask this very question, and also ‘can you afford a security breach’ in whatever environment you’re in? Because the smart, cloud-based technology being used by governments, law enforcement authorities and Border Force have cascaded down through the civil and corporate worlds. Immediate safety and security Security-conscious corps are diligently removing human error, by eliminating the human burden Security-conscious organisations are now diligently removing human error, by eliminating the human burden. With cloud-managed software not only are access points managed through biometric integration, but it also overcomes the issues mentioned earlier around accountability - in that the cloud will always register who has passed through an access point, or out of it. This is critical for both immediate safety and security but also for matters arising from a crisis, or emergency situation. It’s also less admin-heavy, more cost-effective and can manage and store employee records, including background checks. Employee or contractor data is encrypted, their information is safe, their interactions are secure, and businesses are protected. Implement robust access management systems Simple acts of negligence, such as sharing passwords or access cards, can open the door to catastrophic consequences, operational disruption, financial loss, and even, in the case of CNI, national security risks. To mitigate this, organisations can do worse than to implement robust access management systems, and in doing so, release employees from having to be accountable for ensuring the security of the sites in which they work. As artificial intelligence evolves, I see even more robust biometrics coming to the fore, until such time we work and live in environments that are controlled without us even knowing security checkpoints are all around us, and access management happening at every step. It will become as ‘every day’ as an automatic door allowing entry into our local supermarket - but we’re not quite there yet.

Why biometrics are key to a safe Paris Olympics

The European Union’s GDPR (General Data Protection Regulation) framework represents some of the toughest and most secure privacy laws in the world. Their creation signaled the EU’s dedication to individual privacy rights and generated new standards and requirements for an organisation inside or outside of the EU wanting to do business in the region. This privacy stance raises an interesting question: to what extent (if any) will France be utilising AI-assisted security and biometrics - at the upcoming Paris Summer Olympic Games? AI-assisted security systems As France prepares to host this global event, certain AI-assisted security systems like algorithmic video monitoring have been tested and are planned to be in use. The country also passed a law with a legal framework for addressing provisions of hosting the Games in their country, Law No. 2023-380. These technologies will support the police and security efforts to detect sudden crowd movements, irregular vehicle, and pedestrian movements, abandoned objects, people lying on the ground, and more, any of which could indicate an attack. Local connectivity One recent French poll found that an overwhelming number of respondents support the use of smart cameras Such technologies, which can be connected directly to localised command centers and police departments, provide an opportunity to prevent deadly incidents like the bombing at the 1996 Summer Olympics in Atlanta, GA. There are signs that citizens of the EU are becoming more receptive to such measures, with one recent French poll finding that an overwhelming number of respondents (89 percent) support the use of smart cameras in stadiums for this purpose. Biometric technology The general public and private use of biometric technology (face matching, face liveness, and other biometric modalities), on the other hand, continues to run into challenges with adoption in this privacy-focused region. For the most part, laws ban facial recognition, permitting its use only once someone has been convicted or is suspected of having committed a serious crime. Limitations of video surveillance Limiting the utility of biometrics to only video surveillance for an event of the Paris Olympics’ magnitude is not the best approach. The ability to quickly identify or de-identify possible suspects from regular attendees greatly strengthens the activities of the police and helps protect the rights of regular attendees. Consider the example of New York City detectives using facial recognition to identify a man who, in 2019, left a pair of potential bombs in the Fulton Street subway station. Benefits of facial recognition technology Many believe that the use of facial recognition could have expedited and solved the Boston Marathon bombing tragedy With facial recognition technology, it took only one hour to identify the suspect - a process that previously would have taken several hours or even days, leaving the possibility of further damage unresolved. Citing yet another example, many believe that the use of facial recognition could have expedited and solved the Boston Marathon bombing tragedy much sooner and more efficiently, aiding investigators who reportedly had to sift through 120,000 photos and nearly 13,000 videos before identifying the perpetrator. Need for real-time video With an estimated 16 million visitors expected to descend upon France for the Games - and an environment that is harder to protect from bad actors - is limiting the use of biometrics to video surveillance the most effective? In our minds, in certain high-urgency cases, such as tracking kidnapping cases or possible terrorism activity, it should be permissible for facial recognition to be applied to recorded video in real-time and without delay. Public safety Beyond this obvious good that biometrics can bring from a public safety perspective, there are other ways that biometrics can be woven into the fabric of the Summer Olympics to support a faster, more enjoyable experience for all. Consider queue management. In a few weeks, an estimated 2.3 to 3.1 million ticket holders will form lines to access Olympic events. Fingerprint scanning technology People could pass through lines in mere fractions of a second, versus having to fumble for a paper ticket If fingerprints were captured at the point of ticket purchase and fingerprint scanning technologies made available on-site, people could pass through lines in mere fractions of a second, versus having to fumble for a paper ticket that may be torn or damaged, or an e-ticket on a phone with a dead battery. This would speed up the processing of queues tremendously. Olympics-focused gambling activity Another area where biometrics can be beneficial is in Olympics-focused gambling activity, which is expected to gain steam as the Games get underway. Biometrics can be used to match facial images with legal documentation for age verification, and then combined with geofencing to ensure gambling participants are of legal age, depending on where they are physically based. Biometrics also makes it possible to identify, beyond the shadow of a doubt, all of the activities of particular individuals, which can help track nefarious activities like money laundering. Evolution of privacy and security We believe biometrics can be a force for major good in our society and around various facets of the upcoming Paris Olympics, most notably public safety. It will be interesting to watch both the 2024 Summer Olympics and the ongoing evolution of privacy and security laws and practices in France and the European Union.

