Systems and methods for smart antifraud protection

Abstract

A system and a method of operating an anti-fraud, AML and ATF system, through the use of adapted wireless internet devices, using a "selfie" or a picture of the head / face of a user. The system of an example of the invention for operating in an anti-fraud, AML and ATF systems uses set of different methods to generate a unique correct / optimise Encoding Data of the given picture; The set of applied methods are in charge of automating the optimisation of the generated Encoding, improving the authentication system for certain anti-fraud, anti- money laundering (AML) or anti-terrorist financing (ATF) aspects through the use of fixed or wireless devices or portable computers or desktop computers, with at least one built in camera and application software in the device as per examples of this invention, such devices adapted as per examples of this invention, whereas the system server is adapted to include a local processing unit and / or a remote server processing unit adapted as per examples of this invention.

Description

[0001] SYSTEMS AND METHODS FOR SMART ANTIFRAUD PROTECTION

[0002] BACKGROUND OF THE INVENTION

[0003] 1. Field of the Invention

[0004] The field of the invention relates to systems and methods for antifraud protection, for example as part of a payment system, or for use in payment systems.

[0005] 2. Technical Background

[0006] Traditional methods of executing payments, or any type of secure or financial transactions, in existing systems and methods have evolved from physical authentication and execution methods to so-called digital authentication and remote execution through a server and an app on a smartphone.

[0007] The evolution of technology has accelerated the transformation of remote execution of transactions, thanks to the internet and its associated recent clearer regulations (in particular in the financial sector), which have helped to create the right framework for remote transactions from the comfort of the home. Consequently, however, the methods with which to commit fraud have also evolved, resulting in a large number of people attempting to use the available platforms for illicit purposes.

[0008] To counteract this wave of fraud, money laundering or terrorist financing, a multitude of contingency methods have been developed to combat the illicit use of the prior art in online systems and methods used widely, which are worrying for the financial and social "health" of different countries.

[0009] The most advanced methods for anti-fraud and anti-money laundering or anti -terrorist financing, have added in their systems many things for security, from verifying patterns in transactions or checking destination country or destination person’s name against blacklists shared between & with multiple sources, now adding a strong focus on stopping potential fraudsters that correlate more often than not, equating to money laundering as these people who commit fraud hardly were planning to pay any taxes on the amount de-frauded from other non-suspecting honest citizens or businesses or even governments.

[0010] In some cases, preventing illicit financial transactions by participants willing, for example, to fund terrorist activities or to provide their account credentials to organized gangs committing widespread fraud and, in some cases, to charge poorer segments of the population for access to these accounts, is extremely difficult with the state of the art, except after the fact, when someone reports such potential cases to the authorities. Even blocking such accounts is only a drop in the ocean, as they can use many accounts created by genuine users and obtain free or paid access to these accounts in the name of third party willing or un-willing participants for their illicit means of collecting money. A typical example of fraud in underdeveloped countries is the illegal sale of health services without the need to be a user of any public or private health entity, thus obtaining such health or government benefits.

[0011] In general, these fraudulent financial movements are made possible by certain participants who are willing to provide their current account credentials to criminals or gangs who organize such frauds, the funds of which are used for illicit purposes.

[0012] However, there are more worrying cases than others, there are less developed countries where the rate of fraud is unsustainable in the long term, where there are organized gangs and it is very difficult to prevent such fraudulent movements, either because they already know the loopholes in the prior art systems or methods or because they have people within the system involved in these financial movements, which greatly facilitates the misuse of the prior art systems or methods. A specific example of fraud in underdeveloped and some cases also in developed countries is the illegal sale of health services without the need to be a user of any public or private health entity, impersonating another person who is a genuine user of such health or government health or benefits services, in order to benefit from those various services in an abusive and illicit manner. Cases like the above example are much more complex to detect with the currently applied prior art methods or systems. This is, amongst other things, due to the limitations of such methods, which recognize this type of financial services transfers based on the patterns found during data tracking, however, at no time is the actual identity of the parties involved examined properly other than a simple login with the correct credentials or a security code to confirm a given transaction (which in the event of using third party accounts credentials and associating the fraudster’s mobile phone number to that account is a simple way of avoiding detection), keeping the prior art systems and methods ignorant of this information, overlooking relevant and crucial information in the detection of prohibited activity carried out by a person who is, in fact, different than the person associated to the transaction account or different than the person on the debit / credit card.

[0013] That said, it is not sufficient to add end-user identification information to the prior art system, since, as mentioned above, in certain cases individuals within the entity or platform where the fraudulent movements take place could also be involved, as they have greater freedom to facilitate fraud. Therefore, focusing on a single part of the process is neither a complete nor an effective solution. This is why examples of our invention solve most of these previous mentioned shortcomings.

[0014] Some of the shortcomings of the prior art are as follows:

[0015] 1. A typical shortcoming occurs when a user or many users give their credentials to an individual or an organized gang or illicit organization with the intent to commit fraud or to launder money or for terrorist financing, and they execute transactions without proper limits because these illicit persons are not identified by any of the current known methods or systems.

[0016] 2. Yet another shortcoming, when using the prior art methods and / or systems to execute transactions, is that no information is collected from the illicit person actually carrying out the transaction, so if, for example, a user denies having made a payment, there is no way to find the real executor or confirm that it was the actual owner of such health or government benefits account / bank account / wallet / prepaid-debit-credit card, leaving the service provider, government or bank, or an unwilling or non- suspecting account owner as the victim for financial or services transaction fraud with no real data with which to defend the case against the actual perpetrators.

[0017] 3. There is a more worrying shortcoming use-case where the prior art system(s) and / or methods have no control or reliable counter measure when an internal staff of a merchant, service provider, bank or health provider or government benefits provider is involved in facilitating an illicit transaction. This is the case where for example the merchant or one of its employees is the fraud facilitator rather than the paying enduser. There are many similar use-cases where the merchants use their good position / brand / image and consequently little control is exercised over them to carry out sufficient reliable anti-fraud checks on the transactions of all their staff to detect potentially fraudulent activities and abusing of the good position / brand / image of the merchant.

[0018] 4. Another shortcoming is the inability in the prior art to prevent fraud rather than correcting it (meaning detecting it after the fact and invoking refunds amongst other prior art measures), since there is no data to identify the actual fraudster person in this world of online transactions, because for example when paying with a prepaid / debit / credit card or Apple Pay, Google Pay, etc, who have one or more of Cards stored, there is no way for the merchant to recognize if the paying user is the true owner of such paying card before processing the actual payment or any such further transactions on a repeat basis in the same or other merchants.

[0019] 5. Another shortcoming in the potential fraud increase is due to fraudsters getting more experienced and knowledgeable in learning about prior art used transaction pattern recognition systems and / or methods by trial and error and thus being able to identify the actual patterns, limits and detection triggers. By the prior art maintaining the same algorithms for a long period of time, fraudsters know the techniques and methods to use to avoid detection. Techniques and methods that fraudsters, money launderers or terrorist financiers can then use more widely at scale online targeting multiple service providers (in particular financial institutions) which do not have a sufficient reliable prior art method or system to resolve the shortcomings explained here in the prior art and reduce significantly statistically these illicit practices.

[0020] 6. Yet another shortcoming of the prior art is that in all prior art, the facial recognition systems and methods that monitors the creation are not reliable enough in truly identifying the actual person of the account opening versus the person performing each transaction from that account. For example, but not limited to, bank accounts which are then used for fraud or illicit purposes by one or more different unidentified or, in the opposite case, misidentified persons, repeatedly logging in with the details of the person who actually opened that bank account. Prior art online based systems and methods never checked that the encoded generation of the different points of the face of the person's selfie or video during account opening are correctly positioned prior to storing it in the data base, and worse even is that the person making any outgoing transaction (bank transfer or card payment, etc) is not identified at all and no face points data are available at all of the person(s) making actual transaction(s). Therefore, in the event that a person at account creation resembles to any other person due to the facial points data inaccuracy, such other person not being the actual genuine account representative but his / her facial points data resemble above the threshold of comparison to the inaccurate facial data of the actual account owner, then he / she would have irregular and unlimited access to the bank account of someone else, being able to use the account without even having the same appearance as the legal representative of the account.

