Biometric identification technologies

Face (facial recognition, selfie)

Main article: Facial Recognition and Deepfake Systems (DeepFake)

Facial recognition is the automatic localization of a human face in an image or video and, if necessary, identification of a person's identity based on available databases. Interest in these systems is very high due to the wide range of tasks they solve.

Iris identification

Iris biometrics is a fast, contactless, secure and exceptionally accurate identification method that solves a wide range of tasks for customers in various industries.

• Main article: Iris identification

Sweat

The pattern on the fingerpads can be reproduced, and the face can be forged, which cannot be said about sweat. Sweat characteristics are so unique to each person that it can be used for biometric authentication. This conclusion was reached in 2017 by a group of scientists consisting of Juliana Agudelo, Vladimir Privman and Jan Halamek. The results of their study were published in the report "The possibilities and difficulties of long-term tracking using amino acids secreted by skin glands for active multifactor biometric authentication for cybersecurity."^[1][2]

The idea proposed by scientists is very simple: the composition of amino acids in sweat is unique for each person, and if you teach your smartphone to determine this composition and distinguish it from others, the user can be authenticated by sweat. This method of confirming the identity of a user can find very wide application, ranging from unlocking smart devices to protecting data inside applications. In addition, even people with disabilities who are unable to remember the password or are deprived of the ability to control their limbs will be able to use it.

Unlike other methods of biometric authentication, it is not so easy to fake the unique chemical composition of sweat. According to scientists, the idea they proposed will be able to be used in practice for the next 5-10 years.

Smell

In 2009, wanting to improve "the ability to identify individuals planning to harm the nation," the U.S. Department of Homeland Security tested whether body odor could be used as a method of uniquely identifying a person. A change in smell can be evidence of substitution.

Researchers from the Polytechnic University of Madrid, in collaboration with IIia Sistemas SL, presented in early 2014 a method that claims to be in the line of biometric identification technologies along with face, fingerprint and iris recognition^[3].

The system, developed by Madrid scientists, is able to identify people by the smell emanating from the body. The researchers claim that each person's body has permanent discernible "odor patterns" that are not influenced by disease, diet or age.

The researchers created a sensor capable of recognizing "unique patterns" of human body odors and identifying their carrier with 85% accuracy. The sensor was tested on 13 volunteers, of which eight were men and five were women.

Scientists took thirty odor samples from the cleanly washed palms of each of the subjects at different times of the day. According to the developers, the sensitivity of the sensor turned out to be so high that it was difficult to deceive it with soap, deodorant, cologne or other attempts to change the smell.

In an official statement from the university, scientists express confidence that this opens up the possibility of creating "less aggressive" ways of identifying a person than those that currently exist.

Despite the fact that recognition of the iris and fingerprint give high accuracy of identification, in the mass consciousness these technologies are closely associated with forensic science, which causes distrust and protest, scientists say. Facial recognition at the current stage of development gives too much error rate.

Thus, the development of odor sensors that allow you to identify a person passing by them opens up opportunities for the development of more comfortable and invisible identification methods with a fairly high level of accuracy.

The researchers are confident that such technologies can be used at airports, at checkpoints at the border and in any other situations where photo identification is currently used.

Odor identification is one of the oldest methods used to find and identify people, but now in forensic science specially trained dogs are used for this. The development of methods for effective recognition of human odor using electronic devices has started relatively recently.

So, in April 2013, a group of Swiss scientists presented a method for identifying a person by the smell of mouth. Using a laboratory mass spectrometer, scientists took exhaled air samples from 11 subjects for nine days. Scientists have been able to prove that breath smell also has a unique molecular pattern that does not change depending on external factors, such as eating odorous foods or smoking.

Finger microvibration

Engineers at Rutgers University in New Brunswick proposed in the fall of 2017 a method for authorizing people - by finger microvibrations. The researchers proceed from the fact that they will be unique for each user, and accordingly, in this way an individual signature can be obtained, which will be at least very difficult to fake, writes CNews^[4].

The system, called VibWrite, works quite simply: an inexpensive vibration motor and sensor are attached to a hard surface - be it wood, metal, plastic or glass; when a person touches the surface with his finger, interference is introduced into the vibrations of the engine, which are read as unique signatures.

At the same time, they will be unique for each finger, and their short duration provides increased reliability of authorization, especially compared to code input, graphics keys, as well as, as the technology developers assure, with traditional biometric means.

Ultra-cheap finger microvibration authorization technology developed

Another advantage over fingerprint or retinal scanners is that the VibWrite is about ten cheaper to manufacture and operate.

Accuracy at 95%

The tests showed 95 percent accuracy. True, VibWrite often forced users to enter codes several times before authorization worked in principle. Researchers believe it will take another two years before the system reaches commercial readiness.

VibWrite testing was done indoors. How it will function outside in difficult weather conditions is not yet clear. Tests "in the field" are still ahead.

Heart

2019: By heartbeat

The Jetson device, whose developer is kept secret, is able to identify a person by his heart rate using an infrared laser at a distance of up to 200 meters with an accuracy of 95%. This indicator is more reliable than facial recognition systems and fingerprints that can be forged.

Among the disadvantages of the device - a person should be easily dressed and motionless, since it takes about 30 seconds to analyze. In addition, Jetson identifies people whose heartbeat rhythms are already in the database.

2017: Heart Geometry Identification

In September 2017, it became known about a new method of biometric identification - by heart. Its important advantage is continuity: while the owner authorized in the system is near the computer, the device remains functional, and when a person is removed, it is blocked.