View all

Security beat

Anviz Global expands palm vein tech for security

The pattern of veins in the hand contains unique information that can be used for identity. Blood flowing through veins in the human body can absorb light waves of specific wavelengths. Irradiating the human palm with near-infrared light waves yields an image of the vascular pattern. A venous distribution map can be processed and compared to pre-registered data to match and confirm identity. Palm vein biometrics The idea of palm vein biometrics goes back to the 1980s when palm vein scanners emerged as commercial products in the 1990s. The devices gained some traction in military-grade and high-security applications, although the technology was too expensive to become popular among mainstream applications such as physical access control. Anviz Global is expanding palm vein technology into mainstream applications, featuring a lower price point that will open new market opportunities. Palm vein technology “Palm vein technology has been viewed as a boutique product because it is costly and has not developed as a mainstream product,” says Mark Vena, Senior Director of Business Strategy and Development for Anviz Global. “With a more compelling price point, the market can see the technology more broadly. We can change the game in terms of how people think about palm vein technology.” FAR and FRR Palm vein technology is more accurate than either fingerprint (0.0001% FAR) or facial recognition (0.001% FAR) Iris recognition is considered the gold standard in biometric identification, but palm vein technology can almost achieve the accuracy of iris. The false accept rate (FAR) for palm vein technology is 0.00008%, compared to 0.00005% for iris recognition systems. The false reject rate (FRR) is comparable at 0.01% for either technology. Palm vein technology is more accurate than either fingerprint (0.0001% FAR) or facial recognition (0.001% FAR). Palm vein identification occurs in less than half a second. M7 Palm “Anviz is seeking to bring a high level of capability to more mainstream commercial applications,” says Vena, specifically providing a lower-cost palm vein unit. The M7 Palm by Anviz can achieve next-generation biometric access control for greater security and intelligence. The door-mounted unit combines palm vein recognition with an RFID card reader and PIN code (using a 17-button keypad), all configurable for multi-factor authentication. Outdoor applications The robust unit, including a narrow metal exterior design, provides strong vandal resistance (IK10) and stability for outdoor applications. To ensure a weatherproof design, the IP66 rating indicates the product is completely protected against dust and can withstand strong jets of water from any direction. Power-over-ethernet (PoE) enables centralised power management and the ability to remotely reboot devices. Wiegand-out interface The unit communicates using RS-485 or TCP/IP and can connect locks, exit buttons, door contacts, doorbells, etc The unit communicates using RS-485 or TCP/IP and can connect locks, exit buttons, door contacts, doorbells, etc. The Wiegand-out interface can connect to standard third-party controllers, or a built-in relay can trigger a lock directly. Capacity is 500 users. Because palm vein recognition originates from the physiological characteristics of the human body, it represents a “credential” that cannot be forged, lost, or left at home. It is not easily counterfeited or worn out. Multi-factor authentication Applications include industries requiring high security such as government, judicial, and banking. Multi-factor authentication eliminates fraud and ensures safety at banks, data centers, airports, prisons, and government. An end user might opt for higher levels of access control for a laboratory or server room. For dealers and integrators, palm vein technology, including multi-factor authentication, provides an additional option for customers seeking to protect high-security areas on a large campus. Card readers Card readers may suffice for most of the doors, but an important laboratory or cash-handling area requires more protection. Palm vein technology can secure a casino cash cage or a room containing corporate secrets. Time of Flight ToF laser-ranging uses a laser to measure the distance to an object by calculating the time Time of Flight (ToF) laser-ranging achieves accurate measurement to ensure the right distance for operation. ToF laser-ranging uses a laser to measure the distance to an object by calculating the time it takes for the laser light to travel to the object and back. Palm vein technology operates in the range of 10 to 30cm (4 to 12 inches). The proprietary BioNANO palm vein recognition algorithm enables accurate and fast scanning speed. Non-contact biometric The non-contact biometric performs in touchless environments and does not contribute to lingering health concerns in the post-COVID world. The technology is accurate across all hand types and skin tones. Users may feel more comfortable scanning their hands than their eyes. Optimal human-machine interaction M7 Palm’s optimal human-machine interaction and user experience include a low-power-consumption OLED screen to deliver clear text notifications to the user. Advantages of M7 Palm include accuracy, stability, security, and privacy. Features include easy installation with standard processes and no special tools, centralised power management, simplified cabling, and less maintenance. Managing the system is expedited by integration with the Anviz CrossChex Standard access control and time management dashboard, used by more than 30,000 enterprises, and the system can be integrated with any access control system. Security for SMBs Anviz Global is a converged intelligent security provider for SMBs and enterprise organisations Anviz Global is a converged intelligent security provider for small- and medium-sized businesses (SMBs) and enterprise organisations. The company’s product line includes biometrics, video surveillance, and security management solutions based on the cloud, Internet of Things (IoT) and artificial intelligence (AI) technologies. Privacy concerns Anviz is part of the Xthings family of companies, which includes Ultraloq (smart locks), Utec (smart home), Bright (smart lighting), Secu365 (SaaS Platform), and Ulticam (smart cameras). Privacy concerns are minimised because palm vein recognition does not directly store the original image but rather extracts feature information through algorithmic models. Personal data is protected through obfuscation and encryption.