[0021] 7. Another shortcoming is the opposite of not changing the anti-fraud detection for rules for a long period of time, namely when continuous changes are implemented in the anti-fraud rules and verifications by a bank, insurance, health provider, government benefits provider or any financial services company, these continuous changes acts as a firewall of rules on transactions and from an automated strong wall which as much as it benefits detection of potential fraud transaction it typically also comes together with an increase of an unacceptable amount of false positives that affect innocent genuine honest users. Finding an acceptable trade-off between how many false positives are acceptable to catch a certain percentage of potential fraudulent transactions versus the cost of business is a difficult or almost impossible balancing act, which fraudsters exploit and count on to keep committing fraud at the percentage of the trade-off of prior art systems and methods designed to annoy the least possible genuine honest users which opens a reasonable profitable window to keep committing fraud.

[0022] 8. Finally, a different shortcoming of the prior art systems and / or methods are the fact that facial recognition systems and / or methods are highly dependent on the data with which they have been trained and, as is commonly known, those systems and / or methods currently available are highly discriminatory towards certain ethnic groups, or in other words, highly unreliable and inaccurate across ethnic groups thus forcing facial comparison methods to relax the thresholds to avoid too many false positives. This makes the security of the prior art systems and / or methods directly dependent on the ethnic group to which the identified person belongs, leaving a large security gap in certain ethnic groups of users.

[0023] Although traditional transaction methods are perfectly workable as a business as they were profitable businesses in their own right, the fact is that there are still many shortcomings that need to be overcome in a world that evolves towards more wide spread global use of digital technologies.

[0024] In most cases, major transacting companies, even if they operate under a license from a third-party company, are usually liable for any potential fraud carried out through their platform. However, they do not usually have sufficiently powerful or complex tools to counteract the threats posed by fraudsters or money launderers or terrorist financiers, but rather have only the capability of basic tools, which, as mentioned above, are not effective.

[0025] Examples of our invention, system and methods, resolve the above mentioned prior art shortcomings.

[0026] 3. Discussion of Related Art

[0027] WO2022219351A1 discloses a system and method of operating an anti-fraud, AML and ATF system using a “selfie” (photo of one self) or a face extracted from a photo to become the “selfie” as the key information to automate certain anti-fraud, antimoney laundering (AML) or anti -terrorist financing (ATF) aspects through the use of fixed or wireless devices or portable computers or desktop computers, with at least one build in camera and an application software in the device, wherein the system server is adapted to include a local processing unit and / or a remote server processing unit. The “selfie” is converted to replace the original, requiring less memory space and thus less processing -power or -time when comparing pairs of photos or selfies. SUMMARY OF THE INVENTION

[0028] According to a first aspect of the invention, there is provided a system including at least one or more internet enabled devices (e.g. wireless devices, smartphones, portable computer, desktop computer), each internet enabled device including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executable on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, the at least one server including a respective second non-transitory storage medium (NTSM2), and a respective second computer program product embodied on the respective second non-transitory storage medium, the respective second computer program product executable on the at least one server to communicate with a first internet enabled device of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device is configured to use the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, wherein the respective second computer program product is executable on the at least one server to process the extracted digital image data, wherein the respective second computer program product is executable on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response to store the new digital image data in the respective second non-transitory storage medium of the at least one server and to index the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device, and wherein;

[0029] (i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product adapted to analyse image data for a respective different ethnic group to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and

[0030] (ii) wherein each resulting output “encoding data 1 to n” of the n outputs is used as input to a respective fourth computer program product in the at least one server which is executable on the at least one server to compare all the n outputs and to ignore all such outputs that are identical and to calculate an interpolation between outputs that are different to each other to generate a new modified output, referred to as an “auto modified encoding data” (AMED), and to store such “auto modified encoding data” in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device, and

[0031] (iii) wherein, in the event of only one output of the n outputs being different and all other outputs of the n outputs being the same, then no interpolation is performed in (ii), and the one different output of the n outputs becomes the “auto modified encoding data” (AMED) which is stored in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device. Advantages include improved antifraud protection, and improved security, and improved computer system security.

[0032] According to a second aspect of the invention, there is provided a system including at least one or more internet enabled devices (e.g. wireless devices, smartphones, portable computer, desktop computer), each internet enabled device including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executable on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, the at least one server including a respective second non-transitory storage medium (NTSM2), and a respective second computer program product embodied on the respective second non-transitory storage medium, the respective second computer program product executable on the at least one server to communicate with a first internet enabled device of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device is configured to use the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, wherein the respective second computer program product is executable on the at least one server to process the extracted digital image data, wherein the respective second computer program product is executable on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response to store the new digital image data in the respective second non-transitory storage medium of the at least one server and to index the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device, and wherein;

[0033] (i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product adapted to analyse image data for respective different parts of a head of a person to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and

[0034] (ii) wherein each resulting output “encoding data 1 to n” of the n outputs is used as input to a respective fourth computer program product in the at least one server which is executable on the at least one server to compare all the n outputs and to ignore all such outputs that are inconclusive (e.g. in which a match is not produced) and to generate a new modified output, referred to as an “auto modified encoding data” (AMED), and to store such “auto modified encoding data” in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device, and

[0035] (iii) wherein, in the event of all outputs being inconclusive (e.g. in which in all a match is not produced) then the new digital image data becomes the “auto modified encoding data” (AMED) which is stored in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device. Advantages include improved antifraud protection, and improved security, and improved computer system security.

[0036] According to a third aspect of the invention, there is provided a system including at least one or more internet enabled devices (e.g. wireless devices, smartphones, portable computer, desktop computer), each internet enabled device including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executable on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, the at least one server including a respective second non-transitory storage medium (NTSM2), and a respective second computer program product embodied on the respective second non-transitory storage medium, the respective second computer program product executable on the at least one server to communicate with a first internet enabled device of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device is configured to use the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, wherein the respective second computer program product is executable on the at least one server to process the extracted digital image data, wherein the respective second computer program product is executable on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response to store the new digital image data in the respective second non-transitory storage medium of the at least one server and to index the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device, and wherein;

[0037] (i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product adapted to analyse a respective image of a different respective viewing angle of a head when viewed from the respective built-in camera, to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and

[0038] (ii) wherein each resulting output “encoding data 1 to n” of the n outputs is used as input to a respective fourth computer program product in the at least one server which is executable on the at least one server to compare all the n outputs and to ignore all such outputs that are inconclusive (e.g. in which a match is not produced) and to calculate an interpolation between the conclusive (e.g. in which a match is produced) outputs that are different to each other, to generate a new modified output, referred to as an “auto modified encoding data” (AMED), and to store such “auto modified encoding data” in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device, and

[0039] (iii) wherein, in the event of only one output being conclusive (e.g. in which a match is produced) and all other outputs are inconclusive (e.g. in which a match is not produced), then no interpolation is performed and the one conclusive (e.g. in which a match is produced) output becomes the “auto modified encoding data” (AMED) which is stored in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device. Advantages include improved antifraud protection, and improved security, and improved computer system security.

[0040] The system may be one wherein the digital image data of the first internet enabled device, consists or comprises of a single image, or consecutive images, or a short video, wherein in the case of a short video, images are extracted from the short video corresponding to the face or head of a person, or to an object, and the extracted images are converted into respective two-dimensional digital matrix arrays x, y where “x, y” are the coordinates of each matrix array relative to a reference point, and wherein the single image, or the consecutive images, or the converted extracted images, are transmitted through the communication channel between the first internet enabled device and the at least one server, for processing.

[0041] The system may be one wherein the digital image data of the first internet enabled device, consists or comprises of a single image, or consecutive images, or a short video, wherein in the case of a short video, images are extracted from the short video corresponding to the face or head of a person, or to an object, and the wherein the image, or the consecutive images, or the extracted images are converted into respective three-dimensional digital matrix arrays (x, y, d where “x, y” are the coordinates of each array element in the matrix and the 3rddimension “d” corresponds to the respective distance from the respective built-in camera to each respective array image element) and wherein the respective three-dimensional digital matrix arrays are transmitted through the communication channel between the first internet enabled device and the at least one server, for processing.