Such a development was invented at the University of Buffalo (USA). Its essence lies in the use of the so-called low-level Doppler radar, which allows you to determine the shape, size and rhythm of a person's heart every 8 seconds. If during this time the heart of the user whose information has been uploaded to the system is not detected, the computer will be locked.

A system developed at the University of Buffalo continuously scans a person's heart using Doppler radar and, if a person is removed from a computer, blocks the system

The authors of the project claim that the scanner is harmless to the human body, since its radiation is only 5 mW, which corresponds to less than 1% of the radiation from modern smartphones.

The new system has low power consumption, which will allow it to be effectively used not only in stationary computers, but also in mobile devices.

In addition, the working distance for the described method of identifying a person can reach 30 m, which makes it a potentially desirable and convenient way for unmistakable identity verification in crowded places - for example, at airports to speed up passport control.

Two people with the same hearts have not yet been found in the world, "said Wenyao Xu, lead author of the scientific work, Ph.D. and assistant professor of computer science and engineering at the University of Buffalo School of Engineering and Applied Sciences.

He also stressed that the shape of the heart in an adult never changes, unless it is affected by some rare serious heart disease.^[5]

2014

NASA development

In 2014, specialists from NASA's Jet Propulsion Laboratory announced plans to create technology that can recognize individual heartbeat features using a phone. Microwave signals emitted by the phone are reflected from the body, recorded by telephone sensors and amplified to reproduce the heart rate. In addition to authentication, the user also receives warnings about changes in his heartbeat with a recommendation to see a doctor.

Bionym Development

In the fall of 2014, Canadian startup Bionym attracted an investment round A of $14 million. It was led by Ignition Partners and Relay Venture funds. Export Development Canada, MasterCard and Salesforce Ventures also participated in the round. Bionym was founded in 2011. It received its first round of investments in August 2013 ($1.4 million). Around the same time, the company opened a pre-order for the Nymi bracelet. Nymi measures the electrical activity generated by the heart muscle (electrocardiogram) and uses this data for authentication. The creators of the bracelet claim that the electrical activity of the heart is unique to each person and therefore can serve as a password. These unique data do not depend on the heart rate, the company emphasizes.

Nymi offers one of the biometric authentication methods by analogy with fingerprint or iris scanning. The bracelet does not require early activation, since, being on the wrist, it continuously monitors cardiac activity. The device generates an encoded wireless signal and sends it over Bluetooth to the device the user wishes to access.

The bracelet is proposed to be used not only for entering a personal computer, but also for unlocking a house, apartment or car. The developers have endowed the system with three-factor authentication. You must confirm your identity only once a day or after the bracelet is removed.

In addition to access using an electrocardiogram, Nymi supports the function of unlocking locks using spatial gestures - thanks to the built-in accelerometer and gyroscope. That is, the user is invited, for example, to draw a certain figure in the air, which only he knows, after which the bracelet will send a signal to the computer or lock.

Fingerprints

If you look at the pads of your own fingers - if suddenly you have never looked before, do it right now - you will see lines that form an intricate pattern. These lines are called papillary lines. Why they are needed, there is no unequivocal answer (some scientists say that lines of two orders of magnitude increase sensitivity to roughness, others that they allow them to grip smooth objects (for example, handrails) more tightly, but it is important for us that each person has this pattern is unique and does not change with age. It was this fact that allowed the Englishman William Herschel in 1877 to come up with fingerprinting - a way to identify a person by papillary pattern - popular in forensic^[6].

Strictly speaking, fingerprints are greasy marks that leave our hands anywhere, for example, on a smartphone screen. These marks carry information about the papillary patterns of our finger pads. True, not all - a specific imprint depends on the angle at which we poked our finger at the screen, with what force, and so on. But more than a hundred years of forensic experience, as well as all the detectives we have read, say: if there is a reference imprint, then belonging can be determined.

Fingerprint identification technologies have incorporated all the best that is inherent in biometrics in general. A fingerprint identifies a particular person, not a token or card; unlike a password, a fingerprint cannot be "peeked," forgotten, freely or involuntarily transferred to another. By the way, modern scanners have learned to establish the belonging of a fingerprint to a living person, and they cannot be deceived by showing an impression of the fingerprint on paper, gelatin or glass. The probability of mistaken identity is 0, 000000001%, and the time required to scan the fingerprint does not exceed a split second.

A huge step towards abandoning passwords was taken thanks to the introduction of a fingerprint scanner into a smartphone. Despite the fact that the technology was used earlier, it Apple^[7] that managed to popularize and widely implement it^[8]

No matter how opponents of collecting biometric data scold the Touch ID function, the technology is actively used not only to simply unlock the smartphone, but also to make purchases in the App Store or use the Apple Pay service.

In the future, some large banks are also considering switching to a fingerprint scanner as a method of authentication. Moreover, Visa goes even further - the company is working on a prototype iris scanner that will be used for the same purposes in the online payment service Visa Checkout.

RichRelevance conducted a survey of 2,000 consumers on both sides of the Atlantic in early 2016. They answered questions about how the technology could affect their in-store customer service model, explaining the difference between what shoppers think is "cool" and what is "creepy."

Despite being open to perceptions of the new, UK shoppers feel worse about more invasive technologies, such as facial recognition software, which could be used to identify them by an employee while visiting the store.

By evaluating fingerprint recognition technology in the payment process, nearly half (47.5%) of respondents would welcome the technology if it also allowed them to automatically receive a home delivery service.

