AI - On-PREM AMD's Threadripper Halo is a local-AI workstation for researchers with deep pocketsAI workstation promises to put up to 576 GB of HBM3e and 16 TB/s of memory bandwidth on your desk - ai and ml Hugging Face is too important to fall into Nvidia's hands$12.9 billion deal will inevitably cement Nvidia's market dominance and harm competition in the process. Regulators should take note - ai and ml Zuck's Muse to Spark joy with open weights release 'soon'While you wait, Meta says it’s taught the model to stop wasting tokens and ask for help a bit more often - ai and ml With Gemini 3.8 Flash, Google reminds everyone it's still in the raceAI model scores well, runs fast, and doesn't cost too much (yet) - security AI agents carried out every step of this ransomware attack – then left the victim an 80-page security auditAdding insult to injury Infosec - Security Russians are posing as Signal support to launch phishing attacksPLUS: US takes down Iranian propaganda sites; Marketing company asks 'Why Do We Have Your Information?' And more! - Security Microsoft patches failed to fix on-prem SharePoint, which is now under zero-day attackPLUS: China upgrades smartphone surveillance tools; Ring eases anti-snooping stance; and more - Black Hat and DEF CON DEF CON Franklin project enlists hackers to harden critical infrastructureVoting village reports have been so successful, says Jeff Moss, that the whole of DEF CON will now be included - Security EQT buys majority share in Swiss cybersecurity biz AcronisWent at equivalent of $3.5B+ valuation for entire firm, though portion sold not specified - Malware Month Ten years since the first corp ransomware, Mikko Hyppönen sees no end in sightOn the plus side, infosec's a good bet for a long, stable career FOSS - Feel peak Windows
18h ago · via theregister.com
Abstract Deep convolutional neural networks have become widely adopted for food image classification, but their deployment on resource constrained devices remains challenging due to computational overhead, model redundancy, and increased architectural complexity caused by non contributing feature channels. Existing pruning approaches mainly emphasize parameter reduction, with limited focus on identifying and eliminating layer wise redundant feature responses while preserving semantic interpretability. In addition, Explainable Artificial Intelligence (XAI) techniques are largely confined to post hoc visualization and are rarely utilized for model optimization and structured compression. To address these limitations, this work proposes a Grad-CAM (Gradient-weighted Class Activation Mapping) inspired channel importance criterion for structured pruning of our SE-DenseNet121 model for food classification. The proposed framework identifies redundant feature responses, prunes non performing channels, reduces model cost and complexity, and retains only the most informative representations while preserving Dense Block connectivity. Using this importance score, 160 of 896 transition layer channels (17.9%) were removed. The pruned model was further optimized using Knowledge Distillation and cross entropy fine tuning. Experimental results show that the proposed framework reduces the parameter count by 8.39% and MACs by 5.15%, while achieving an accuracy of 98.13% (98.61% with test-time augmentation). Across five independent runs, the model attains a mean accuracy of 98.56% ± 0.07%, which is adopted as the primary reported metric. McNemar’s test shows a statistically significant but modest divergence between the pruned and teacher models, confined to 1.9% of the test images. Applied zero-shot to 3,045 images from three independent public datasets, the pruned model reaches 86.21% accuracy against the teacher’s 87.78%. These results demonstrate that explainability guided pruning improves computational efficiency while maintaining comparable recognition accuracy and interpretability for food image recognition. Funding Open access funding provided by SRM University-AP. No external funding was received for this research. The authors acknowledge institutional support
20h ago · via nature.com
A few years ago, artificial intelligence began making its way into manufacturing through image-based data comparison. Not exclusively, but especially for quality control applications, neural networks have since been trained to detect deviations from target specifications by comparing a component's actual state with reference images. Today, generative AI based on large language models helps users solve typically well-defined problems step by step. In an iterative process, humans provide prompts and instructions until a satisfactory solution is reached. Agentic AI operates very differently. For complex problems, it autonomously selects the AI tools (agents) required and develops solutions... ...with minimal human intervention. Everyday chatbots such as ChatGPT already illustrate the transition toward agentic AI. Users see how their request is broken down into multiple steps, such as searching, analyzing, and summarizing. However, they do not see the logic by which the system selects and sequences the appropriate