[0042] The system may be one wherein the digital image data of the first internet enabled device, consists or comprises of a single image, or consecutive images, or a short video, wherein in the case of a short video, images are extracted from the short video corresponding to the face or head of a person, or to an object, and the wherein the image, or the consecutive images, or the extracted images are converted into respective three-dimensional digital matrix arrays (x, y, c where “x, y” are the coordinates of each array element in the matrix and the 3rddimension “c” corresponds to a respective colour code of each respective array element in the matrix) and wherein the respective three-dimensional digital matrix arrays are transmitted through the communication channel between the first internet enabled device and the at least one server, for processing.

[0043] The system may be one wherein the digital image data of the first internet enabled device, consists or comprises of a single image, or consecutive images, or a short video, wherein in the case of a short video, images are extracted from the short video corresponding to the face or head of a person, or to an object, and the wherein the image, or the consecutive images, or the extracted images are converted into respective four-dimensional digital matrix arrays (x, y, c, d where “x, y” are the coordinates of each array element in the matrix and the 3rddimension “c” corresponds to a respective colour code of each respective array element in the matrix and the 4thdimension “d” corresponds to the respective distance from the respective built-in camera to each respective array image element) and wherein the respective four- dimensional digital matrix arrays are transmitted through the communication channel between the first internet enabled device and the at least one server, for processing.

[0044] The system of the first aspect of the invention may be one wherein the n sub-program products, are pretrained neural network program products (PTNNPP1 to n), wherein each respective pretrained neural network program product is pretrained for a specific respective ethnic or sub-ethnic-group of an ethnic group (e.g. Hispanic-Spain, Hispanic descent-Mexico, Hispanic descent-Peru, etc., African descent-North America, African descent-United Kingdom, African descent-Nigeria, African descent- Kenya, African descent South-Africa, etc., Asian descent-China, Asian descent-Japan, Asian descent-Indonesia, Asian descent- Singapore, Caucasian descent-North America, Caucasian descent-United Kingdom, Caucasian descent-Germany, Caucasian descent- South Africa etc.) and is usable to generate a respective mathematical representation of the face of a person, to generate a respective output of the n outputs.

[0045] The system of the second aspect of the invention may be one wherein the n subprogram products, are pretrained neural network program products (PTNNPP1 to n), wherein each respective pretrained neural network program product is pretrained for a specific respective area or sub-area of a head of a person or a face of a person (e.g. area enclosing both eyes, or the sub-area of each enclosure of each eye separately, the area above the eyes / forehead or sub-area of the eye-brows separately from the area above the eye-brows, the area below the eyes but above the lips or the sub-areas of the nose separately to the sub-area of each cheek, the area of the nose to include the nostril and bridge of the nose, the area mouth or the sub-area of the upper and lower lip separately to the area left and right of the lip endings towards each lower cheek area, the area of the chin, the area of the neck, the area of each ear or sub-area of each earlobe separately) and is usable to generate a respective mathematical representation of the face or head of a person, wherein the n sub-program products are usable to generate the n outputs.

[0046] The system of the third aspect of the invention may be one wherein the n sub-program products are pretrained neural network program products (PTNNPP1 to n), wherein each respective pretrained neural network program product is pretrained to analyse an image of a respective viewing angle of a reference point on the face of a person when viewed from the respective built-in camera, wherein: for a two-dimensional image data (x,y) such reference point on the face to calculate an angle to the respective built-in camera is the position of iris or centre point of the pupil versus the background sclera of the eye, where the angle is defined as:

[0047] -90 degrees LEFT when the “iris or centre point of the pupil” (IRCPP) is on the far left of the sclera of the eye, +0 degrees RIGHT / LEFT when the “iris or centre point of the pupil” (IRCPP) is in the horizontal centre of the sclera of the eye, +90 degrees RIGHT when the “iris or centre point of the pupil” (IRCPP) is on the far right of the sclera of the eye, and in between, the degrees are a linear interpolation of the mathematical calculation between the defined -90 degrees LEFT, +0 degrees RIGHT / LEFT and +90 degrees RIGHT, and

[0048] -90 degrees DOWN when the “iris or centre point of the pupil” (IRCPP) is on the lower end of the sclera of the eye, +0 degrees UP / DOWN when the “iris or centre point of the pupil” (IRCPP) is in the vertical centre of the sclera of the eye, +90 degrees UP when the “iris or centre point of the pupil” (IRCPP) is on the upper end of the sclera of the eye, and in between, the degrees are a linear interpolation of the mathematical calculation between the between the defined -90 degrees DOWN, +0 degrees UP / DOWN and +90 degrees UP, and resulting in a two-dimensional horizontal angle (left / right) and a two-dimensional vertical angle (up / down), and is usable to generate a respective mathematical representation of the angle(s), wherein the n sub-program products are usable to generate the n outputs.

[0049] The system of the third aspect of the invention may be one wherein the n sub-program products are pretrained neural network program products (PTNNPP1 to n), wherein each respective pretrained neural network program product is pretrained to analyse an image of a respective viewing angle of a reference point on the face of a person when viewed from the respective built-in camera, wherein: in a three-dimensional image data (x,y, z) such reference point on the face to calculate the angle to the respective built-in camera is the distance z from the respective built-in camera to each of those points x, y, where the angle is defined as; the linear interpolation between the delta of the distance zl of a reference point on the upper half of the face and the distance z2 of a mirror reference point on the lower half of the face as the UP / DOWN angle, and the linear interpolation between the delta of the distance z3 of a reference point on the left half of the face and the distance z4 of a mirror reference point on the right half of the face as the LEFT / RIGHT angle, and resulting in 3 dimensional horizontal and vertical angle representation, and is usable to generate a respective mathematical representation of the horizontal and vertical angle representation, wherein the n sub-program products are usable to generate the n outputs.

[0050] According to a fourth aspect of the invention, there is provided a system including at least two or more internet enabled devices (e.g. wireless devices, smartphones, portable / desktop computer), each internet enabled device including a respective built- in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executable on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, wherein each respective first computer program product is configured to communicate with a second computer program product stored in a second non-transitory storage medium (NTSM2) of the at least one server, the second computer program product executable on the at least one server to communicate with a first internet enabled device (IED1) of the at least one or more internet enabled devices, and to communicate with a second internet enabled device (IED2) of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device is configured to use the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, and wherein the respective first computer program product when executing on the second internet enabled device is configured to use the respective built-in camera of the second internet enabled device to extract digital image data from the respective built-in camera of the second internet enabled device, and to transmit the extracted digital image data through a communication channel between the second internet enabled device and the at least one server, wherein the respective second computer program product is executable on the at least one server to process the extracted digital image data, wherein the respective second computer program product is executable on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response to store the new digital image data in the respective second non-transitory storage medium of the at least one server and to index the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device; wherein the respective second computer program product is executable on the at least one server to detect new digital image data being received from the second internet enabled device of the at least one or more internet enabled devices, and in response to store the new digital image data in the respective second non-transitory storage medium of the at least one server and to index the new digital image data such that the new digital image data is associated with an originating user account of the second internet enabled device, and wherein:

[0051] (i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product adapted to analyse image data to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and

[0052] (ii) wherein each such output of the n outputs is used as input to a fourth computer program product in the at least one server which is executable on the at least one server to compare the n outputs with corresponding historical past n outputs of the user account of the first internet enabled device and / or of the user account of the second internet enabled device, stored in the second non-transitory storage medium, indexed to the corresponding user account (in example, compares EDI with ED1-1 to EDl-n, where EDI is the current new input data and ED1-1 is the previous historical input data and EDl-n is most past historical input data when that user account was opened, AND compares ED2 with ED2-1 to ED2-n, where ED2 is the current new input data and ED2-1 is the previous historical input data and ED2-n is most past historical input data when that user account was opened), AND