In addition, 62% of customers surveyed want to be able to scan a product from their devices to see feedback and guidance on other items they might like, while 52% of shoppers don't mind the pop-up feature starting to arrive on their mobile devices when they enter the store.

2024: Biometric systems have learned to hack on the sounds of finger movement on the touch screen

In mid-February 2024, a team of researchers from the United States and China announced the development of technology to recreate fingerprints from the sounds of their movements on the touch screen of mobile devices. The method, called PrintListener, theoretically allows you to bypass biometric security systems.

Fingerprint-based security is widespread and highly trusted. According to estimates, by 2032 the volume of the corresponding segment can reach almost $100 billion. Meanwhile, attackers are experimenting with new ways to steal or reconstruct fingerprints. Against this background, some users even try not to show their hands in photos.

𝓲

Unsplash

To implement a PrintListener attack, a cybercriminal needs to obtain audio recordings of sounds that occur when a user interacts with a touch screen through swipes in applications like Discord, Skype, WeChat, FaceTime, etc. For this, for example, any malware that has access to the microphone of a smartphone or tablet can be used. Then special algorithms come into play to analyze sounds and extract the necessary information from them. In particular, the spectral analysis method is involved. In addition, the physiological and behavioral features of the user are taken into account.

The study was attended by employees of Huazhong University, Wuhan University, Tsinghua University, as well as the Universities of Colorado and Denver. Experts say that PrintListener technology makes it possible to partially recreate fingerprints in 27.9% of cases and completely in 9.3% of cases. These results are said to far exceed the likelihood of success in MasterPrint attacks that offer random search of prints using a template library.^[9]

2022: Apple, Google and Microsoft switch to password-free authorization

May 5, 2022, and Apple Google Microsoft announced that they are switching to no authorization on password all mobile, dextic platforms and in. browsers Messages appeared on the official websites of the American corporations. PIN The company plans to introduce a single authorization standard using a fingerprint, face scan or code from 2023. Read more. here

2019: Hackers take fingerprints from a glass in a restaurant to unlock someone else's smartphone

At the end of October 2019, Tencent Security X-Lab's cybersecurity team showed how to fingerprint a glass in a restaurant and unlock someone else's smartphone with them.

At a special event in Shanghai, researchers invited viewers to touch the glass. Then the head of Tencent Security X-Lab, Chen Yu, took out his smartphone, photographed fingerprints and led them through a special application. After 20 minutes, the team was able to get a physical version of the fingerprints, which easily deceived three smartphones and two cars equipped with fingerprint scanners. Hardware for such a hacker attack will cost only $140, and only one application is needed from the software. Tencent Security X-Lab refused to clarify which method they used to create fingerprint copies.

Tencent Security X-Lab cybersecurity team shows how to take fingerprints from a glass in a restaurant and unlock someone else's smartphone with them

X-Lab experts claim they were the first to crack the ultrasonic fingerprint sensor used in smartphones along with two other types: capacitive and optical sensors. Earlier, Chen's team managed to crack older sensors with opaque reflective material. The design flaw found put about half a dozen smartphone models at risk, including the Huawei Mate 20 Pro.

Qualcomm-developed ultrasonic fingerprint sensors were considered the most reliable and fastest option than previous sensor types. They reflect sound waves from the tip of your finger to create a 3D image. X-Lab researchers said they had been developing an app to hack that system for months. They also noted that getting fingerprints from the surface of a smartphone is even easier than from a glass. All that researchers can advise for now is to regularly erase all prints.^[10]

Web Authentication (WebAuthn) is a biometrics standard for site authorization

The W3C Consortium (World Wide Web Consortium) and the FIDO Alliance (Fast IDentity Online) began work on Web Authentication (WebAuthentication) back in 2015. This API allows users to log into Google, Facebook, Dropbox, GitHub, and so on using the YubiKey hardware keys^[11].

Soon, the authorization process on many sites will be simplified and will be carried out using hardware keys in laptops, special applications or even fingerprint sensors on smartphones. On April 10, 2018, a number of organizations and companies, including Microsoft and Google, supported the new standard, thanks to which web developers will implement additional authorization mechanisms on their sites designed to reliably protect user accounts and data^[12].

users, it will be a natural transition. People already use their fingers and faces everywhere to unlock their mobile phones and PCs. […] What they use to unlock today will soon allow them to log in to their favorite sites and native applications, which already include Bank of America, PayPal, eBay and Aetna, "Brett McDowell, executive director of the FIDO Alliance, which is involved in the development of the standard, told Motherboard reporters

A standard called Web Authentication (WebAuthn) will allow sites or web services to use applications, hardware tokens or biometric data to authorize users instead of passwords or as the second stage of two-factor authentication.

The functionality is already available Mozilla Firefox in and will appear in and within the next few months Microsoft Edge. The Google Chrome developers also announced support for WebAuthn. Opera In addition, the Webkit engine development team (used in and) recently joined the corresponding working group. Apple Safari App Store

Based on the FIDO 2.0 Web API, WebAuthn was developed, which has broader capabilities and, in theory, allows you to abandon the use of passwords at all. For example, WebAuthn offers to use hardware keys, fingerprints, face recognition, iris scans and other biometrics for authentication on sites and in applications.

As McDowell explained, web developers wishing to implement WebAuthn support must implement JavaScript API support. Users will have to use the corresponding devices or applications to authorize on sites or services with support for the new standard. In this regard, Google and Microsoft will present the so-called "authenticators" of the FIDO2 by the end of this month. All devices running Windows 10 will receive them through the Windows Hello function, and most versions of Android will receive them through the Android Fingerprint API.