tools. Agentic AI is already demonstrating its strengths in software development, customer service, IT operations, knowledge management, and administrative business processes. It excels wherever complex tasks consist of multiple steps and require coordination across various systems. Customer Service: In customer service, agentic AI can autonomously process a large portion of customer requests. It can do more than answer questions; it can execute complete service workflows and make independent decisions within predefined rules. Software Development: Modern coding agents analyze requirements, write code, run tests, and iteratively identify and fix defects. Retail: In retail environments, well-orchestrated AI agents can support inventory management, demand forecasting, logistics, and product administration. These agents independently combine information from different systems, prepare operational decisions, and in some cases execute them. Orcha-3: Bringing Agentic AI into Industrial Manufacturing The Orcha-3 platform, developed at Fraunhofer IWU, transfers this approach to manufacturing engineering. Orcha-3 is designed to address questions that cannot easily be solved using
21h ago · via idw-online.de
Police expand use of facial recognition technology - Published Police have used live facial recognition (LFR) for the first time in Devon. The technology, which scans faces in crowds from a live camera feed and checks them against a watchlist of wanted people, was used in Exmouth town centre at the end of August, officers said. Sgt Angela Galasso, from Devon and Cornwall Police, said the technology was probably "scanning far more faces than a police officer ever could", adding: "It's a tool. One of the many in our armour which help us to do our job." Any people identified by LFR still needed to be verified by humans, police said. Human rights group Liberty has called for legislation to clarify how police forces can use such systems. Last month the police used the same technology near the Boardmasters festival site in Cornwall. Galasso said a camera van had been deployed in response to anti-social behaviour, violent incidents and shoplifting in the town. She said: "We've got technology on board which is running with a predetermined watchlist of people. There's about 8,000 on the list. "The alert comes up, you're scanning the picture held on our system and the live picture. You're looking at that similarity scoring. "If it falls within a certain point, we then ask our uniformed officers to engage with that person and a second verification happens." The BBC spoke to shoppers on The Strand in Exmouth, where the cameras were used. "If you've done nothing wrong, then you shouldn't be afraid of it," said Denise Newman, from Budleigh Salterton. Paul Miller, from Exmouth, said he was "in favour of facial recognition with strict safeguards". Police said images of people who were not of interest would be "automatically and permanently deleted within seconds" and the force would
22h ago · via bbc.co.uk
Dreame Technology unveiled a broad product portfolio at IFA 2026 in Berlin, introducing four new product lines: the Aqua20 Pro Ultra Roller X Complete vacuum and mop robot, the A4 AWD Pro robotic lawn mower series, the LEAPTIC Cube mini camera, and the new T-series of wet/dry vacuums. The manufacturer is thus increasingly positioning itself as a provider of a connected “smart living” ecosystem that goes beyond traditional household robotics and ranges from floor cleaning and yard care to mobile imaging. Aqua20 Pro Ultra Roller X Complete: Robot Vacuum with a 180°C Steam Source The Aqua20 Pro Ultra Roller X Complete takes center stage at the presentation. According to Dreame, it is the first manufacturer to incorporate a central steam source that generates steam up to 180 °C within seconds and dispenses it through four nozzles. In a four-step process, the steam first loosens dried-on dirt before the mopping roller wipes it away with warm water using the AquaRoll system. According to the manufacturer, the process removes 99.999 percent of bacteria, as tested by the independent laboratory SGS. The mop head can be extended up to 8 cm using MopExtend technology—twice as far as on its predecessor, the Aqua10 Ultra Roller Complete—and is designed to better reach edges, corners, and areas under furniture. Other features include AI image recognition capable of identifying over 320 object types, 40,000 Pa of suction power, and the ability to climb obstacles up to 10 cm high. The Aqua20 Pro Ultra Roller X Complete will be released in Germany on September 21 for 1,499 euros, with a 300-euro introductory discount during the first two weeks. Other manufacturers are also showcasing new generations of vacuuming and mopping robots at IFA 2026, further intensifying competition in the high-end segment. A4 AWD Pro Series: Robotic Mowers for Large
23h ago · via basic-tutorials.com