[0053] (iii) wherein, in the event of a mismatch (meaning the match is below a predefined threshold “x”) between the n outputs with the corresponding historical past n outputs of the user account of the first internet enabled device OR a mismatch between the n outputs with the corresponding historical past n outputs of the user account of the second internet enabled device, then the transaction of the user account corresponding to the mismatch is denied by the system (e.g. if EDI shows a mismatch it means it is not the same person as the authorised person of the user account of device IED1 and this user account will be denied the transaction that it was intending to do with device IED2, if ED2 shows a mismatch it means it is not the same person as the authorised person of the user account of device IED2 and this user account will be denied the transaction that it was intending to do with device IED1), AND

[0054] (iv) wherein, in the event there is a match above a predefined threshold “y” between the n outputs with the corresponding historical past n outputs of the user account of the first internet enabled device AND a match between the n outputs with the corresponding historical past n outputs of the user account of the second internet enabled device, then the transaction between the users of the first internet enabled device and the second internet enabled device is allowed to proceed by the system, allowing the first internet enabled device and the second internet enabled device to instruct any such transaction between their corresponding users, respectively through the respective communications channels (CCH1 and CCH2) for execution by a fifth computer program product of the at least one server, wherein the users of the accounts corresponding to the first internet enabled device and to the second internet enabled device could be two end-users, two merchants, an end-user and a merchant, etc. and wherein the transaction requested for execution could be a bank transfer, an online payment, a physical point of sale payment, a government benefits request, an insurance claim, a mortgage application, a loan application, etc. Advantages include improved antifraud protection, and improved security, and improved computer system security.

[0055] The system may be one wherein only the check of a match or mismatch is performed on the encoding data corresponding to an end-user of a device (IED1), comparing EDI with ED 1-1 to EDl-n and in the event of a match, the system will allow the transaction request to proceed between IED1 and IED2 through SI.

[0056] The system may be one wherein the system does not do any checks but instead sends the encoding data of the end-users of the device (IED1) to the screen of the merchant device IED2 and it is the merchant of device IED2 that checks visually, and if the merchant confirms the user in front of him (in example in a physical point of sale usecase) he matches with the image corresponding with the encoding data of the user in front of him (ED1-1 of a past successful transaction or with EDl-n the original image encoding data when end-user of device IED1 opened his account), if merchant confirms manually by clicking on his device IED2 then through communication channel CCH2 it notifies the computer program product in SI and the transaction can proceed, otherwise the transaction is denied and communicated by SI computer program product to both user devices IED1 and IED2 respectively through communication channel CCH2.

[0057] The system may be one wherein the check of a match or mismatch is performed on the encoding data corresponding to an end-user of a device (IED1), comparing EDI with ED1-1 to EDl-n and wherein; i. in the event there is a match above a predefined threshold “y” the system will allow the transaction request to proceed between IED1 and IED2 through SI. ii. in the event of a mismatch (meaning the match is below a predefined threshold “x”) between EDI and ED1-1 to EDl-n the system will deny the transaction request. iii. in the event there is a match above a predefined threshold “x” BUT below a predefined threshold “y” the system will send the encoding data of the end-users of the device (IED1) to the screen of the merchant device IED2 and it is the merchant of device IED2 that checks visually, and if the merchant confirms the user in front of him (in example in a physical point of sale use-case) he matches with the image corresponding with the encoding data of the user in front of him (ED1-1 of a past successful transaction or with EDl-n the original image encoding data when end-user of device IED1 opened his account), if merchant confirms manually by clicking on his device IED2 then through communication channel CCH2 it notifies the computer program product in SI and the transaction can proceed, otherwise the transaction is denied and communicated by SI computer program product to both user devices IED1 and IED2 respectively through communication channel CCH2.

[0058] According to a fifth aspect of the invention, there is provided a computer- implemented method of generating and storing a new modified output, the method using a system including at least one or more internet enabled devices (e.g. wireless devices, smartphones, portable computer, desktop computer), each internet enabled device including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executing on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, the at least one server including a respective second non-transitory storage medium (NTSM2), and a respective second computer program product embodied on the respective second non-transitory storage medium, the respective second computer program product executing on the at least one server to communicate with a first internet enabled device of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device uses the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and transmits the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, wherein the respective second computer program product executes on the at least one server to process the extracted digital image data, wherein the respective second computer program product executes on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response stores the new digital image data in the respective second non-transitory storage medium of the at least one server and indexes the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device, and wherein;

[0059] (i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product executing to analyse image data for a respective different ethnic group to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and

[0060] (ii) wherein each resulting output “encoding data 1 to n” of the n outputs is used as input to a respective fourth computer program product in the at least one server which executes on the at least one server to compare all the n outputs and to ignore all such outputs that are identical and to calculate an interpolation between outputs that are different to each other to generate the new modified output, referred to as an “auto modified encoding data” (AMED), and to store such “auto modified encoding data” in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device, and

[0061] (iii) wherein, in the event of only one output of the n outputs being different and all other outputs of the n outputs being the same, then no interpolation is performed in (ii), and the one different output of the n outputs becomes the “auto modified encoding data” (AMED) which is stored in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device. Advantages include improved antifraud protection, and improved security, and improved computer system security.

[0062] According to a sixth aspect of the invention, there is provided a computer- implemented method of generating and storing a new modified output, the method using a system including at least one or more internet enabled devices (e.g. wireless devices, smartphones, portable computer, desktop computer), each internet enabled device including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executing on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, the at least one server including a respective second non-transitory storage medium (NTSM2), and a respective second computer program product embodied on the respective second non-transitory storage medium, the respective second computer program product executing on the at least one server to communicate with a first internet enabled device of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device uses the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, wherein the respective second computer program product executes on the at least one server to process the extracted digital image data, wherein the respective second computer program product executes on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response stores the new digital image data in the respective second non-transitory storage medium of the at least one server and indexes the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device, and wherein;

[0063] (i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product analysing image data for respective different parts of a head of a person to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and

[0064] (ii) wherein each resulting output “encoding data 1 to n” of the n outputs is used as input to a respective fourth computer program product in the at least one server which executes on the at least one server to compare all the n outputs and to ignore all such outputs that are inconclusive (e.g. in which a match is not produced) and to generate the new modified output, referred to as an “auto modified encoding data” (AMED), and to store such “auto modified encoding data” in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device, and

[0065] (iii) wherein, in the event of all outputs being inconclusive (e.g. in which in all a match is not produced) then the new digital image data becomes the “auto modified encoding data” (AMED) which is stored in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device. Advantages include improved antifraud protection, and improved security, and improved computer system security.

[0066] According to a seventh aspect of the invention, there is provided a computer- implemented method of generating and storing a new modified output, the method using a system including at least one or more internet enabled devices (e.g. wireless devices, smartphones, portable computer, desktop computer), each internet enabled device including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executing on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, the at least one server including a respective second non-transitory storage medium (NTSM2), and a respective second computer program product embodied on the respective second non-transitory storage medium, the respective second computer program product executing on the at least one server to communicate with a first internet enabled device of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device uses the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, wherein the respective second computer program product is executing on the at least one server to process the extracted digital image data, wherein the respective second computer program product is executing on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response to store the new digital image data in the respective second non-transitory storage medium of the at least one server and to index the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device, and wherein;

[0067] (i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product analysing a respective image of a different respective viewing angle of a head when viewed from the respective built-in camera, to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and

[0068] (ii) wherein each resulting output “encoding data 1 to n” of the n outputs is used as input to a respective fourth computer program product in the at least one server which is executing on the at least one server to compare all the n outputs and to ignore all such outputs that are inconclusive (e.g. in which a match is not produced) and to calculate an interpolation between the conclusive (e.g. in which a match is produced) outputs that are different to each other, to generate the new modified output, referred to as an “auto modified encoding data” (AMED), and to store such “auto modified encoding data” in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device, and

[0069] (iii) wherein, in the event of only one output being conclusive (e.g. in which a match is produced) and all other outputs are inconclusive (e.g. in which a match is not produced), then no interpolation is performed and the one conclusive (e.g. in which a match is produced) output becomes the “auto modified encoding data” (AMED) which is stored in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device. Advantages include improved antifraud protection, and improved security, and improved computer system security.