As big tech impacts physical access control, there may be more to come

In the competitive world of physical access control, Big Tech companies are seeking to play a larger role. Physical access competition Apple Wallet continues to stake its claim on mobile credentialing. Amazon One Enterprise is pushing a palm-based identity service. Google/Nest offers smart locks for home access control, with identity and access management provided in the Google Cloud. The entry of these big companies in the historically fragmented physical access control market is causing disruption and foreboding new levels of competition. Apple Wallet impacting credentialing trends The popularity of mobile wallets and contactless technologies in general has grown, creating more demand At Apple’s Worldwide Developer Conference in June 2021, the company announced support for home, office and hotel keys, including corporate badges and student ID cards, in Apple Wallet. Later, the company announced Hyatt as the first hotel partner to support the technology. Since then, the popularity of mobile wallets and contactless technologies in general has grown, creating more demand for a seamless solution such as Apple Wallet. Easy access In 2023, HID Global announced the availability of their employee badge in Apple Wallet, allowing staff and guests to easily access corporate spaces with their iPhone or Apple Watch, including doors, elevators, turnstiles, etc. Employees just need to hold their iPhone or Apple Watch near the reader to unlock. Factors affecting the rate of adoption However, implementing and maintaining an Apple Wallet-based access control system can incur costs for hardware updates, software licencing, and ongoing maintenance. Factors affecting the rate of adoption include the need to upgrade existing infrastructures to accommodate the technology, and the necessity for access control manufacturers to develop and implement integrations with Apple Wallet. Benefits of adoption Keys in Apple Wallet take full advantage of the privacy and security built into the iPhone and Apple Watch With larger companies leading the way, some smaller ones might take longer to catch up. There is also a need to educate building owners and administrators to see the value and benefits of switching to Apple Wallet-based access control. Convenience and greater security can accelerate adoption. Keys in Apple Wallet take full advantage of the privacy and security built into iPhone and Apple Watch. Sensitive data protection A compatible app, specific to the building’s access control system, is needed. Once added, credentials are securely stored in the iPhone's Secure Enclave, a dedicated hardware chip designed for sensitive data protection. Holding an iPhone near an NFC-enabled reader enables transmission of encrypted credentials. In addition to Near Field Communication (NFC), some systems also utilise Bluetooth Low Energy (BLE) for added security, longer read range, and hands-free unlocking. Phones need sufficient battery charge to function. Amazon One Enterprise Enables Palm-Based Biometrics In November 2023, Amazon Web Services Inc. (AWS) announced an identity service providing comprehensive and easy-to-use authentication for physical and digital access control. The system enables users to employ their palm as an access control credential, allowing organisations to provide a fast and contactless experience for employees and others to gain access to physical locations as well as digital assets. Physical and digital locations Physical locations include data centres, office and residential buildings, airports, hotels, resorts Physical locations include data centres, office and residential buildings, airports, hotels, resorts, and educational institutions. IT and security administrators can easily install Amazon One devices and manage users, devices, and software updates using AWS’s Management Console. Elimination of physical credentials An advantage of the Amazon approach is