More than a quarter of England and Wales police forces now involved in Palantir pilots Finding comes as concerns rise about the dependence on overseas suppliers for critical public-sector services Twelve English police forces are piloting Palantir’s data analytics software, according to research by the FT. The newspaper found that forces in Bedfordshire, Hertfordshire, Leicestershire, Cambridgeshire, Essex, Kent, Norfolk, Suffolk, Derbyshire, Lincolnshire, Northamptonshire and Nottinghamshire have all engaged in pilot projects with Palantir, in addition to the Metropolitan Police, whose attempt to procure software from the company is being challenged in court. In some cases, the trials are part of regional Special Operations units, through which neighbouring forces share services and data to counter organised crime and terrorism. In others, forces have undertaken pilots with Palantir on an individual basis. Most of the trials are ongoing, while Lancashire police said it had discontinued the pilot. There are 43 territorial police forces in England and Wales, meaning that more than one quarter are currently participating in Palantir pilots. Palantir has not disclosed previously how many police forces it has been working with. The company – founded by Peter Thiel and counting among its customers the Israeli military and US Immigration and Customs Enforcement (ICE) – is known for its strategy of offering its services to public service bodies for a very low initial cost in order to gain a foothold. For example, it won an initial NHS contract in 2020, for or a nominal £1. Three years later the company was offered a £330 million contract to build the NHS Federated Data Platform (FDP). A group of MPs is urging the government to exercise a break clause in the contract over issues of transparency, vendor lock-in, value for money and data security. In May 2026, The London Mayor’s Office for Policing
1d ago · via computing.co.uk
IAS Residential Fellow Dr Alice Ojwang delivers a seminar on their research, fully titled "Carbo AI: An Artificial Intelligence-Based Photographic Application to Support Dietary and Lifestyle Education for the Management of Type 2 Diabetes" - Type 2 diabetes is a rapidly growing global health challenge, with dietary management central to its prevention and control. However, estimating the carbohydrate content of meals remains difficult, particularly for locally consumed foods. Carbo AI is an artificial intelligence-powered mobile application that uses photographic image recognition, computer vision, and machine learning to identify foods, estimate portion sizes, calculate carbohydrate content, and provide instant dietary guidance. The project was led by the Human Nutrition and Dietetics Team in the Department of Health and Biomedical Sciences, Technical University of Kenya, in collaboration with experts in computer science, medical physics, and software engineering. Development was supported by Amazon Web Services (AWS) cloud infrastructure and KENET's High-Performance Computing (HPC) platform. Carbo AI aims to improve carbohydrate literacy, strengthen diabetes self-management, enhance nutrition counselling, and demonstrate the potential of interdisciplinary AI-driven innovations to improve health outcomes. Arrivals from 11:45 am for a 12:00 noon start. For those joining in-person, lunch will be served after the seminar from 1:00pm. This event is hybrid format, please use the required booking button at the bottom of the page to choose either in-person or online attendance. (Please note that in-person spaces are limited and booking is required, so we can manage numbers for catering and also the space in the seminar room) By booking a place at this event, attendees agree to behave in a respectful manner such that everyone feels comfortable contributing as they wish. The IAS reserves the right to eject anyone who does not abide by this policy. IAS events are typically recorded, minus any Q&A sessions at the end, again
1d ago · via lboro.ac.uk
The German chemical company’s optical sensing unit has launched a legal fight with the iPhone maker, in a dispute over technology used to prevent facial recognition systems from being fooled by masks, photographs and 3D replicas. If you don't have a login or your access has expired, you will need to purchase a subscription to gain access to this article, including all our online content. For more information on individual annual subscriptions for full paid access and corporate subscription options please contact us. To request a FREE 2-week trial subscription, please signup. NOTE - this can take up to 48hrs to be approved. For multi-user price options, or to check if your company has an existing subscription that we can add you to for FREE, please email Atif Choudhury at achoudhury@worldipreview.com 24 August 2026 The iPhone maker has hit back at an attempt to derail its lawsuit, as new filings shed more light on the allegations at the heart of the Silicon Valley dispute, ahead of an October hearing. 7 September 2026 As Trump publicly backs generative AI developers, The Seattle Times and Newsday are the latest to claim that OpenAI and Microsoft scraped their websites to use in ChatGPT, Copilot and Bing. The expanding litigation could shape how publishers protect valuable content—and how AI companies train and commercialise their models.