[0070] According to an eighth aspect of the invention, there is provided a computer- implemented method of verifying a transaction, the method using a system including at least two or more internet enabled devices (e.g. wireless devices, smartphones, portable / desktop computer), each internet enabled device including a respective built- in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executing on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, wherein each respective first computer program product communicates with a second computer program product stored in a second non-transitory storage medium (NTSM2) of the at least one server, the second computer program product executing on the at least one server to communicate with a first internet enabled device (IED1) of the at least one or more internet enabled devices, and to communicate with a second internet enabled device (IED2) of the at least one or more internet enabled devices, and wherein the respective first computer program product executes on the first internet enabled device to use the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, and wherein the respective first computer program product executes on the second internet enabled device to use the respective built-in camera of the second internet enabled device to extract digital image data from the respective built-in camera of the second internet enabled device, and to transmit the extracted digital image data through a communication channel between the second internet enabled device and the at least one server, wherein the respective second computer program product executes on the at least one server to process the extracted digital image data, wherein the respective second computer program product executes on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response stores the new digital image data in the respective second non-transitory storage medium of the at least one server and indexes the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device; wherein the respective second computer program product executes on the at least one server to detect new digital image data being received from the second internet enabled device of the at least one or more internet enabled devices, and in response stores the new digital image data in the respective second non-transitory storage medium of the at least one server and indexes the new digital image data such that the new digital image data is associated with an originating user account of the second internet enabled device, and wherein:

[0071] (i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product analysing image data to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and

[0072] (ii) wherein each such output of the n outputs is used as input to a fourth computer program product in the at least one server which executes on the at least one server to compare the n outputs with corresponding historical past n outputs of the user account of the first internet enabled device and / or of the user account of the second internet enabled device, stored in the second non-transitory storage medium, indexed to the corresponding user account (in example, compares EDI with ED1-1 to EDl-n, where EDI is the current new input data and ED1-1 is the previous historical input data and EDl-n is most past historical input data when that user account was opened, AND compares ED2 with ED2-1 to ED2-n, where ED2 is the current new input data and ED2-1 is the previous historical input data and ED2-n is most past historical input data when that user account was opened), AND

[0073] (iii) wherein, in the event of a mismatch (meaning the match is below a predefined threshold “x”) between the n outputs with the corresponding historical past n outputs of the user account of the first internet enabled device OR a mismatch between the n outputs with the corresponding historical past n outputs of the user account of the second internet enabled device, then the transaction of the user account corresponding to the mismatch is denied by the system (e.g. if EDI shows a mismatch it means it is not the same person as the authorised person of the user account of device IED1 and this user account will be denied the transaction that it was intending to do with device IED2, if ED2 shows a mismatch it means it is not the same person as the authorised person of the user account of device IED2 and this user account will be denied the transaction that it was intending to do with device IED1), AND (iv) wherein, in the event there is a match above a predefined threshold “y” between the n outputs with the corresponding historical past n outputs of the user account of the first internet enabled device AND a match between the n outputs with the corresponding historical past n outputs of the user account of the second internet enabled device, then the transaction between the users of the first internet enabled device and the second internet enabled device is allowed to proceed by the system, allowing the first internet enabled device and the second internet enabled device to instruct any such transaction between their corresponding users, respectively through the respective communications channels (CCH1 and CCH2) for execution by a fifth computer program product of the at least one server, wherein the users of the accounts corresponding to the first internet enabled device and to the second internet enabled device could be two end-users, two merchants, an end-user and a merchant, etc. and wherein the transaction requested for execution could be a bank transfer, an online payment, a physical point of sale payment, a government benefits request, an insurance claim, a mortgage application, a loan application, etc. Advantages include improved antifraud protection, and improved security, and improved computer system security.

[0074] A method of any aspect of the fifth, sixth, seventh or eighth aspects of the invention may include using a system of any aspect of the first, second, third or fourth aspects of the invention.

[0075] The present invention is designed to solve real issues in people's lives, such as (i) enhance the security of people's digital assets to protect them from fraudulent activities by illegal acts or scams by other people, (ii) protect user accounts across the full spectrum of sectors where users store items of electronic value, such as e-money (legacy bank accounts and central bank issued currency “CBDC” accounts), point of sale payments platform accounts, online payments, health services accounts, national health accounts, government issued benefits accounts, cryptocurrencies portfolio, insurance portfolio, equity / ETF portfolio, etc, (iii) improve compliance levels with current and future governments or financial institutions' KYC / KYB (know your customer / know your business), AML (anti-money laundering) and ATF (antiterrorism financing) regulations, whilst reducing the financial and mental health impact exposure of potential fraud to users of a given system / platform or potential fraud from an account of a given system to a user of a different system / platform, (iv) confirm the identity of the person performing a transaction (be it a login, execute a given service from a given account as a health or benefits service that costs money to an institution / government, etc) or authenticate an electronic transaction online, e.g. digital banking payment transaction, or any other online platform such as point of sale / online payment or online products & services sales platforms.

[0076] In summary, examples of the present invention are designed to solve the above in addition to providing solutions to problems faced by companies & institutions offering services in the digital world, such as the prevention of (i) fraud with high risk to illicit money movements, (ii) identity theft in processes with personal authorization, (iii) fraudulent use of digital accounts, etc.

[0077] Examples of the present invention are designed to overcome the shortcomings of the prior art and to provide a fully or semi-automated way of resolving the shortcomings of the prior art specifically in the prevention and detection of potential identity fraud in any securitised or non-securitised Internet process. A method and system, in one example, are based on the manual correction of facial recognition data, converting that corrected encoding data into fully automated form thereafter for that user. In another example it is based on the confirmation of the user and / or merchant manually, or fully automated, as the case may require a certain high or even higher security protection level. In yet another example, of many more variants, the encoding data confirmed or corrected is auto fed to the corresponding artificial intelligence algorithm corresponding to that user ethnicity facial features and used for all future transactions, thus allowing for further auto improvements in the system’s and method’s accuracy in reducing false positives. In a further example the system applies simultaneous different methods, wherein each method is adapted and trained specifically for each different ethnic features group; such methods would run in parallel and the system would apply an additional method that would select the outcome of the previous method with the highest statistical success rate (=lowest false positive rate) for each given system input, and such parallel methods would include a method that includes in the encoding data in addition to the data points also adding the colour of such data points allowing the method to add additional steps to check the statistical highest probability of automating the ethnic group method to be used, trained specifically for that ethnic group data encoding, AND an additional optional step to identify the colour of the eyes (pupil and / or iris) to reduce further any requirement for manual intervention if at all required and reduce even further the false positives and thus increase the reliability of the system over the prior art.

[0078] The devices included herein relating to examples of the invention apply to fixed- or wireless- devices, smartphones, tablets, laptops- or desktops- and any other different devices that have an integrated camera and can download an application software that includes the method(s) of examples of this invention or has it integrated by the manufacturer and is adapted to communicate with the cloud hardware and cloud application software of the system of examples of this invention. Aspects of the invention may be combined.

[0079] BRIEF DESCRIPTION OF THE FIGURES

[0080] Aspects of the invention will now be described, by way of example(s), with reference to the following Figures, in which:

[0081] Figure 1 represents a diagram relating to a typical example of the present invention. Figure 1 shows adapted wireless devices with a respective built-in camera (IED1 to lEDn), the cloud (Internet), and Server with computer program (SI to Sn) stored in the respective non-transitory memory (NTSM2), wherein each wireless device with a respective built-in camera (lEDn) is connected to the cloud, through a communication channel (CCH1 to CCHn). The Server Subprograms generates a new output (AMED). Figure 2 represents a typical flow-chart of an example of the prior art on how the digital data is processed and the encoding is created.

[0082] Figure 3 represents a flow-chart of an example of the present invention for obtaining the ENCODING of digital data, considering several ethnic groups.

[0083] Figure 4 represents a flow-chart of an example of the present invention for obtaining the ENCODING of digital data, considering several body parts.