the elimination of physical credentials such as fobs and badges, and digital elements such as personal identification numbers (PINs) and passwords. AWS says security is built into every stage of the service, from multi-layered security controls in the Amazon One device, which is the same technology used in the Amazon Go retail stores, where shoppers can pay for purchases by scanning the palm of their hands. The devices combine palm and vein imagery for biometric matching and deliver an accuracy rate of 99.9999%, which exceeds the accuracy of other biometric alternatives, says the company. AI and ML The palm-recognition technology uses artificial intelligence and machine learning to create a “palm signature” that is associated with identification credentials such as a badge, employee ID or PIN. Boon Edam, a revolving door and turnstile manufacturer, offers Amazon’s palm biometric technology on its equipment, and IHG Hotels & Resorts uses the technology to provide employees a convenient way to identify themselves and gain access to software systems. Google and Nest Devices in Access Control When the Nest × Yale Lock is connected to the Nest app, a resident can unlock a door from their phone Google’s Nest devices include smart locks for home access control. The Google Nest × Yale Lock allows access control via both physical keys and passcodes accessible through the Google Home app. When the Nest × Yale Lock is connected to the Nest app, a resident can unlock a door from their phone. Passcodes can be created for family, guests, and other trusted persons. Alerts can be provided whenever someone unlocks and locks the door. When Nest “knows” a resident is away, the door can lock automatically. Voice control, Google Home app Voice control, using Google Assistant integrated with various Nest devices, enables use of voice commands to lock and unlock doors, thus adding another level of convenience. Smart home devices from various manufacturers can be controlled through the Google Home app. SMART Monitoring ADT’s Self Setup smart home security systems integrate Google Nest smart home products with ADT security and life safety technology, including SMART Monitoring technology. Microsoft Azure is another company that could impact access control. The Microsoft Azure Active Directory is an identity and access management platform that could be extended to physical access control, leveraging existing user credentials. Long-Range Impact on the Security Marketplace Big Tech companies are creating platforms for managing access control data, integrating with other security systems Increasingly, Big Tech companies are creating platforms for managing access control data, integrating with other security systems, and offering analytics for optimising security and building operations. Big Tech is also actively researching and developing new technologies for access control, such as facial recognition, voice authentication, and AI-powered anomaly detection. Access control communication and integration As their involvement in physical access control grows, Big Tech companies could potentially gain more influence in setting industry standards for access control communication and integration, similar to how they have become dominant in other areas such as mobile platforms. Given their expertise in user interface design and data analysis, Big Tech companies could help to direct how future access control systems are managed and how users interact with them, including more intuitive and user-friendly operations. Future of physical access control Existing concerns about privacy, security, and potential dominance by a few Big Tech companies could spill over into physical access control. However, traditional security companies, startups, and industry consortiums are also actively developing innovative solutions. Ultimately, the future of physical access control will likely be shaped by a combination of many different players and technologies – large and small.