1d ago · via worldipreview.com
Quizzes·TriviaFrom JFK To Ghandi, Only People With SUPER Facial Recognition Can Match These Historical Figures To Their Childhood PhotosImagine this small girl with a toy airplane...Posted 14 hours agocomment Press to add the post to your bookmarks Saveby Jeremy HayesBuzzFeed Contributor There are some faces in history that are timeless. You know these historical figures with a simple glance. But can you identify them from a childhood photo? Test your facial recognition skills: Share This QuizComments
Comments
1d ago · via buzzfeed.com
Although the Milwaukee Police Department halted its use of facial recognition technology in February, there are still cases being prosecuted in Milwaukee County in which the technology was used at some point in the investigation. This means concerns about the technology – and the potential errors it can introduce – still remain, community members told NNS. Facial recognition technology uses artificial intelligence to identify someone by comparing a photo of an unknown face to a database of images of known faces. “The reliability of these systems is certainly in question, and particularly the disproportionate risk of inaccuracy when identifying non-white faces,” said Amanda Merkwae, policy and advocacy director for the American Civil Liberties Union of Wisconsin. Questions also remain about whether defendants and their lawyers have always been told when facial recognition technology played a role in an investigation. MPD and facial recognition In April 2025, the Milwaukee Police Department acknowledged it had, for two to three years, used other agencies’ licenses to obtain facial recognition search results without having a written department policy governing the practice. At the same time, MPD was considering expanding its access to facial recognition technology through an agreement with the company Biometrica. The proposal provoked substantial opposition from residents and civil rights groups who raised concerns about misidentification, racial bias, surveillance and public trust. Milwaukee Police Chief Jeffrey Norman announced in February that MPD would prohibit its members from using facial recognition technology and would not consider acquiring the technology until after the department develops a policy with the Fire and Police Commission. However, the moratorium did not eliminate the technology from criminal cases in which it had already been used. MPD did not respond to questions about its policies and practices relating to disclosing the use of facial recognition to the DA’s office.
