BT/ Apple patent points to brainwave sensors for headset

Paradigm

Follow

Published in

Paradigm

24 min readMar 25, 2024

--

Biometrics biweekly vol. 85, 11th March — 25th March

TL;DR

Apple is working on technology that could turn the Apple Vision Pro into a brainwave reader to improve mental health, assist with training and workouts, and help with mindfulness
Suprema and Qualcomm power improved under-display biometrics in the new Samsung Galaxy. Samsung Wallet now supports mobile driver licenses in South Korea
EU Court approves fingerprint mandate for ID cards. EU cybersecurity agency outlines good practices for remote identity proofing. Regulators aim for frictionless age verification, interoperability
Mastercards from Thales and FPC, Idex Biometrics launch in Turkey
Smart Engines claims new facial recognition system uses no biometric information
Precise adds palm biometrics as a third modality for access control, developers
Paravision’s biometric age estimation released to expand online safety market
Facia 3D biometric liveness detection passes iBeta Level 2 PAD compliance test
Patent for ultrasonic fingerprint biometrics among thousands of Huawei files each year
A new crop of latent fingerprint algorithms set new highs in NIST testing
Facia 3D biometric liveness detection passes iBeta Level 2 PAD compliance test
ZeroBiometrics integrates ROC face biometrics and liveness detection
Sumsub integrates novel key management capabilities for crypto compliance
Trulioo unveils identity verification platform network growth, customer successes
Worldcoin makes core components of iris biometric imaging software open-source
GSA names 8 identity proofing companies for $194.5M in Login.gov contracts
2.7M people in India’s Assam state may get digital ID after activation of citizenship law
Norwegian BankID digital ID is now biometric, available on smartphones
Australia’s national digital identity receives new name and logo
Toddlers to get digital ID in Philippines, Caymans
Online-only processing of biometric IDs proposed in Israel
Peru partners with MOSIP for national digital ID pilot program
Nigeria, Rwanda use digital IDs for social welfare payments
Research shows improved facial recognition accuracy and low-resolution performance
Humans are more likely to notice fakes among familiar faces, research suggests
Biometric industry events. And more!

Biometrics Market

The Biometric system market size is projected to grow from USD 36.6 billion in 2020 to USD 68.6 billion by 2025; it is estimated to grow at a CAGR of 13.4% during the forecast period. Increasing use of biometrics in consumer electronic devices for authentication and identification purposes, the growing need for surveillance and security with the heightened threat of terrorist attacks, and the surging adoption of biometric technology in automotive applications are the major factors propelling the growth of the biometric system market.

Biometric Research & Development

Latest Research:

Research shows improved facial recognition accuracy and low-resolution performance

The rise of biometric identification and verification systems deployed in the edge infrastructure has created a gap in accurate and efficient facial recognition models. Traditional facial recognition technology based on deep convolutional neural networks (DCNN) has limitations. These include susceptibility to external factors like occlusions, variations in lighting conditions, and facial expressions, which can compromise the accuracy of identification. Therefore, engineers need a new method that can overcome these challenges.

Engineers need a face attribute recognition technology that uses optimized feature extraction and fusion techniques to improve accuracy further. These techniques involve extracting unique features from facial images and their combination to create a comprehensive representation of the facial attributes. In addition, the software model must support improvements in information flow between features at different scales. The paper proposes a solution to these challenges.

The new facial recognition method from Yizhuo Gao of China’s Jilin Police College’s Department of Criminal Science and Technology utilizes bilinear pooling in a DCNN to extract facial features. It involves setting up networks at three different scales to capture multiscale features. This multiscale approach ensures that features are extracted at various resolutions, capturing global and local attributes of the face. Features extracted from different convolutional layers within the network are combined to form a holistic feature set.

*Architecture of facial recognition method using the deep convolutional neural network*

Figure shows the architecture process for a facial expression recognition system, which begins with a database of face images intended for training the model. The model has six components.