Newest systems-on-chips (SOCs) to expand AI inside cameras, says Ambarella

When it comes to security cameras, the end user always wants more—more resolution, more artificial intelligence (AI), and more sensors. However, the cameras themselves do not change much from generation to generation; that is, they have the same power budgets, form factors and price. To achieve “more,” the systems-on-chips (SoCs) inside the video cameras must pack more features and integrate systems that would have been separate components in the past. For an update on the latest capabilities of SoCs inside video cameras, we turned to Jérôme Gigot, Senior Director of Marketing for AIoT at Ambarella, a manufacturer of SOCs. AIoT refers to the artificial intelligence of things, the combination of AI and IoT. Author's quote “The AI performance on today’s cameras matches what was typically done on a server just a generation ago,” says Gigot. “And, doing AI on-camera provides the threefold benefits of being able to run algorithms on a higher-resolution input before the video is encoded and transferred to a server, with a faster response time, and with complete privacy.” Added features of the new SOC Ambarella expects the first cameras with the SoC to emerge on the market during early part of 2024 Ambarella’s latest System on Chip (SOC) is the CV72S, which provides 6× the AI performance of the previous generation and supports the newer transformer neural networks. Even with its extra features, the CV72S maintains the same power envelope as the previous-generation SoCs. The CV72S is now available, sampling is underway by camera manufacturers, and Ambarella expects the first cameras with the SoC to emerge on the market during the early part of 2024. Examples of the added features of the new SOC include image processing, video encoders, AI engines, de-warpers for fisheye lenses, general compute cores, along with functions such as processing multiple imagers on a single SoC, fusion among different types of sensors, and the list goes on. This article will summarise new AI capabilities based on information provided by Ambarella. AI inside the cameras Gigot says AI is by far the most in-demand feature of new security camera SoCs. Customers want to run the latest neural network architectures; run more of them in parallel to achieve more functions (e.g., identifying pedestrians while simultaneously flagging suspicious behavior); run them at higher resolutions in order to pick out objects that are farther away from the camera. And they want to do it all faster. Most AI tasks can be split between object detection, object recognition, segmentation and higher-level “scene understanding” types of functions, he says. The latest AI engines support transformer network architectures (versus currently used convolutional neural networks). With enough AI horsepower, all objects in a scene can be uniquely identified and classified with a set of attributes, tracked across time and space, and fed into higher-level AI algorithms that can detect and flag anomalies. However, everything depends on which scene is within the camera’s field of view. “It might be an easy task for a camera in an office corridor to track a person passing by every couple of minutes; while a ceiling camera in an airport might be looking at thousands of people, all constantly moving in different directions and carrying a wide variety of bags,” Gigot says. Changing the configuration of video systems Low-level AI number crunching would typically be done on camera (at the source of the data) Even with more computing capability inside the camera, central video servers still have their place in the overall AI deployment, as they can more easily aggregate and understand information across multiple cameras. Additionally, low-level AI number crunching would typically be done on camera (at the source of the data). However, the increasing performance capabilities of transformer neural network AI inside the camera will reduce the need for a central video server over time. Even so, a server could still be used for higher-level decisions and to provide a representation of the world; along with a user interface for the user to make sense of all the data. Overall, AI-enabled security cameras with transformer network-based functionality will greatly reduce the use of central servers in security systems. This trend will contribute to a reduction in the greenhouse gases produced by data centres. These server farms consume a lot of energy, due to their power-hungry GPU and CPU chips, and those server processors also need to be cooled using air conditioning that emits additional greenhouse gases. New capabilities of transformer neural networks New kinds of AI architectures are being deployed inside cameras. Newer SoCs can accommodate the latest transformer neural networks (NNs), which now outperform currently used convolutional NNs for many vision tasks. Transformer neural networks require more AI processing power to run, compared to most convolutional NNs. Transformers are great for Natural Language Processing (NLP) as they have mechanisms to “make sense” of a seemingly random arrangement of words. Those same properties, when applied to video, make transformers very efficient at understanding the world in 3D. Transformer NNs require more AI processing power to run, compared to most convolutional NNs For example, imagine a multi-imager camera where an object needs to be tracked from one camera to the next. Transformer networks are also great at focussing their attention on specific parts of the scene—just as some words are more important than others in a sentence, some parts of a scene might be more significant from a security perspective. “I believe that we are currently just scratching the surface of what can be done with transformer networks in video security applications,” says Gigot. The first use cases are mainly for object detection and recognition. However, research in neural networks is focussing on these new transformer architectures and their applications. Expanded use cases for multi-image and fisheye cameras For multi-image cameras, again, the strategy is “less is more.” For example, if you need to build a multi-imager with four 4K sensors, then, in essence, you need to have four cameras in one. That means you need four imaging pipelines, four encoders, four AI engines, and