[0084] Figure 5 represents a flow-chart of an example of the present invention for obtaining the ENCODING of digital data, considering several angles.

[0085] Figure 6 represents a flow-chart of an alternative example of the present invention for obtaining the ENCODING of digital data, considering several ethnic groups.

[0086] Figure 7 represents a flow-chart of an alternative example of the present invention for obtaining the ENCODING of digital data, considering several body parts.

[0087] Figure 8 represents a flow-chart of an alternative example of the present invention for obtaining the ENCODING of digital data, considering several angles.

[0088] Figure 9 represents a flow-chart of an alternative example of the present invention for obtaining the ENCODING of digital data, considering several ethnic groups.

[0089] Figure 10 represents a flow-chart of an alternative example of the present invention for obtaining the ENCODING of digital data, considering several body parts.

[0090] Figure 11 represents a flow-chart of an alternative example of the present invention for obtaining the ENCODING of digital data, considering several angles.

[0091] Figure 12 represents a diagram relating to an alternative example of the present invention. Figure 12 shows adapted wireless devices with a respective built-in camera (IED1 to lEDn), the cloud (Internet), and Server with computer program (SI to Sn) stored in the respective non-transitory memory (NTSM3), wherein each wireless device with a respective built-in camera (lEDn) is connected to the cloud, through a communication channel (CCH1 to CCHn). The Server Subprograms generates a new output (AMED_NEW_OUT).

[0092] Figure 13 represents a flow-chart of an example of the present invention to decide whether or not to approve the execution of a transaction, considering the historical data.

[0093] Figure 14 represents a flow-chart of an example of the present invention to decide whether or not to approve the execution of a transaction, considering the historical data and merchant’s validation.

[0094] Figure 15 represents a flow-chart of an example of the present invention to decide whether or not to approve the execution of a transaction, considering the historical data and merchant’s validation and two different thresholds.

[0095] Figure 16 represents a use-case of an example of the invention wherein a merchant and a user are involved in a transaction process.

[0096] Figure 17 depicts a use-case of an example of the invention on a physical payment.

[0097] Figure 18 showcase a flow-chart of the user case of an example of the invention system in an online payment.

[0098] Figure 19 depicts a use-case for an online payment, wherein the user goes into the payment process.

[0099] Figure 20 showcase a representation of the different parts of the head, wherein different areas are taken depending on the user representation.

[0100] DETAILED DESCRIPTION

[0101] In an example, the invention includes a system and method for operating an anti-fraud detection where verification of the identity of the person or persons involved in the transaction is mandatory to avoid certain fraud, money laundering or terrorist financing (ATF), processing aspects through the use of fixed- or wireless- devices or portable- or desktop- computers. Such wireless devices (e.g. smartphones, tablet computers or any such other mobile device) with at least one built in camera, Near Field Communication (NFC) technology or alternatively Bluetooth, and such computers with at least one built in camera, wherein such devices are adapted as per examples of this invention, and there is included as part of the system a local- or remote processing unit or server to process the data of the transaction and capturing or displaying the data of the user and / or the merchant for a given transaction request. Aspects of disclosures refer, in particular, to a system and methods that are able to obtain relevant information about the person who requests, approves, or executes a transaction (e.g. payment, refund, cash deposit or cash withdrawal), or any such other transaction of a bank, fintech, foreign exchange, brokerage system, payments platform or any such other that requires user’s authentication or identification.

[0102] In examples of this invention, contrary to the prior art, transactions cannot be executed without a verification of the identity of the person (e.g. user) or persons (e.g. user and / or merchant) involved in the transaction. The identification of such persons can be, depending on the use case, through the visual confirmation of the corresponding user by a merchant, or in a different case, automatically by means of the system or method itself, processing a “photograph” or "selfie" taken at the moment of the transaction request from the user and / or merchant.

[0103] Identity verification is considered of vital importance, as it is one of the major reasons why fraud is still prevailing, and this is precisely one of the shortcomings of the state of the art, where basic security methods apply, e.g., at user's login to the application or platform, a two-step authorization factor (2FA), entering a personalized user code sent by the platform to the user's application or his phone number by SMS, via e-mail or via a one-time password (OTP) code sent to the user’s device. In an example of our invention however, the identity of the user depends on the usecase, and consists or comprises of one, two or even three extra security steps. In a first use-case, where the user-initiated transaction does not involve any merchant (e.g., an outgoing bank transfer payment, online payment, etc), the user is obliged to take a photo of himself showing his face clearly within the outlined photo boundaries; this type of photo is commonly known as a 'selfie'. The selfie is processed by a neural network, which is trained to establish from the image the required recognition points, also known as “encoding”. These points represent the identification of the face as detected in an image and are stored in the system as the unique biometric identification file.

[0104] The selfie, taken by the user who wants to execute a transaction, is processed through the neural network generating the unique biometric identification (ID) of the mentioned selfie of that transaction and, in case of a successful transaction is stored together with the verified biometric identification dataset of the user in the database of the system. This ID is compared with the existing one in the system database of the corresponding user from the time he passed his KYC (that included a selfie) and the previous times of past successful transactions, then the transaction is executed as long as the data matches (meaning the unique biometric identification selfie at the time of the transaction matches the past KYC biometric identification selfie and / or the past successful transactions biometric identification selfies) and at this stage the validation processes are considered as passed, otherwise the transaction request is rejected. If rejected, then the new ID (biometric identification selfie at time of this transaction) is compared to all blacklist databases, that contains all past rejected transactions biometric identification selfies, to ensure that the person of this transaction is not already on the blacklist system, if the biometric information extracted from the picture taken from the device matches with above a certain set threshold with any biometric information of the blacklist databases then the rejected transaction biometric identification selfie is added to the blacklist database(s), the blacklist database includes all the selfies or photos and biometric identification of the potentially fraudulent transactions, on the other hand if the system does not match it with any biometric information of the blacklist database, the biometric information of the rejected transaction (e.g., the picture taken from the devices) and the transaction itself is pending to be reviewed by a support- or Compliance- personnel, the review of the transaction is another step of the method where the different teams review the “encoding” generated for the “selfie” of the transaction, to ensure visually that it is truly the same person or not. Also, the blacklist database functions as a firewall on the transaction system that uses an example of the invention allowing to configure different rules sets to automatically detect potential fraud on the transaction verification process without any manual revision. In certain case however, e.g. in the use-case of where the transaction is performed after a user login authentication into his account and thereafter the transaction is rejected for the same reason as described above, then both the rejected transaction biometric identification selfie and the KYC biometric identification selfie are put in the backlist database, and that logged in user account is blocked, which means (a) that user account can't be used anymore unless de-blocked manually by a Compliance officer and (b) both of that transaction selfie and that account KYC selfie will be used to automatically block any transaction type from that moment onwards by the system and methods of examples of this invention. One of the optional tasks to be performed manually by the compliance officer is to confirm, in addition to all the information collected from both the user and the person who has attempted to perform the transaction, that the generation of the encoding from the selfie has been correct. Using, for example, the platform described with respect to examples of this invention for this purpose, where the different facial points detected by the system can be visualized in a simple manner. With this simple task, if the biometric ID ("encoding") is correct, the efficiency of the system is ensured in the future comparisons with that data and if not, it is corrected and replaced by the defective one (saving it in the history of that ID), avoiding future comparison problems.

[0105] In a second use-case, where the transaction does involve a merchant, the system of an example of this invention offers a set of innovative methods to follow, which can be used together or independently, as long as at least one is performed. The set of methods includes a visual confirmation of the user's identity performed by the merchant, ensuring that the real identity of the paying user account (as per his KYC selfie or past successful transactions selfies) and the person, in front of the merchant, who wants to carry out the transaction is in fact the same person. For this purpose, the merchant is shown on his device, as described within an example of this invention, the image of the corresponding user account KYC selfie or a recent approved transaction selfie or both, which was stored in the system's database during the user's account registration process (KYC) and / or during previous successful transactions selfies. Once the image is displayed, the merchant confirms through his device, described in examples of this invention, that the person who wants to perform the transaction is the same person as the one displayed on that same device or not. In case this would be the only security method activated or used by the platform, the transaction would be approved for execution, only in case the merchant confirms that they are the same person, through the method described above. Otherwise, the transaction would be rejected.