The initial face pre-processing stage involves pre-processing the image, which can be done through face detection and improvement in image quality or cropping the image only to include the face to prepare the data for feature learning.

At the feature learning step, the network processes individual images to learn various facial expressions. Techniques such as rotation, scaling, and color adjustment diversify the training dataset. The images are then passed through two separate convolutional neural networks operating in parallel for feature extraction. These DCNNs, often used for transfer learning, combine the extracted features to improve the model’s performance.

Following feature learning, the combined output is fed to a softmax layer, responsible for the final prediction in the classification tasks.

In parallel to the learning, a pre-trained model analyzes the input image from the database to make a prediction. This model prediction is integrated with the output from the softmax layer to enhance prediction accuracy.

The next step is expression recognition, the final step where the facial expression is classified.

The model then outputs a classification score for each class of facial expression, which, after the softmax layer, can be interpreted as the probability that indicates the likelihood of each facial expression.

The study compared the performance of their facial recognition technique with existing algorithms, such as VGG16-SSN, VGG16-PSN, and APS, focusing on attributes such as physical facial features and overall facial structure. The results shown in Figure 2 reveal that the average accuracy rates for VGG16-SSN, VGG16-PSN, APS, and the proposed method are 86.79 percent, 87.13 percent, 91.55 percent, and 97.11 percent, respectively.

The paper credits its method’s superior accuracy to its use of global and local features, facilitated by a combination of a shared sub-network and two task-specific sub-networks. This approach not only enhances accuracy over the APS algorithms but also addresses the challenges of low-resolution facial recognition.

The research examines how the method performs across a spectrum of resolutions, ranging from 15×15 to 100×100 pixels. Recognizing faces at lower resolutions, such as 15×15 to 30×30, is particularly challenging due to the lack of image details. However, the paper’s algorithm achieved a 54.03 percent accuracy rate at the lowest resolution of 15×5. Figure 3 presents data on the relationship between image resolution and the accuracy rate.

*Comparison of accuracy rate for different resolutions*

In today’s biometric applications, there is a constant need for power-efficient and highly accurate facial recognition models suitable for deployment on resource-constrained edge devices. Where traditional CNN-based models fell short, according to Gao, the new generation of facial recognition methods excels by leveraging detailed facial attributes for both global and local feature extraction. Implementing these algorithms ensures the desired level of accuracy in challenging environments where external factors may otherwise compromise the performance.

Humans more likely to notice fakes among familiar faces, research suggests

Deepfakes and synthetic data, and how they impact biometric systems were the focus of the Norwegian Biometrics Laboratory Annual Workshop 2024, hosted by the EAB last week.

Presentations discussed the benefits, challenges, and security threats to facial recognition from synthetic data.

Sascha Frühholz of the University of Oslo presented ongoing research being conducted in partnership with several other scientists under the title: “Human Cognitive and Neural Mechanisms for Identity Recognition in Synthetic Faces.”

Their work examines how people perceive identity across synthetic images in comparison to real faces, including what is happening from a neuroscience perspective.

Some past work indicates that people are capable of discriminating between synthetic and real people, and the brain sometimes indicates it can tell a difference even when the individual cannot consciously do so. People’s ability to tell the real and fake apart, however, is modest at best, and some studies indicate it is negligible, or even non-existent.

Research by Frühholz and partners bears out the latter point, that high-quality synthetic faces are judged by people as real nearly as often as genuine ones.

To advance the research, they sought “synthetic faces that had a certain percentage level of similarity to the original face.” Faces were generated along a continuum of similarity, from 0 to 100 percent similarity, at intervals of 20 percent. These were provided by Sebastian Marcel of Idiap.

Faces were also divided between familiar (celebrities) and unfamiliar ones.

They found that people took the longest to judge faces at 40 to 60 percent similarity and the shortest amount of time with those of 100 percent or 0 percent similarity.