Method of "cheating" fingerprint sensors

As you know, fingerprints in each person are unique and two of the same do not exist. Still, even overlapping fingerprints can fool biometric authentication systems, according to researchers at New York University and Michigan State University. Those are, smartphones and other electronic devices with fingerprint sensors are not as secure as is commonly believed^[13] has been^[14]).

The problem is that because of their too small sizes, the sensors do not scan the entire surface of the finger, but only a small part of it. In addition, some mobile devices allow you to save several prints at once, and user identification occurs if any of them matches.

As the researchers suggested, in many people, fragments of fingerprints can coincide, and attackers are able to create a universal "master fingerprint" to bypass the biometric authentication system. Scientists analyzed 8.2 thousand fragments of fingerprints and found many matches. Using commercial fingerprint verification software, they found 92 potential "master prints" for each group of 800 randomly selected prints. The researchers considered a "master fingerprint" fragment that coincided with at least 4% of the rest in the group.

Note that in the course of the study, out of 800 full fingerprints of the entire surface of the finger, only one "master fingerprint" was found. The researchers analyzed a universal "master print" and divided into fragments. As it turned out, using fragments of a universal fingerprint, it is even easier to bypass the authentication system.

Cybercriminals learn to read fingerprints and iris patterns

According to the fall of 2016, Kaspersky Lab (Kaspersky) has already found at least 12 sellers on the black market offering skimmers who can steal fingerprint data and at least three researchers who are working on technologies that allow hacking of vein pattern recognition systems on the wrist and iris. According to experts, in September 2015, pre-sales testing of the first versions of biometric skimmers was already carried out on the black market. Then several errors were found, but the main problem was the use of GSM modules for transferring biometric data - they could not cope with large amounts of information, which means that new versions of such skimmers will use other, faster data transfer technologies, the company believes^[15].

It also became known that in the communities of cybercriminals, issues of developing mobile applications that allow masking human faces are being actively discussed. Such programs help to use photos of real people posted on social networks to deceive the facial recognition system.

"Unlike passwords or PINs, which are easy to change in the event of a hack, fingerprints or iris patterns cannot be changed. Accordingly, if biometric data is once in the wrong hands, its further use will be fraught with serious risk. That's why they need extremely reliable ways to protect themselves, "said Olga Kochetova, an information security expert at Kaspersky Lab. "The danger also lies in the fact that they are introduced into modern electronic passports and visas, which means that the theft of such documents leads to the fact that in the hands of the attacker is actually all the information on which the person's identity can be established."

Fingerprint sensors of some smartphones can be tricked using a printer

Computer security experts from the University of Michigan (USA) in 2015 hacked the fingerprint scanner, which is equipped with many modern smartphones, using a conventional inkjet printer. An article dedicated to the study was published on the website of the University^[16]

Until now, it was possible to deceive fingerprint scanners by creating a fake fingerprint manually, for example, from latex or glue, but this process takes a lot of time, and the quality of the resulting prints is sometimes too poor.

American researchers have found a simpler and more effective method. To do this, they scanned the index fingerprint of one of their colleagues with a resolution of 300 dpi, and then printed it on glossy paper, replacing the usual printer ink with conductive ones.

The resulting images were able to successfully deceive the built-in smartphones Samsung Galaxy S6 Huawei and Honor 7 scanners. For additional verification of the method, the researchers made fingerprints of other project participants, and in all cases they allowed them to bypass the built-in protection of devices, but smartphone sometimes a little more attempts were required to deceive the Huawei Honor 7 sensor.

According to the authors of the study, the method they discovered can be adopted by hackers, and smartphone manufacturers should think about improving the fingerprint scanners with which they equip their devices.

Voice biometrics

Voice is natural,
it does not require special knowledge and skills,
provides all people with equal opportunities

Voice biometrics is one technology that develops very quickly and allows different companies to use its solutions to identify customers. In the biometric system, individual behavioral, psychological and some other characteristics are used to determine or confirm identity. There are many biometric measurements, including iris, fingerprint, face, voice, signature, etc. Voice biometrics allows, by examining a person's voice characteristics, to identify the client. It is a relatively simple and cost-effective way to solve a number of practical problems.

Voice biometrics and speech technology are far from toys, it is a highly developed technology that can be used to improve the quality of the service to such an extent that the customer can feel this improvement. The company must provide the customer with an automated service, and speech technologies can help in this. No one makes the client wait, redirects or offers to use the menu. Voice communications are convenient for the customer.

The system understands the client and is able to check his words. He may not even remember the password or number. Voice biometrics, which is used during the conversation, allows you to establish who is calling. This reduces the conversation time. So the client does not need to introduce himself and name the password. His password is his voice! At the same time, he feels that his call is important and the company immediately makes a decision.

The technology is most widely used in the banking sector, in insurance companies, in telecom. Airlines are showing significant interest. The mobile application market for cell phones is also promising, where voice technologies are in full demand. In the automotive industry, voice systems allow you to use navigation devices on the go, are able to turn on music, air conditioning, help, without distracting from car control, record and send SMS, etc.

In medicine, speech technologies are used to record customer information, create electronic patient records. This makes it possible to optimize the work of doctors and creates clear advantages for clients. The doctor does not use the computer keyboard, he simply dictates medical indicators and diagnosis. The speech recognition system translates the voice into text and records it.