four sets of CPUs to run the higher-level software and streaming. Of course, for cost, size, and power reasons, it would be extremely inefficient to have four SoCs to do all this processing. Therefore, the latest SoCs for security need to integrate four times the performance of the last generation’s single-imager 4K cameras, in order to process four sensors on a single SoC with all the associated AI algorithms. And they need to do this within a reasonable size and power budget. The challenge is very similar for fisheye cameras, where the SoC needs to be able to accept very high-resolution sensors (i.e., 12MP, 16MP and higher), in order to be able to maintain high resolution after de-warping. Additionally, that same SoC must create all the virtual views needed to make one fisheye camera look like multiple physical cameras, and it has to do all of this while running the AI algorithms on every one of those virtual streams at high resolution. The power of ‘sensor fusion’ Sensor fusion is the ability to process multiple sensor types at the same time and correlate all that information Sensor fusion is the ability to process multiple sensor types at the same time (e.g., visual, radar, thermal and time of flight) and correlate all that information. Performing sensor fusion provides an understanding of the world that is greater than the information that could be obtained from any one sensor type in isolation. In terms of chip design, this means that SoCs must be able to interface with, and natively process, inputs from multiple sensor types. Additionally, they must have the AI and CPU performance required to do either object-level fusion (i.e., matching the different objects identified through the different sensors), or even deep-level fusion. This deep fusion takes the raw data from each sensor and runs AI on that unprocessed data. The result is machine-level insights that are richer than those provided by systems that must first go through an intermediate object representation. In other words, deep fusion eliminates the information loss that comes from preprocessing each individual sensor’s data before fusing it with the data from other sensors, which is what happens in object-level fusion. Better image quality AI can be trained to dramatically improve the quality of images captured by camera sensors in low-light conditions, as well as high dynamic range (HDR) scenes with widely contrasting dark and light areas. Typical image sensors are very noisy at night, and AI algorithms can be trained to perform excellently at removing this noise to provide a clear colour picture—even down to 0.1 lux or below. This is called neural network-based image signal processing, or AISP for short. AI can be trained to perform all these functions with much better results than traditional video methods Achieving high image quality under difficult lighting conditions is always a balance among removing noise, not introducing excessive motion blur, and recovering colours. AI can be trained to perform all these functions with much better results than traditional video processing methods can achieve. A key point for video security is that these types of AI algorithms do not “create” data, they just remove noise and clean up the signal. This process allows AI to provide clearer video, even in challenging lighting conditions. The results are better footage for the humans monitoring video security systems, as well as better input for the AI algorithms analysing those systems, particularly at night and under high dynamic range conditions. A typical example would be a camera that needs to switch to night mode (black and white) when the environmental light falls below a certain lux level. By applying these specially trained AI algorithms, that same camera would be able to stay in colour mode and at full frame rate--even at night. This has many advantages, including the ability to see much farther than a typical external illuminator would normally allow, and reduced power consumption. ‘Straight to cloud’ architecture For the cameras themselves, going to the cloud or to a video management system (VMS) might seem like it doesn’t matter, as this is all just streaming video. However, the reality is more complex; especially for cameras going directly to the cloud. When cameras stream to the cloud, there is usually a mix of local, on-camera storage and streaming, in order to save on bandwidth and cloud storage costs. To accomplish this hybrid approach, multiple video-encoding qualities/resolutions are being produced and sent to different places at the same time; and the camera’s AI algorithms are constantly running to optimise bitrates and orchestrate those different video streams. The ability to support all these different streams, in parallel, and to encode them at the lowest bitrate possible, is usually guided by AI algorithms that are constantly analyzing the video feeds. These are just some of the key components needed to accommodate this “straight to cloud” architecture. Keeping cybersecurity top-of-mind Ambarella’s SoCs always implement the latest security mechanisms, both hardware and software Ambarella’s SoCs always implement the latest security mechanisms, both in hardware and software. They accomplish this through a mix of well-known security features, such as ARM trust zones and encryption algorithms, and also by adding another layer of proprietary mechanisms with things like dynamic random access memory (DRAM) scrambling and key management policies. “We take these measures because cybersecurity is of utmost importance when you design an SoC targeted to go into millions of security cameras across the globe,” says Gigot. ‘Eyes of the world’ – and more brains Cameras are “the eyes of the world,” and visual sensors provide the largest portion of that information, by far, compared to other types of sensors. With AI, most security cameras now have a brain behind those eyes. As such, security cameras have the ability to morph from just a reactive and security-focused apparatus to a global sensing infrastructure that can do everything from regulating the AC in offices based on occupancy, to detecting forest fires before anyone sees them, to following weather and world events. AI is the essential ingredient for the innovation that is bringing all those new applications to life, and hopefully leading to a safer and better world.