While people tend to correctly match faces with 100 percent similarity and perceive the difference between faces with low similarity, Frühholz says there are insights to be gained from the middle of the curve. People were more conservative with matching familiar faces to those with similarity between 20 and 80 percent, perhaps more easily noticing differences. For unfamiliar faces with 40 percent similarity, study participants were still slightly more likely than not to identify the images as the same person.

The preliminary examination of brain activity (in the fusiform face area, occipital face area and superior temporal sulcus) during these processes suggests that different processes are happening when familiar and unfamiliar faces are being assessed as matching or not, at least for difficult judgments.

The different areas of the brain are associated with different functions. The FFA analyzes faces holistically. The OFA is active when people consider the details of a face, as in more difficult comparisons. The STS is active when the individual tries to place a face into a social context. Frühholz’ research indicates that the OFA is used more for difficult matching decisions with unfamiliar faces, while the FFA and STS are activated in different degrees according to the similarity or dissimilarity of the faces presented.

A deepfake detection collaboration between PXL Vision and the Idiap Research Institute was announced earlier this week.

Main News:

Apple patent points to brainwave sensors for headset

A new Apple patent suggests the company is developing technology that could integrate brainwave and biometric sensors into its Apple Vision Pro headset or future AR/VR devices, aiming to enhance mental and physical health. The patent outlines a brain-computer interface capable of monitoring various bodily systems, including heart, lungs, and brain activity, to support mental health, mindfulness, and physical training without directly reading thoughts.

The envisioned sensors could offer new insights into a user’s health and activity, akin to how the Apple Watch tracks physical metrics. Applications could range from aiding trauma therapy to assisting individuals with neurodivergent learning.

Apple Vision Pro could someday scan your brain and body to help improve your health.

Apple is working on technology that could turn the Apple Vision Pro into a brainwave reader to improve mental health…

appleinsider.com

Samsung Galaxy smartphones integrate Suprema biometrics

Suprema has supplied its latest in-display fingerprint recognition technology, BioSign 6.0, to Samsung for the Galaxy S24 series. The technology uses the Qualcomm 3D Sonic Sensor Gen 2 for fingerprint recognition. Suprema has been providing fingerprint recognition tech for Samsung Galaxy S series phones since 2019. BioSign 6.0 boasts significant improvements to both speed and accuracy when compared to the previous version. This is thanks in part to an optimized AI-based fingerprint analysis algorithm. Suprema believes this technology will lead to wider adoption in the market due to its superior performance and user experience.

Precise adds palm recognition to biometric portfolio

Precise Biometrics has expanded its biometric technology portfolio to include palm recognition, complementing its existing fingerprint and facial recognition modalities. The technology will be integrated into Precise Biometrics’ YOUNiQ Access and YOUNiQ Visit products, and is designed to work with a range of devices, including IP cameras and tablets. The introduction of palm recognition also enables multi-factor authentication solutions by allowing the use of multiple biometric modalities, enhancing security and providing customizable options for organizations. Precise Biometrics has partnered with Hand.ID, an American company specializing in palm recognition readers, to further develop and deploy this technology.

EU Court approves fingerprint mandate for ID cards

The Court of Justice of the European Union has ruled that mandating the inclusion of two fingerprints in identity cards is in line with fundamental rights to privacy and personal data protection, due to its benefits in combating identity fraud and ensuring system interoperability. However, the regulation enacting this requirement was deemed invalid because it was based on incorrect legal grounds and followed the wrong legislative process. Despite this, to prevent negative impacts on EU citizens and maintain security, the Court has decided to keep the regulation’s effects until a new, correctly reasoned regulation is implemented by December 31, 2026, at the latest. The ruling came about after a German citizen contested the requirement for fingerprints in identity cards, leading to a review of the EU regulation’s validity.

EU cybersecurity agency outlines good practices for remote identity proofing

The European Union Agency for Cybersecurity, ENISA, has released a report outlining good practices for remote identity proofing (RIDP), as the EU proceeds with a digital transformation that will rely heavily on biometrics, digital identity and reliable remote identity verification services.