1d ago · via milwaukeenns.org
Meta sued over alleged facial recognition training for smart glasses Meta Platforms is facing a proposed nationwide class action lawsuit accusing the company of using photographs from Facebook and Instagram to extract biometric information for an unreleased facial recognition system intended for its smart glasses. The suit further alleges that Meta has used images to train generative AI models without obtaining consent. The 66-page complaint, filed in the U.S. District Court for the Northern District of Illinois, alleges Meta used images depicting both users and non-users to train and test facial recognition technology known internally as NameTag; to generate facial templates for the system; and to develop image-generation models including Emu and Muse Image. The lawsuit, Alvarez et al. v. Meta Platforms, Inc., was brought by Illinois resident Francisco Alvarez, his minor child, California resident Jeremy Wahl, and Wahl’s minor daughter. It alleges violations of the Illinois Biometric Information Privacy Act (BIPA), California publicity and misappropriation law, and the California Constitution’s privacy protections. At the center of the smart glasses allegations is NameTag, facial recognition technology developed for Meta’s Ray-Ban and Oakley glasses, as well as the Meta AI companion app. The unreleased system’s existence became public in June, when WIRED reported finding inactive NameTag code in the Meta AI app. The code showed that the system was designed to recognize people seen through the glasses’ camera. When activated, it would convert a captured face into a biometric signature, or faceprint, and compare it with faceprints stored on the wearer’s phone. According to the complaint, Meta went substantially beyond developing software capable of recognizing faces. The plaintiffs allege that the company obtained photographs from Facebook and Instagram, extracted numerical representations of the faces in those photos, and created biometric identifiers that could be supplied to NameTag to identify people encountered
1d ago · via biometricupdate.com
5 Cool New Gadgets From IFA 2026 You Need To Try IFA 2026 is the largest consumer and home technology event, and it's full of interesting new products. From beauty devices to everyday tech and even future innovations, the trade show also marks the launch of several new devices for Europe and America. With some products immediately available and others arriving in the next few months, BGR saw several cool gadgets that might become must-have products for consumers. With over 2,000 exhibitors at IFA 2026, BGR went across the 25 different halls of Berlin's annual trade show to uncover some of the most interesting gadgets. Many of these products have been released by companies based in Asia, as the region continues to export innovative ideas and products across the globe. From Lovesense's AI doll to upcoming cases for the foldable iPhone, we've seen a bit of everything being highlighted in Berlin. While ShowStoppers already brought a cool first look of what to expect from this year's IFA, there are even more technologies that could soon make part of people's everyday lives, whether they just want to easily store data, look stylish, be hydrated, or clean their homes efficiently. These are the coolest new gadgets we've found at IFA 2026; our top picks include Lexar's Muse portable SSD, L'Atitude's latest smart glasses, and even a smart water bottle with facial recognition. 1. Lexar's Muse portable SSD Popular storage brand Lexar continues to navigate the memory crisis that's been plaguing the industry — and raising product prices — with more innovative solutions. At IFA 2026, the Chinese company introduced Muse, the thinnest portable SSD in the world that also comes with a very fashionable design. This release is part of the company's 30th anniversary celebration. The Lexar Muse measures a maximum of
1d ago · via bgr.com
Sie können Operatoren mit Ihrer Suchanfrage kombinieren, um diese noch präziser einzugrenzen. Klicken Sie auf den Suchoperator, um eine Erklärung seiner Funktionsweise anzuzeigen. Findet Dokumente, in denen beide Begriffe in beliebiger Reihenfolge innerhalb von maximal n Worten zueinander stehen. Empfehlung: Wählen Sie zwischen 15 und 30 als maximale Wortanzahl (z.B. NEAR(hybrid, antrieb, 20)). Findet Dokumente, in denen der Begriff in Wortvarianten vorkommt, wobei diese VOR, HINTER oder VOR und HINTER dem Suchbegriff anschließen können (z.B., leichtbau*, *leichtbau, *leichtbau*). For radiology residents, the steep learning curve of cross-sectional anatomy (especially in CT and MRI) demands more efficient and engaging educational tools. In recent years, game-based learning (GBL) and gamification have emerged as powerful strategies to enhance postgraduate education by leveraging ‘active recall’ and ‘spaced repetition’ within a competitive framework. Here we describe the context of ANATOMIADI and assess the perceived educational impact. Materials and methods Image datasets were anonymized, archived, and manually labeled to highlight specific anatomical structures. Over 400 unique games were developed, categorized by anatomical system and tiered into three levels