Banking contact centers use voice technologies. If the client needs basic information, then it is provided to him freely. But if he wants to carry out a financial transaction or some kind of transaction with his account, then his [status] needs to be checked. Voice biometrics is one type of client verification by which it is possible to identify whether it is a person alive, or a voice recording is broadcast.

From the presentation "What vacancies can natural speech processing systems apply for" Alexey Guryanov, Director of Voice Digital Technology at BSS

The voice biometrics system can identify the need for additional client verification. You can also create a "blacklist" of the votes of customers seen in fraud or in attempts to unauthorized access to other customers' accounts. This makes it possible to ensure the security of banking operations.

Experts predict a great future for voice biometrics, which in 2012 may play a leading role in user authentication. People are already used to using their voice in mobile search, device control and dictation, so the right approach to voice authorization could soon become an important part of the human identification process. These are the results of a recent study, "Voice Biometrics Authentication Best Practices: Overcoming Obstilles to Adoption." The main purpose of the study was to evaluate previous projects on the use of voice biometrics, as well as to analyze the current state of this industry and assess its prospects. According to the authors of the report, Valid Soft, voice biometrics can become part of a multi-level identification process in order to reduce the risk of online payment fraud. The data shows that the number of registered "prints" of the vote will increase from 10 million in 2012 to more than 25 million in 2015.

Speech Recognition (Technology, Market)

Main Article: Speech Recognition (Technology, Market)

Physician speech recognition

Main article: Recognition of doctors' speech

Speech technology

Speech Technology - Systems and Project Catalog

2018

Created software that clones a person's voice

In June 2018, a company Lyrebird software on a basis artificial intelligence that can copy the voice of anyone. The program records a person's voice and is able to copy it in a few minutes. The result of the system is so realistic that even relatives do not hear the difference and cannot determine who they are talking to on the phone, the developers say.

British tax authorities collected samples of votes of 5.1 million citizens

The UK Government Office for Tax and Customs Duties (Her Majesty's Revenue and Customers, HMRC) collected samples of 5.1 million British votes with the help of its new service Voice ID, launched in January 2017^[17].

Voice ID allows employees of call control call centers to identify citizens who have called on the phone by their voice. As reported on the HMRC website at the launch of the service, citizens themselves will be able to choose whether to use Voice ID or ordinary methods to identify them. However, UK rights organisation Big Brother Watch found^[18]that when trying to contact HMRC's call centre by^[19] you don't have to choose anything and those who call must provide a voice sample to use Voice ID.

The only way to avoid voice recording is to repeat "no" three times during the recording process. It is noteworthy that human rights activists have identified this method on their own, since it is not mentioned anywhere. According to experts at Big Brother Watch, without providing an easy way to choose whether to use Voice ID or not, HMRC violates the General Data Protection Regulation (GDPR), which recently came into force in EU countries .

Citizens can refuse to use voice authentication, but to do this they need to go through a long and difficult procedure. However, even after not using Voice ID, the voice sample still remains in the tax office database.

Human rights activists sent a request to HMRC under the Freedom of Information Act (FOIA), but the department refused to say how the user could remove a sample of his voice from the database. HMRC also refused to tell which government and private organisations it was handing over the samples.

2018: Chinese neural network clones people's voices in 4 seconds

Chinese specialists from Baidu have created artificial intelligence that can copy human voices in a few seconds,^[20] reported in March 2018]^[21].

A neural network called Deep Voice "learned" to simulate voices using a sound file with 2400 different voices lasting more than 800 hours. As a result, AI can copy any human voice based on a 4-second listen (a year ago, a neural network took about half an hour), and, according to the developers, a voice copy will be impossible to distinguish from the original. You can check the statement here^[22]

Artificial intelligence differentiates male and female voices and distinguishes British English from American.

Those skilled in the art believe that the invention can be applied in various fields: for example, to help people who have lost their voice; for simultaneous translation; for digital assistants, etc.

2016: HSBC customers will log in by voice

Retail customers of the British bank HSBC and its subsidiary First Direct will soon no longer need to type a password in order to gain access to a current account and conduct a transaction through a smartphone. The password will be replaced by voice identification. A massive transition to a biometric verification system will take place in early summer, writes Tim Wallack in an article published by The Telegraph^[23][24]

The voice verification method at the initial stage will be available for 15 million personal account holders. And as HSBC officials say, it will be faster, simpler and much safer. Bankers place particular emphasis on the latter circumstance. They know from experience that many customers often use the same password for several accounts linked to a mobile device at once. And because of this, they become easy prey for scammers.

Technically, the transition to the new system will take place as follows. A client who wishes to use it will need to provide a record of his voice to the bank. Based on this sample, the speed of speech, modulation and pronunciation features will be analyzed, which make the sound of speech of each person unique.

The system will be able to recognize the voice of the client, even if he freezes his throat.

After that, writes The Telegraph, the client will begin to access his accounts by pronouncing the agreed text. A pass, for example, can be the phrase "My voice is my password." According to Joe Gordon, the system will be able to recognize the voice of the client, even if he freezes his throat, which will certainly affect his speech. "More than 100 parameters are taken into account," he says. "A person's speech tract remains unchanged even in the case of a cold, and behavioral factors such as speech speed, accent or pronunciation remain in place."

And in those extremely rare situations, when the system still does not cope with the task, it will always be possible to use the usual verification, he adds.