View all

Round table discussion

What’s new with biometrics?

Technological leaps in the last several decades have revolutionised biometrics. The technologies are constantly evolving, spanning facial recognition to iris scanning to fingerprints, to provide new levels of security and convenience. Biometrics are everywhere, from smartphones to border control, constantly evolving to meet the needs of our increasingly digital world. They are also more accurate and easier to use than ever. We asked this week's Expert Panel Roundtable: What’s new with biometrics?

How can security technologies protect lone workers?

There is safety in numbers, or so the expression goes. Generally speaking, several employees working together tend to be safer than a single employee working alone. Even so, some environments require that workers complete their jobs alone, thus presenting a unique combination of security vulnerabilities. The U.S. Occupational Safety and Health Administration (OSHA) defines a lone worker as “an employee working alone, such as in a confined space or isolated location.” We asked this week’s Expert Panel Roundtable: How can security technologies help to protect "lone workers?"

Are schools safer because of physical security systems?

Headlines of violence in our schools are a reminder of the need to keep educational institutions safe. In fact, if there is a positive aspect to the constant bombardment of headlines, it is that it keeps our attention perpetually focused on how to improve school security. But what is the role of physical security systems? As the new school year begins, we asked this week’s Expert Panel Roundtable: Are schools safer because of physical security systems? Why or why not?

View all