of difficulty. A structured feedback survey was designed and distributed to the participants after the event to capture data on different domains regarding the educational value, gamification and engagement, platform usability and team collaboration, technical workflow, logistics, scientific accuracy, fairness, and quality assessment. Results ANATOMIADI challenge registered an attendance of 141 participants representing 28 residency programs across Italy. The evaluation of the event’s perceived value yielded a unanimous positive response; indeed, 100% of the participants recommended that the competition be repeated in future academic years. Based on the required level of knowledge, the event was recommended to residents in their 3rd (54/63 responses) and 4th year (48/63 responses) of postgraduate education. Conclusion By integrating high-fidelity imaging with interactive game mechanics, the ANATOMIADI challenge not only fostered anatomical learning
1d ago · via springermedizin.de
The concourses are silent when, deep in the Emirates Stadium’s bowels, Chief Inspector Pete Dearden’s briefing begins. Arsenal’s 4.30pm kick-off against Chelsea is still hours away but the fixture carries significance as a London derby and because the Premier League champions are hosting a realistic challenger. Dearden is a senior officer in the Metropolitan police’s public order command and Sunday’s matchday commander. He is, to borrow football jargon, giving the pre-match team talk, running through intelligence and in-game tactics. “Set the right style and tone,” he says. After Dearden has finished, the police room becomes a pseudo-station for the afternoon, with all incidents triaged by the same crime officer. Football offences can be found in niche legislation, and the Met wants London’s matchdays to be, so far as possible, policed uniformly. Matchday is the iceberg’s tip. The Met’s 17 dedicated football officers (DFOs) each work full-time on one club, planning each fixture with the matchday commander. All 500-plus games the Met covers per season are risk-graded using factors such as risk level of supporters, proximity, recent history and kick‑off time. Dearden labels Sunday’s game “medium medium”. “Our mantra is safety, security and service,” Dearden says. “We appreciate that with 60,000 people in the stadium not everybody is going to behave, but we want to reduce the overall risk of the likelihood of disorder.” Three police support units (PSUs), each containing an inspector, three sergeants and about 21 constables, have been assigned to the match. In addition, there are the senior team and 20 operational football officers, or “spotters”, on duty. Clubs also have their own in-house safety teams, with, in effect, both parties doing their homework before collaborating. Arsenal’s operation involves about 1,000 stewards and response team members. They are the first responders to any incidents within Arsenal’s footprint. When
1d ago · via theguardian.com
Fibre optic sensing represents a significant advancement in the utilisation of fibre-optic infrastructure. The same fibre technology that has revolutionised telecommunications can now be leveraged to continuously monitor and detect physical changes and events in the environment surrounding critical infrastructure. What is fibre optic sensing? The concept of fibre optic sensing is to treat fibre itself as the sensor by creating thousands of continuous sensing points along the fibre length. This phenomenon is called distributed fibre optic sensing, where the optical fibre itself acts as a distributed fibre optic sensor. Principle of operation Fibre sensing work is based on the phenomenon of backscattering that occurs due to interaction between light and the glass fibre. Different sensing technologies analyse different components of the backscattered optical signal. An instrument generally called fibre sensing interrogator is used to send optical pulses into a fibre and analyses the light returning from different points along the fibre. Changes in the characteristics of this returned light can reveal what is happening at specific locations. The main purpose is to use a standard or specific fibre for measuring the strain, temperature or vibrations along its length, using Raman, Brillouin or Coherent Rayleigh backscattering fibre sensing techniques. - DTS (Distributed temperature sensing): It incorporates a Raman OTDR. Raman scattering produces temperature-induced changes in photons scattered back to the source in the Stokes band. By measuring the difference between the intensity of backscattered light in the Stokes and anti-Stokes bands, the temperature can be accurately determined at any given location along the fibre length. - DTSS (Distributed temperature and strain sensing): It incorporates Brillouin OTDR. Brillouin scattering is a phenomenon where the backscattered light wavelength is influenced by the external temperature and acoustic stimulation in a predictable way. This data, when coupled with background knowledge of temperature at the
1d ago · via tele.net.in