The report, sensibly titled “Remote ID Proofing Good Practices,” points to a general shift toward digitization across European society and the economy, which was accelerated by the sudden need for remote identity verification during pandemic-related restrictions, resulting in “a period of intense transformation.” Lockdowns and other public health security measures, it says, “highlighted the significance of well-regulated and standardized remote identification processes, along with trustworthy digital identities on which public and private sector organizations may rely.”

With eIDAS 2.0 on the horizon, promising to provide all EU citizens with safe, transparent access to the EU Digital Identity Wallet (EUDIW) and other digital services, Europe is well on its way to goals set for 2030, and the European Commission’s goal of setting concrete targets for a “secure, safe, sustainable and people-centric digital transformation.” According to ENISA, this facilitates the need for “secure and reliable identity proofing services, deployable quickly, at scale and in a cost-efficient manner,” which is “a key enabler for electronic transactions in the Single Digital Market.”

In a world beset by deepfakes and injection attacks, all of this comes with a dark cloud of criminal potential. ENISA says recent developments in the attack landscape motivated its report. Fraud techniques can make RIDP methods unreliable, it says, and stakeholders want to know how to mount an effective digital defense based firmly in established good practices. In a concise expression of the core problem across the board, ENISA notes how “the rate of changes in technology and threat landscape outperforms the legislative cadence.”

To that end, the report identifies key objectives: to increase stakeholder awareness, assist in risk analysis practices in a rapidly changing threat landscape, and contribute to the development of stronger RIDP countermeasures.

EU announces deals on digital tech funding, health data

As the AI Act nears enactment, the EU is pouring money into the industry while regulating health data. Meanwhile, the EUDI Wallet is being received with some skepticism from fintech experts.

EU announces deals on digital tech funding, health data | Biometric Update

The European Union is planning to invest €67.5 million in supporting businesses working in the artificial intelligence…

www.biometricupdate.com

EU regulators aim for frictionless age verification, interoperability

Changes in age verification are on the minds of many legislators, regulators and providers. A recently released on-demand webinar presented by Biometric Update and Goode Intelligence explores age verification and estimation in the context of lessons from deployments and regulatory moves in the UK, but the flurry of age verification debate extends to Europe and beyond.

The European Commission-funded euConsent project has released a feasibility study investigating the viability of potential modifications to its architecture that would enable interoperability between age verification providers (AVPs).

Presently, euConsent provides a “distributed interoperable model,” based on eIDAS architecture for secure information exchange between nodes, which allows AVPs to reuse previous age checks performed by other providers as long as both are part of the euConsent network.

The “Feasibility Study for AVP Interoperability between Native Mobile Applications” is concerned with enhancements that would extend this capability to mobile apps. “This goal is not trivial,” reads the report, “since data sharing between different apps has many restrictions, and the two major mobile operating systems (Android — iOS) have different limitations.”

Functionally, the proposed system must be able to recognize when a user has not previously signed in to an age-restricted app that is part of the euConsent network, and therefore requires an age verification prompt. Once a user has signed into an app that is part of the network, other apps that require age verification will recognize that sign-in and apply it. User authentication on a device via PIN, password, or biometric authentication can provide conditional limits to access, or the transfer of permission can be seamless.

Newly downloaded apps using other euConsent AVPs for verification must be able to recognize that an age check has already been performed by the user on another euConsent AVP. For all of this to work, the different AVPs must be able to communicate.

Mastercards from Thales and FPC, Idex Biometrics launch in Turkey

A pair of banks in Turkey are rolling out biometric payment cards from Mastercard.

Fingerprint Cards and Thales will collaborate to launch biometric payment cards in the country, marking the eleventh biometric payment card rollout globally between the two companies. Also in Turkey, DenizBank is launching biometric payment cards powered by Idex Pay.

Smart Engines claims new facial recognition system uses no biometric information

Facial recognition algorithms traditionally depend on biometric data to make matches. However, with the implementation of the Artificial Intelligence Act within the European Union, the legal status of facial recognition systems has changed, especially if they pose a risk to personal freedom and privacy.