[0106] Another identification method is that the merchant takes a photograph of the user face and this image (transaction user ID) is processed by the system and / or methods of examples of this invention, the transaction is executed as long as the system identifies the person who wants to perform the transaction as matching to be the corresponding registered user of the transaction user account, otherwise, the transaction is rejected and the transaction user biometric ID (photo made by merchant) is compared to an all blacklist databases, that contains all past rejected transactions biometric identification selfies, to ensure that the person of this transaction is not already on the blacklist system; if the biometric information extracted from the picture taken from the device matches with above a certain set threshold with any biometric information of the blacklist databases then the rejected transaction biometric identification selfie is added to the blacklist database(s), on the other hand if the system does not match it with any biometric information of the blacklist database, the biometric information of the rejected transaction (e.g., the picture taken from the devices) need to be manually reviewed by the support / compliance personnel to verify that the encoding generated from the picture is correct, in the case that is not correct, correcting the encoding. And if the system and / or methods detects that the identified rejected transaction user performed a user login into the register user account in the system as part of the transaction (meaning just prior to the transaction as a prior method before the selfies check method), in this case that user account associated to the rejected transaction will be blocked not allowing any outgoing transactions from that account, and can only be de-blocked manually by an authorized Compliance officer of the system.

[0107] The next method contemplated in examples of this invention is the use-case where the merchant identifies the transaction person's selfie on screen (regardless if the transaction person or the merchant took that transaction person's selfie or if the system herein provided the stored user account selfie of that transaction) as being the same present person of the transaction, confirming that the present user and the selfie shown to the merchant device are the same person but the system of examples of this invention detects that the transaction person's selfie is in fact not matching with the database transaction user account selfie (KYC selfie or last successful transactions selfies). This previous scenario, in the vast majority of cases, corresponds to potential fraud by the merchant itself, typically without the actual transaction account user's knowledge but possibly with the knowledge of the potential fraudulent transaction person's knowledge. This is a far more serious and bigger potential financial fraud impact when a merchant is in fact the fraudulent party or collaborating party in fraudulent transactions. That is why, before showing to the transaction person and the merchant that the transaction has been rejected, the merchant is requested to take a 'selfie1of himself to confirm such transaction as if it was a routine random simple confirmation of that transaction. This additional method step is added when the previous detection is triggered to obtain the biometrical identification (ID) of the merchant of that specific transaction, and then, put the merchant ID selfie in a merchant's blacklist database as well as in the user's blacklist database, and mark the transaction user ID used to carry out that same transaction, also as part of the blacklist. In this case where the merchant is unreliable or potentially the origin of fraudulent transaction(s), the system of examples of the invention will let the merchant and user finish that transaction visually on both their device as a normal transaction showing on screen “pending execution” even though the system knows from the outset that the transaction will in fact be later on reported as not executed = rejected, allowing to improve the blacklist databases of the system in examples of the invention.

[0108] In another use-case as an example of the invention, is where the merchant and user are involved in the transaction process, there is automated the previous explained use- case / method auto-triggering following different rules: if the merchant or user has made previous transactions through the system, the past decisions taken on the past transactions are used as a first step of this method by the system for the current new transaction. Also, the past action of past transaction(s) is used as the next step of method used by the transaction firewall, where an identified transaction by user and merchant is executed but the firewall system rejected such similar or same type of transaction in the past, trigger an additional step in the method forcing the merchant and / or user to be identified or not in the current and / or next transact! on(s). E.g.: in a past or current transaction where the merchant or user was identified, but the firewall detected through for example the Global Positioning System (GPS) position or country of the device detection of one or both parties that they are not in the same country or city at the moment of the transaction, the method will reject such current transaction, but for the next transactions, involving that same rejected transaction merchant and / or user the method will force the identification of both parties in the current transaction. However, legitimate users or merchants where the method or system of examples of this invention has not had any potential fraud detected in the past “n” transactions, the method herein will skip further steps and allow execution of the current transaction without any further identification checks of both parties (merchant and user) of the current transaction. The system in this example of the present invention automates the different identification steps of the method, for user or merchant identification requirement steps depending on the historical transactions results of the user and / or merchant.

[0109] Amongst others, one of the positive aspects of examples of the present invention, is that the system and methods are continuously learning and improving the potential fraud detection success rate as the number of transactions with biometric identification data increases over time (in fact a self-learning method), that is why the blacklist database(s) with the help of the processed biometric identification data (ID) and a method with steps that makes the system improve its detection secernate (meaning be able to distinguish, recognize as being different) after every transaction through the system of examples of the invention herein. On the other hand the system and methods herein function as a transaction firewall where it automatically reduces the manual review process of the system and methods herein as the transactions increase thus increasing the auto detected potential fraudulent transactions and thus reducing the manual intervention accordingly and moreover by adding a method step allowing manually by a system operator the insertion of certain rules adopted for the automation of certain of the detection rules at any time.

[0110] Other aspects to improve the avoidance of fraud in transactions, added to examples of the invention's methods, is a transaction firewall system, which is performed as a security hedging at the end of the method as an additional step, once the participants of the transaction have been identified; this system consists or comprises of a proprietary algorithm based on a neural network decision, where the transactions are characterized with all the information associated with it, including, among other data, the GPS position or country of the devices (if available), with which the transaction is attempted, IPs of the same, data collected from the user and / or the merchant devices, transaction history, etc.

[0111] With all this data, the artificial intelligence network, proprietary algorithm, is trained to evaluate the value and weight of each of the different data provided in order to make the decision with the highest potential fraud detection success rate. Each detected case of potential or confirmed fraud is used to train and improve the accuracy of the artificial intelligence network, proprietary algorithm, in the next decisionmaking method for all transactions after the current transaction on the system.

[0112] Other aspects of the disclosure are the techniques and methods used for improving and upgrading the facial recognition system (artificial intelligence network, proprietary algorithm), as per examples of this invention, constantly improving the level of potential fraud detection efficiency / success rate as the number of transactions and corresponding available user and / or merchant data increases over time. One of the main objective of examples of this invention is to identify potential fraud, AML or ATF in as many use-cases as possible before it occurs, thus preventing it in the first instance from occurring, by comparing the biometric identification of the person who is performing a given transaction and / or the person performing the merchant function of the same transaction, with that of the corresponding user and / or merchant account authorized KYC person(s) and with the blacklist database(s), generated over time with the available historical biometric identification data.

[0113] Biometric identification is obtained from a photo or series of consecutive photos taken at the time of a given transaction and processed by a so called deep neural network, which is trained by a proprietary method to establish all the, pre-established or adjusted over time, recognition points necessary to identify a static face in the event of a single photo taken or a dynamic live face in the event of several consecutive taken photos.

[0114] The shortcomings of the state of the art are several, such as for example, but not limited to, that these prior art neural networks used to obtain the points that identify the face, although they have a superlative data processing capacity, obtaining, in general, highly effective, and efficient results, all have been trained however with an algorithm that results in a large deviation of the data points. That is, the data on which the neural network's consistency training was performed was not balanced with respect to one parameter in particular, namely the ethnic group of the person to be identified. For example, if the neural network tries to obtain the facial points from a photograph or selfie of a person of African descent, the system detects the face and may recognize parts of it, but, as it has learned from prior art algorithms, it will identify the facial features common among the population of Caucasian descent, features very different from those of the identified person. Consequently, the result obtained by the prior art neural network is not sufficiently reliable for transactional anti-fraud measures since the points obtained do not correctly identify the person, however could be sufficient for non-financial transactions such as certain accounts login for certain applications / use-cases where combating money fraud is not a prevailing objective.