Summary - Canon uses deep learning for subject recognition and tracking - Its technology covers noise reduction, demosaicing and optical correction - The goal is a more faithful capture rather than invented content - This is not a blanket rejection of every form of generative AI - The distinction concerns AI’s role in the photographic process Canon is positioning artificial intelligence primarily as a tool for recording the real world more effectively: recognising subjects, reducing noise, reconstructing colour and correcting optical imperfections. In comments from Go Tokura and Canon’s official technical material, deep learning works before or during image processing. The distinction matters: the system attempts to recover capture information more accurately rather than adding people, objects or scenes that were never in front of the lens. Three areas of application Canon groups its technology into recognition and detection, image optimisation and optical correction. In autofocus, deep-learning models help a camera identify people, animals and vehicles and maintain tracking. Noise reduction and colour reconstruction For noise reduction, the challenge is separating unwanted information from real detail, particularly at high ISO settings. In demosaicing, which reconstructs full colour information from sensor data, deep learning can reduce false colour and jagged edges. Lens and diffraction correction A third area addresses lens aberrations and detail lost to diffraction. The camera or software combines knowledge of the optical system with trained models to restore information altered during image formation. Not a blanket rejection of generative AI The position should not be read as a promise that Canon will never use generative AI. It describes the company’s current imaging priorities: tools that serve the photographer and improve fidelity or capture efficiency. The line between correction and invention remains central to photographic trust. What we think AI is already an invisible part of modern cameras. The
1d ago · via pttl.gr
Police are celebrating a significant milestone in the use of technology to aid investigations and arrests. Essex Police have reached 200 arrests using live facial recognition (LFR) technology, which helps officers identify wanted individuals and progress ongoing investigations more efficiently. Detective Superintendent Stephen Jennings said: "Reaching 200 arrests represents a significant milestone in our use of live facial recognition. "That’s 200 suspects brought into custody, 200 investigations progressed and, potentially, 200 victims a step closer to getting justice – all thanks to this technology." In August alone, 26 arrests were made through the use of LFR. Arrests have included individuals wanted for violent crimes and theft offences. Det Supt Jennings also praised the public’s support for the force’s use of this technology, inviting residents to engage with officers during deployments to learn more about how it works. Before each deployment, a watchlist is compiled containing images and data of individuals police want to locate. This includes people wanted for offences, those subject to court orders, those who pose a risk to the public, or are vulnerable or at risk of harm. If a person is not on the watchlist, their image is deleted instantly and is not stored or saved on any database. Read more Det Supt Jennings added: "No-one has ever been arrested as a result of a false positive alert during any of our deployments. "This technology is so clever that even if someone doesn’t have their whole face visible, they can still be correctly and accurately identified. "If you’re not on our watchlist then your image is deleted in a fraction of a second." He also highlighted the operational value of LFR, particularly in freeing up officers’ time. He said: "Police officers spend a significant amount of time trying to locate individuals who are either wanted for
1d ago · via halsteadgazette.co.uk
The technology is being introduced locally by Flownamix Group. (Image source: 123RF) Continuous Facial Recognition with Liveness is now available in South Africa, giving organisations a way to verify that the right person remains active throughout a digital session, not only at login. The technology is being introduced locally by Flownamix Group as part of a launch that also includes Secure Enterprise Messaging. Both solutions are powered by YEO Messaging’s patented technology. The introduction comes at a time when organisations are re-evaluating how they establish trust in digital environments. The rise of identity fraud, AI-enabled impersonation and increasingly sophisticated cyber crime is driving demand for technology that provides stronger identity assurance beyond the point of login. According to the South African Banking Risk Information Centre (SABRIC), digital banking fraud incidents increased by 86% in 2024, while associated losses rose by 74% to nearly R1.9 billion. As organisations continue to expand digital services and customer interactions, maintaining confidence in a user’s identity throughout an interaction is now just as important as verifying it at login. Together, Continuous Facial Recognition with Liveness and Secure Enterprise Messaging enable organisations to strengthen identity assurance across the digital journey. While Continuous Facial Recognition with