Smart Engines has identified this market gap as a need for facial recognition that does not depend on the use of biometrics. The company claims to have addressed this issue by utilizing a neural network, trained to compare facial images in a manner similar to human visual recognition.

According to Smart Engines, the AI-powered process involves three steps: image acquisition, document recognition, and face matching. The process being with acquiring a live image through a standard camera, and extracting the facial photo from an ID document. AI algorithms then compare the live image and the document image “end-to-end” to determine their similarity in a way the company compares to human matching “by eye.”

The Smart Engines highlights that its technology solves the key issue of user privacy (as defined in the AI Act) by avoiding the use of sensitive biometric information. The company believes this approach will enhance user trust, regulatory compliance and security, facilitating the widespread adoption of its technology across the European Union.

Furthermore, the demand for surveillance technologies capable of operating autonomously, without an internet connection, is growing. Smart Engines’ AI-powered solution is designed to function in such autonomous settings, promoting its wider implementation and integration across various industries.

The Smart ID Engine’s face verification software module leverages data synthesis to eliminate the bias in the training of its algorithms. This involves creating artificial data points algorithmically to support the training of machine learning modules, which is used to generate a diverse set of facial images representing a wide range of demographic groups.

The Smart ID Engine is compatible with a wide array of devices and platforms, including those based on x86 and Arm architecture, making the solution more accessible to a wide user base. It supports integration on devices running Windows, Linux, Android and iOS operating systems.

Adopting a privacy-focused approach, Smart Engines ensures that no personal and biometric information is transmitted to third-party services, nor is it saved or stored. In an effort to enhance user trust, the solution complies with leading privacy regulations such as HIPAA, GDPR, and CCPA.

Precise adds palm biometrics as third modality for access control, developers

Precise Biometrics is expanding into a new modality with the launch of palm biometrics software to run on any device with a quality camera.

The new modality is being integrated into Precise’s portfolio of access control and visitor management products. In addition to being offered with YOUNiQ Access and YOUNiQ Visit, Precise will sell its palm biometric algorithm to developers and system integrators, and notes payments as another possible application.

Palm biometrics provide touchless convenience, fast and reliable authentication, the company says, and will allow it to address the needs of more customers and use cases, resulting in higher sales.

Precise can also offer more multi-factor authentication options now, between its palm, face and fingerprint biometrics.

California-based palm recognition scanner-maker Hand.ID has been identified as one of Precise’s initial development partners. Hand.ID’s website says it has products coming this summer.

“Being able to extend our product offering reflects our commitment to keep on pushing forward and offer the most secure and user-friendly authentication and identification methods to our customers,” says Precise CEO Joakim Nydemark. “With palm recognition we will be able to address even more customers and use cases, which is in line with our ambition to accelerate our sales.”

Paravision’s biometric age estimation released to expanding online safety market

Biometric age estimation from Paravision is now available to manage access to age restricted services in a privacy-preserving way, according to a company announcement.

The Age Estimation solution is developed with an approach prioritizing ethics, the company says, to address the needs of a growing range of businesses to apply reliable age assurance to their digital services. In addition to traditional age-restricted industries, legislation like the U.S. Kids Online Safety Act and the UK’s Online Safety Act propose and impose (respectively) new obligations for social media networks, gaming platforms and other online service providers.

The technology was developed in collaboration with Persona, and integrates with Paravision’s recently launched biometric Liveness and Deepfake Detection products to provide trust in age assurance.

“As online activities have become ubiquitous for adults and children alike, we’re proud to release a product that can meaningfully help protect children while enabling safe, fast, and private access for age-appropriate users,” says Paravision CPO Joey Pritikin. “Paravision Age Estimation not only helps to create safe and user-friendly online experiences but also helps partners meet compliance requirements and local laws while fighting multiple types of fraud.”
“Our solution is ethically trained on a proprietary dataset of hundreds of thousands of properly-consented images, ensuring broad race, gender, and diversity. It is designed to keep people safe while preserving their privacy,” says Paravision CTO Charlie Rice. “And it is developed with our partners in mind, enabling rock-solid performance in their production systems.”