Final tests of the voice verification system should be completed within the next few weeks so that by the beginning of summer 2016 the bank can offer it to customers. In addition, HSBC recently introduced fingerprint verification for account holders linked to. iPhone

According to The Telegraph, other large credit institutions have also taken the course of abandoning traditional passwords for mobile banking. Banking group Lloyds is testing a biometric verification system for debit cards linked to a smartphone. To demonstrate the capabilities that biometrics can provide, Lloyds even developed a device for recognizing a person by heart rate.

RBS also intends to use fingerprint verification. And Barclays, among other things, is experimenting with a scanner that, before allowing the client to make payments for a substantial amount, identifies it by the blood flow in the finger, The Telegraph previously reported^[25].

Micromovements

The goal of the project, implemented at the New York Institute of Technology, is to analyze the micro-movements and vibrations of the hand holding the smartphone, by which the user could be identified. Gestures and movements with which a person controls the phone are studied, as well as pauses between these gestures when viewing content.

Researchers at Cornell University programmed the popular Kinect sensor to analyze common household chores such as cooking and brushing. Their goal is to use motion recognition in smart homes and personal robotic helpers, although critics have maligned that this is clear and immodest proof that the sunset of society will begin with video games.

Gait

Artificial intelligence taught to determine personality by gait

At the end of May 2018, it became known about the ability of artificial intelligence to recognize people by gait. The development is proposed to be used primarily at airports.

Researchers from the University of Manchester (University of Manchester), together with colleagues from the Spanish Autonomous University of Madrid (Universidad Autónoma de Madrid), collected the world's largest database of people's gait, which included about 20 thousand foot movements of 127 people, made using special floor sensors and high-resolution cameras.

Airport passengers are recognized by their gait. We have already collected 20 thousand steps

All this data was uploaded to the neural network, which after training was able to recognize people by gait with almost 100 percent accuracy. The work of the AI system is based on the principle of deep residual learning, which allows you to identify a person by the spatial and temporal characteristics of his trace.

According to the author of the study from the University of Manchester Omar Castilla Reys (Omar Costilla Reyes), during the walk of each person, approximately 24 different movement parameters can be distinguished. Simply put, each person really has a unique model of movement.

Experts say the developed system will come in handy in crowded areas for security. Since the system is very accurate, with its help you can quickly notice a person who is wanted, as well as prevent terrorist attacks and other types of crimes.

Negotiations are already underway to use the neural network at airports to pass passport control. Compared to other methods of biometric identification, such as scanning fingerprints and retinas, gait recognition has clear advantages, since passengers do not need to stop to pass the check - they just need to go along a special touch track.^[26]

Development by Japanese researchers

Japanese researchers have found that with the help of 3D shooting, a person can correctly identify him by gait in 90% of cases. Moreover, the bare foot on the ground identifies the owner 99.6% of the time. This can help airport security - lines of people in socks march through their frames every now and then.

Analyzing User Actions

SRI International uses accelerometers and gyroscopes built into smartphones to obtain unique data describing a person's condition at times when they are walking or standing. The length of the step, the forces applied to maintain equilibrium, and the speed of movement are all individual parameters. Additional sensors can record other physical characteristics, such as the orientation of the hand or the physical position of the user - the proximity of him to other people, being in a sitting or standing position, trying to raise something, typing or talking on the phone.

Look:

• UnifyID
• Google Abacus

Stylometry. Keyboard handwriting

The features of the style are quite enough to distinguish people from each other. Specialists at Drexel University are trying to recognize the author's individual handwriting when typing text on a smartphone or tablet. Words, the use of grammatical constructs, the construction of phrases and even repeated errors are analyzed. This technology can be combined with other methods of keyboard authentication, such as analyzing the speed of typing and the duration of pauses between letter types.The use of this kind of technology makes the authentication system even more secure.

The contents of the password entered may not be the only unique difference between the user. By analyzing the speed and rhythm of keystrokes when entering a password, you can improve the reliability of authorization.

Indian scientists from the Chennaya College of Engineering suggested that a special set of print characteristics for each person will also help replace the standard password entry and save users from having to remember it. Everyone prints differently, and that could be key to creating a new way to authenticate.

The algorithm calculates the speed of printing, the duration of pressing the key and the pause between presses. The resulting statistics are assigned to a specific user and serve as its identifier.

Posture analysis

Japanese scientists have developed a system of 400 sensors in a seat that accurately identifies the contours and support zones of the human back and fifth point. The bum analyzer, which researchers say has 98% accuracy, can be adapted in automotive antitheft systems.

Lips

Lip identification technology was developed and patented at the Baptist University of Hong Kong in 2015. It is based on the recognition of characteristic movements of the lips, taking into account texture and facial expressions, for example, during the pronunciation of the password. Scientists argue that it is almost impossible to fake this kind of biometric trace^[27] of ^[28].

Lip identification can be used to improve the effectiveness of safety systems and serve as an addition to access methods such as facial recognition, retinal scanning, fingerprinting. The pilot application of the technology is planned to be introduced for service at ATMs and for controlling access to public places.

Nose

You can most accurately identify people by the retina of the eye, but British scientists have found an interesting replacement. They used the PhotoFace program and divided all noses of volunteers into six main types: Romanesque, Greek, Nubian, eagle, snub-nosed, twisted. The virtue of the method, they believe, is that noses are harder to hide or disguise. The drawback is also obvious - a nose scan gives a much less accurate result than checking the retina.