Liveness continuously verifies that the authorised user remains present throughout a session, Secure Enterprise Messaging protects communications through verified identities and end-to-end encryption. “Our partnership with YEO Messaging, combined with Continuous Facial Recognition (CFR) technology, represents an important step towards secure, identity-assured communication where the person receiving, viewing or acting on sensitive information can be continuously and confidently verified. For Flownamix Group, this partnership is not simply about adding another layer of authentication. We believe this approach will become increasingly important for regulated industries, enterprises and service providers that need to protect confidential information, reduce impersonation risk and strengthen accountability across
1d ago · via itweb.co.za
From AI-Assisted Inspection to AI-Native Metrology Artificial intelligence is rapidly changing the role of metrology within manufacturing. What began as the application of AI to individual inspection tasks is evolving into something considerably more significant: measurement systems designed around artificial intelligence, data and increasingly autonomous decision-making. This transition from ‘AI-assisted inspection to AI-native metrology’ could become one of the most important developments in industrial measurement. Rather than simply using AI to improve an existing inspection process, the longer-term opportunity is to design measurement, interpretation and decision-making as part of a single intelligent workflow. Examples of this transition are already emerging across industrial inspection. ZEISS, for example, has integrated AI-based defect detection into its INSPECT X-Ray software through its ZADD Segmentation application, while KEYENCE is combining conventional rule-based vision inspection with AI-based inspection within its VS platform. Nikon has also introduced automated microscopy incorporating AI-powered image analysis. These developments demonstrate that AI is already moving beyond experimentation and into practical quality-control applications. From Automation to Intelligence Traditional automated inspection typically follows a predictable sequence. A component is presented to the measurement system, a predefined inspection routine is executed, results are compared with specifications, and a report is generated. AI-assisted inspection improves this process by introducing capabilities such as automated feature recognition, image classification, defect detection and adaptive inspection. These applications can reduce programming requirements and help systems deal with greater variation in components and manufacturing conditions. KEYENCE, for example, describes its AI vision technology as an extension of conventional machine vision, allowing AI models to learn acceptable part appearance and reduce the need for continual manual adjustment of inspection logic. Its VS platform allows traditional rule-based inspection and AI inspection to operate together. However, the underlying architecture remains essentially the same: the inspection system performs a predefined task and AI assists
1d ago · via metrology.news
Unico opens IDCloud biometrics to smaller Brazilian businesses through Didit Brazilian startups and smaller companies are gaining self-service access to a facial biometrics network that was previously offered only through customized enterprise engagements. The Unico and Didit partnership makes IDCloud available as a database validation service within Didit’s identity and fraud platform. Businesses can add the service to a configurable workflow through an API or no-code interface. Unico says it has concentrated on consultative sales to large organizations, including major Brazilian banks and retailers. Didit’s self-service model gives Unico a distribution channel to reach companies that may not have huge transaction volume. Brazilian users can complete a biometric identity check without photographing an identity document. This is possible only if Unico is able to retrieve a facial record associated with the person’s CPF. The 11-digit taxpayer identifier gives the reference needed to search the network. A newly captured selfie supplies the facial biometrics used for comparison. The process begins when the user enters a CPF into a workflow configured through Didit. Didit then captures a selfie and applies passive or active liveness detection to on-screen instructions. The platform sends the CPF and captured image to Unico IDCloud, which compares the face with the biometric record connected to that number. IDCloud can return a match, a biometric mismatch, or an inconclusive result. Didit applies the configured policy to determine if the session should be approved, rejected, or referred for review. IDCloud covers 96 percent of the economically active population of Brazil. Didit’s documentation describes the figure as 96 percent of CPF holders and approximately 96 percent of Brazilian adults. In terms of pricing, a conclusive database query costs $0.20. Passive liveness adds $0.10, bringing the combined cost to $0.30, while an active method adds $0.15 for a total of $0.35. Didit
1d ago · via biometricupdate.com