Facia 3D biometric liveness detection passes iBeta Level 2 PAD compliance test

3D liveness detection software from Facia has passed a Level 2 biometric presentation attack detection standard compliance test from iBeta.

UK-based startup Facia says in the announcement that the false rejection rate (FRR) of its software is below 1 percent, and the false accept rate is 0 percent, making it a leading identity verification provider.

iBeta tested Facia’s SDK v1.0.3 with a OnePlus Nord 200 smartphone running Android 12 and an iPhone 12 Pro Max running iOS 16.

Facia also recently unveiled a new algorithm for deepfake detection, called Morpheus 2.0. The company’s 3D facial recognition and liveness detection are complemented by age verification and iris detection software.

“We always focus on meeting the industry certifications that propel us towards the top of the industry,” says Facia CEO Mujadad Naeem. “Our rigorous testing ensures that our clients can immediately onboard our solutions, with utmost level of satisfaction.”

Facia’s customer base consists of banking, KYC and crypto providers, along with airports and security agencies the company describes as government-level. The company also lists Intellicheck among its clients.

Patent for ultrasonic fingerprint biometrics among thousands Huawei files each year

A Huawei patent for ultrasonic fingerprint identification technology has been recently published. The patent application for CN117058725A describes “an ultrasonic fingerprint identification module, an ultrasonic fingerprint identification system and electronic equipment” that uses piezoelectric layers and stacked circuit substrates to “improve the accuracy of identifying fingerprint information.”

The move suggests Huawei may be moving to challenge Goodix, which many companies including Samsung currently rely on for ultrasonic fingerprint tech, but is restricted by Qualcomm patents, per a report in Gizmochina. An in-house ultrasonic fingerprint system would make Huawei less dependent on other manufacturers and positioned to serve the Chinese market.

According to the patent application, “compared with the optical fingerprint recognition module, the ultrasonic fingerprint recognition module is more suitable for the technical development trend that the transmittance of the current screen is continuously reduced,” because “the thickness of the ultrasonic fingerprint recognition module is thinner.”

Yet, while Huawei’s ultrasound-based system may be more spatially efficient for ever-more-wafer-thin devices and screens, there are problems.

One is juice: the “penetrating power of the ultrasonic wave” is presently not enough for certain terminal screens that can cause distortion, leading to inaccuracy. Furthermore, noise is an issue: “the echo signal has a large number of signals reflected by other devices, so that the original signal-to-noise ratio (SNR) corresponding to the current fingerprint image is small, and therefore the ultrasonic fingerprint recognition module cannot accurately and effectively recognize fingerprint information.”

This is, indeed, a problem — and a technical problem the applicants at Huawei aim to solve.

New crop of latent fingerprint algorithms set new highs in NIST testing

The U.S. National Institute of Standards and Technology has completed eight evaluations of biometric algorithms for latent fingerprint matching since the start of this year, which set new highs in accuracy and speed.

NIST has tested new submissions from Hisign, Dermalog, ROC, Neurotechnology, Griaule, Idemia, Peking University and Innovatrics since January 24, 2024 for its Evaluation of Latent Fingerprint Technologies (ELFT).

Idemia scored the lowest false positive identification rate (FPIR), at just under 8 percent. Hisign was next at 11 percent. ELFT assessments are carried out with the false non-identification rate (FNIR) set at 1 percent. ROC had the fastest mean mated search duration, at 15 seconds, over a hundred times faster than the next-fastest algorithm.

The top rank-1 hit rate was 96.5 percent, from Idemia and Hisign.

Idemia says in an announcement that its result is the fourth consecutive time it has achieved the best accuracy and the top speed among the most accurate algorithms.

The algorithm submitted by Idemia is already available in its flagship Multi-Biometric Search Services (MBSS), according to an announcement from Idemia Public Security, the group’s biometric solutions division.