Analysis of the vein pattern in the palm of your hand

2023

In the metro of Tashkent launched a payment system using the palm

On November 30, 2023, in the khokimiyat of the city of Tashkent, it was reported that a fare payment system was launched in the local metro using a palm scan. The solution was developed by Uzinfok, a single integrator for the creation and support of state information systems in Uzbekistan. Read more here.

Japanese stores have a payment system for recognizing veins on fingers

On August 29, 2023, Japanese companies Hitachi and Tobu Railway announced the development of a new biometric identification system that will identify users by subcutaneous veins on their fingers and faces. In pilot mode, the platform has already begun to be used in stores in Japan. Read more here.

2018

In July 2018, a major Japanese supermarket chain, Aeon Group, announced the introduction of Fujitsu PalmSecure's vein pattern payment system. It is noted that we are talking about the first launch of such technology in retail stores. As part of the pilot project, some of Aeon's Ministop outlets began to be equipped with biometric scanners. Customers pre-register with the Aeon Credit Service, providing it with their personal data and images of the vein pattern from their palms. A credit card is linked to your account.

• Main article > >

2016

School cafeterias usually shine nothing at all, but one cafe in Florida stands out from the general row at least by using hand scanners, which replace students with money for lunch. The system will replace cards and PINs with a scanner in the red light range, and it does not require physical contact with the palm. It remains to modernize the school breakfasts themselves.

Fiserv (Faiserv) In the spring of 2016, a technology provider for financial services announced the launch of the palm scannerVerifast, a system by biometric authentications which financial institutions can reduce fraud, reduce transaction time and improve the quality of services offered.

Fiserv (Faiserv) hopes to change the quality of banking when customers visit a branch in person, speeding it up and making it safer by applying palm scanning technology.

Fiserv reported that palm authentication differs from traditional biometric authentication systems, which often work only in manual mode and are not very effective. This system will provide "fast, safe and accurate" identification of consumers by using an infrared sensor device that will read a unique pattern of veins in the palm of your hand.

Karl Guynn, director of product development at financial services firm Gesa Credit Union, which recently started testing the Verifast system in three of its branches, said: "We tested the traditional authentication process we use and found that it takes about 15 seconds to identify a person in the cash center. Authentication from the pattern of veins in the palm of your hand takes about one second. In other words, we accelerated each transaction by 14 seconds. Authentication by drawing veins in the palm of your hand significantly speeds up transactions, so a person can briefly look into the office to quickly carry out the operations he needs, and go further on their business. This process is characterized by higher reliability, our customers and cashiers rated it very highly. "

A study by financial group Raddon found that about 83% of consumers view palm authentication as one of the most useful features for bank transactions. Almost 97 percent of the service's beta testers reported that they are likely to use this technology after it is implemented in the banking environment.

Market estimates

2023: Global biometric lock market growth by 10% to $2.66 billion

In 2023, sales of biometric locks in the world reached $2.66 billion. The indicator of the previous year, when the market volume was estimated at $2.41 billion, was exceeded by 10%. This is stated in a study by Market Research Future, the results of which were published in early November 2024.

The biometric lock industry is experiencing significant growth, driven by an increasing demand for more efficient and convenient means of restricting access - both in residential and commercial premises. At the same time, there is a noticeable shift towards the integration of advanced technologies such as artificial intelligence and machine learning, which expand functionality and improve the effectiveness of biometric systems. Public and private investment in biometrics further fuels this trend, making such protections not just a preference, but a necessity in many areas.

By the type of technology used, the industry is segmented into recognition of fingerprints, faces, iris and voice. In 2023, the first of these sectors dominated with an estimate of $0.9 billion. Face recognition tools brought in $0.7 billion, iris - $0.5 billion. Revenue in the voice recognition segment amounted to $0.56 billion. North America led with revenues of about $1.06 billion. Next come Europe with $0.78 billion and Asia-Pacific with $0.62 billion. South America, the Middle East and Africa accounted for $0.2 billion. Significant market players are:

• Axis Communications;
• Kisi;
• Amano;
• Gunnebo;
• Kaba;
• Nedap;
• Amazon;
• Covenant Security;
• Tyler Technologies;
• Sierra Wireless;
• SentrySafe;
• ZKTeco;
• SMART Technology;
• Hikvision;
• Nymbus.

Market Research Future analysts believe that in the future, the CAGR (average annual growth rate in complex percentages) will be 10.25%. Thus, by 2032, sales of biometric locks in the world can reach $6.4 billion.^[29]

Cognitive imprint

No one considers memorizing long strings of numbers and letters as outdated as the US defense agency DARPA. It is developing the concept of "cognitive fingerprints," which can combine iris scanning, keyboard handwriting and even web surfing habits for continuous user authentication.

System "trolling"

Of course, no one wants to receive error messages, but these messages can play an important role in ensuring security. Randomly displaying error messages and observing the reaction of users, specialists from the Southwestern Research Institute are preparing to identify users and recognize intruders. Therefore, the next time the PC informs you of a memory shortage and asks if you want to fix the problem, carefully check everything. Perhaps they are checking you.

Individual interface settings

Specialists from the University of Maryland use visual streams for additional authentication of a PC user or phone. On a desktop computer, the program analyzes typical ways for its owner to arrange and resize windows, work patterns and restrictions when moving the mouse. Three video streams are checked on the phone at once: the image taken by the front camera, environmental details (as well as shoes and clothes) taken using the camera on the back panel, and manipulating the screen. The researchers expect that combining these three threads will be enough to authenticate each individual user and conduct repeated checks while he is using his device.

DNA test (in hours of hair)

This method almost unmistakably identifies the owner, but it never became everyday due to the cost of time and high cost. But several groups of researchers are actively working to make the process cheaper and faster.

Biologists from the Lawrence National Laboratory in Livermore (USA) have learned to almost accurately determine a person's identity not only by DNA, but also by the proteins contained in the hair. The method, according to the creators, will revolutionize forensics and archeology, reports in the fall of 2016 the journal PLoS One^[30]

Although the technique of protein analysis is still far from perfect, it allows you to accurately establish your identity not only in the first hours and days after the death of a person, but even a few centuries after his death. Scientists have successfully identified several people who died more than 250 years ago.

The technique works as follows: hair is dissolved in special substances that do not destroy the chemical structure of proteins inside them, and analyze the composition of this "soup" for the presence of 185 mutations in the structure of proteins, the combination of which is unique for each inhabitant of the Earth.

According to experts, such a set of proteins is redundant - in fact, about a hundred such markers will be enough for accurate identification. Now scientists are working to simplify and reduce the cost of the technique so that it can be conveniently applied in practice during criminal investigations and during excavations.

The method has already been tested on six dozen European Americans who agreed to donate blood and hair for DNA and protein tests. According to biologists, the personalities of each of them managed to correctly establish what opens the way for the use of proteins in calculating the personalities of criminals for very small portions of hair. For this, a sample weighing only 1 milligram is enough, which is slightly more than contained in one hair.

Auricle

The auricle is not only suitable for protecting the auditory canal. The developed system remembers the tubular structure of the middle ear and the general shape of the auricle to create an "ear print" that allows accurate identification of the owner in 99.6% of cases.

Losing or stealing a gadget can cause the owner a lot of problems. And this is not only material losses due to the loss of an expensive gadget, but also the threat of unauthorized access to personal or corporate data, an online banking application or an electronic wallet.

In the spring of 2015, Yahoo Labs proposed a new option - to identify the owner of the smartphone by its auricles. Scientists used the fact that two identical auricles do not exist in nature - this is unique biometric information^[31].

Since the touch screen touches the ear during the call, this can lead to chaotic pressing of the on-screen buttons with the ear, so the smartphone screen is always turned off in talk mode. At this point, scientists at Yahoo Labs suggested that the owner be authenticated. Testing of the development showed that the accuracy of biometric identification using the imprint of the auricle is 99.52%. The created software was called Bodyprint.

Interestingly, Yahoo Labs also proposed modifying the procedure for taking a call. So, until now, all phones always demanded to press the receive button. Now you can just attach your smartphone to your ear: Bodyprint software recognizes that your ear is pressed and starts broadcasting your voice.

Since Yahoo itself does not show interest in the production of gadgets, it is likely that either licensing of the technology will be carried out, or the corresponding software will be distributed as a mobile application through the Apple Store and Google Play stores. The main advantage of the created technology is, in the absence of the need to install a special sensor for fingerprint. The touch screen is in any smartphone, the technology can be used in any gadget, so its prospects are very significant.

In June 2015, it became known that Amazon received a patent for technology that will unlock the phone with the ear. If the technology is implemented, it will work as follows: a person brings the phone to his ear, the front camera takes a picture of the auricle and compares the photo with the one in the database. That is, the technology is similar to the one used for authorization using a fingerprint^[32].

The video for the PatentYogi project explains how authorization will work. Perhaps the company will use this technology so that when answering a call, it is enough to simply bring the phone to your ear.

Neural connections instead of fingerprints

A unique system of connections in the individual's brain can be used to identify individuals, like fingerprints. Such conclusions were reached by American scientists, authors of an article in the journal Nature Neuroscience, conveys the Lenta.ru^[33]

Neurophysiologists traditionally use a number of methods to represent the structure, function, and biochemical characteristics of the brain (e.g., computed tomography) to compare brain activity in different groups of people. These methods, or neuroimaging, make it possible to understand one or another feature of the brain that is characteristic of all people, and individual features are usually ignored.

However, Emily Finn and her colleagues found that the unique features of the connections of various parts of the human brain are stable enough to accurately establish the identity of the subject. Finn worked with 126 participants in the Human Connect project. It turned out that the pattern of connections obtained during one of the neuroimaging sessions (at rest, when passing tests for memory, emotions and language skills) in subsequent sessions remains unchanged - and it is easy to determine an individual among other participants in the experiment.

Moreover, the connection drawings helped predict the level of mobile intelligence (the ability to perceive and remember new things, to solve problems that a person had not encountered before). The main prognostic factors of this ability were connections between the frontal, parietal and temporal lobes of the brain.

In addition to Oxford University, universities in Washington and Minnesota take part in the Human Connectome project. Connectom refers to the totality of all connections between neurons, the number of which in the human brain is estimated at quadrillion. The project was launched in 2010, with a budget of $40 million.

This technology has another significant advantage. 'A fingerprint could be stolen and a person would then be unable to grow a new finger to replace it. Brain prints, however, can easily undergo changes: the user simply comes up with a new thought combination, "said Professor Sarah Laszlo, one of the participants in the project to develop brain identification.

Implant a chip or swallow a microcomputer

The most original and unusual method of identity identification was offered by PayPal. The company simply offered to implant chips or swallow microcomputers. This will solve the problem of authorization radically and forever. Chips and microcomputers will analyze the pulse, the composition of gastric juice and other internal biometric information. The details, however, are not specified, but such a bold approach to solving the problem is at least admirable.

Notes