[0115] The main problem with the inaccuracy of the facial points is that at no time are these cases identified as an error, the prior art neural network(s) provide the resulting points, and these are taken as good, causing all comparisons and calculations made with such data from that moment on to be unreliable for use in secure transactions, such as but not limited to financial transactions. The only way to improve the efficiency of the system when using existing prior art is to retrain the neural network with the data obtained through another additional neural network specialized in obtaining this type of data. This method, although the accuracy of this network is better in other ethnic groups, it does not ensure an improvement in the efficiency of the system for use in security- or financial transactions to combat fraud, money laundering or terrorist financing, since at no time in the prior art process does it confirm that the resulting one or multiple concurrent neural networks resulting facial points coincide with the actual facial features of the corresponding person.

[0116] In a different example of the present invention, our system solves the previous shortcomings as follows:

[0117] During the registration method of new user's account creation onto the system platform, a selfie or consecutive selfies of the person being registered is required, in order to store, together with the data defining the user, his or her image and the corresponding biometric identification, in particular it stores also examples of this invention’s proprietary neural network's resulting facial points that coincide with the actual facial features of the corresponding person's selfie or consecutive selfies. This identification is compared with the multiple biometric identification, specifically including comparing examples of this invention’s proprietary neural network's resulting facial points that coincide with the actual facial features of the corresponding person's account selfie or consecutive selfies already existing in the system (server) of examples of this invention, and in case the proprietary algorithm of an example of this invention detects similarity with any of them in any other pre-existing accounts or in any of the blacklist databases, the registration is set to pending for review by a Compliance officer or is rejected automatically, and after account creation the same is applied to transactions.

[0118] The objective of the match detection by a system is not sufficient to have only a correct or not result, because it is not enough to confirm or deny the match, since the similarity is calculated by a system from two biometric IDs. Therefore, if it is confirmed that they are different identities, it is because in the prior art generation of one or both biometric IDs was or could have been not accurately enough calculated for the reasons explained above, in contrast to the system and methods of examples of the present invention.

[0119] To solve this prior art shortcoming, re-generating a new ID is not useful, since the neural network will recalculate the facial points in the same position. This is why, in an example of this invention, a visualization method allows for an authorised system operator or Compliance person to modify manually the facial points that make up the biometric identification of a given user of a given ethnical origin.

[0120] This method as part of the system of examples of this invention displays, on the selfie of the person, the different facial points previously calculated by the system's neural network, differentiating each set of points with the associated colours of that facial point, numbers of distance from each given point from a reference point (two dimensional as x,y or three dimensional adding depth as x,y,z or adding as c,x,y,z) and / or geometric shapes depending on the different facial features they represent (e.g., eyes, eyebrows, mouth, lips, nose, chin, forehead etc).

[0121] The visualization of the different dots (with the associated data described above) on the image of the face allows an operator or compliance person to check if the biometric identification of the user is correct or if there is a deviation in any set of dots. If so, this tool allows in one example the user and in another example an authorised system person to modify the position of the points that are not correctly positioned, for example, by simply clicking on the point and dragging it to its final position, at which moment when releasing the point in its new position the method of an example of this invention has an additional setup and auto recalculates the coordinates of the new point 2D or 3D position as well as all associated data with that new point (colour, geometrical shapes etc).

[0122] Once the dots are correctly positioned, the method, after recalculating the new position data has an additional step to auto save the new position data of all the points and shapes, replacing the inaccurate biometric ID with the corrected one, fit for combating fraud, money laundering, terrorist financing in financial transactions or any such other security transactions that require such high accuracy as per examples of this invention. This previous method of this example of the invention allows the system of this example of the invention to achieve the required efficiency, given that, with higher accuracy biometric identifications of this example of the invention, comparisons of photos / selfies can be carried out with greater precision, ensuring that the identity of the actual person is statistically more likely the one the system and methods herein works with, obtaining more accurate results in accordance with the resemblance of the users.

[0123] Once the system of an example of the invention is fed with enough modified encoding, as described in above methods, as a method part of the main neural network of an example of this invention, in charge of the biometric identification, sufficiently accurate biometric data can be retrained with the newly manually corrected encoding, improving the accuracy of the results obtained from the network by training the neural network with the manually corrected encoding resulting in less manual intervention required as more manual data is fed into its training programming method of the neural network of an example of this invention, that resides within the proprietary server of an example of this invention. Thus, defining an action pipeline for retraining, the neural network, eliminates the system's learning peak, given that it will be a continuous improvement.

[0124] In a further example of the present invention a method for improving the neural network in charge of generating the biometric identification of users is described, as being a deep neural network capable of detecting the recurrent modifications made during the manual corrections of the stored encodings and auto incorporate those as being the original entry for each such photo or selfie data set and deleting the originally first calculated for that same photo, selfie or short video. That is, a neural network with the method(s) of an example of this invention, independent of the neural network that was responsible for generating the first encodings of the manually modified encoding, and such independent neural network is trained with both the old original first encodings and their manually corrected versions and / or their corrected versions only, specializing the network in the detection of defective encodings by means of a proprietary method step of correlation between the old original first encodings and their manually corrected versions of any given photo or selfie, and how to correct them.

[0125] In another example of the present invention, the manually corrected points data, is replaced by an additional method step using the correlation result between the old original first encodings and their manually corrected versions to determine ethnic origin / group and use that data to auto detect ethnic origin / group and automatically switch to a method dedicated to a neural network pre-programmed to be most efficient for that specific ethnic origin / group.

[0126] These last methods and system of examples of this invention automate the process of encoding correction, and consequently, of the different biometric identifications of the system and methods, automating, after a first batch of data, and improving the accuracy of the main neural network in charge of the generation of biometric identifications, as well as reducing the need of the correction of the identifications as the input data increases over time.

[0127] In another example of the present invention, a system includes at least one or more internet enabled devices (wireless devices, smartphones, portable / desktop computer), with a built-in camera and at least one or more internet enabled server 1, wherein;

[0128] The first internet enabled device (IED1) includes a first non-transitory storage medium, and a first computer program product embodied on the first non-transitory storage medium, the first computer program product executable on the first internet enabled device when executed communicates with the server, and with a second non- transitory storage medium, and a second computer program product in the second non-transitory storage medium, the second computer program product executable on the server when executed communicates with the first internet enabled device 1, and wherein the first computer program product when executed on IED1 uses a built-in camera to extract the “digital data from the camera” of device 1, and transmit such digital data from the camera through the communication channel between device 1 and server 1, and is processed by a server computer program product and where in the event the server computer program product detects new data being received from device 1 to device n, such new input data is stored in the server non-transitory storage medium and indexed such that each such new input data is associated to the originating user account of device 1 to device n of the system, and wherein;

[0129] (i) a third computer program product in server adapted to include “n” subprogram products, each sub-program product adapted to “n” different ethnic groups to process new data in parallel for each new input data, resulting in “n” outputs referred to as “encoding data 1 to n”, and

[0130] (ii) wherein each such output “encoding data 1 to n” is used as input to a fourth computer program product in server 1 to n which compares all the “encoding data 1 to n” and ignores all such outputs that are identical and calculates an interpolation between those outputs that are different to each other and generates a new modified output, referred to as a “auto modified encoding data”, and stores such “auto modified encoding data” in the second non-transitory storage medium indexed to the corresponding user account of device 1 to device n of the system, and

[0131] (iii) wherein, in the event of only one output being different and all other outputs are the same, then no interpolation is performed and the one different output becomes the “auto modified encoding data” to be stored in the second non-transitory storage medium indexed to the corresponding user account of device 1 to device n of the system.

[0132] In a further example of the present invention, a system includes at least one or more internet enabled devices (e.g. wireless devices, smartphones, portable / desktop computer), with a built-in camera and at least one or more internet enabled server, wherein;

[0133] The first internet enabled device 1 includes a first non-transitory storage medium, and a first computer program product embodied on the first non-transitory storage medium, the first computer program product executable on the device 1 when executed communicates with the server, and with a second non-transitory storage medium, and a second computer program product in the second non-transitory storage medium, the second computer program product executable on server when executed communicates with device 1, and wherein the first computer program product when executed on device 1 uses a built-in camera to extract the “digital data from the camera” of device 1, and transmit such data through the communication channel between the device and the server, and is processed by the server computer program product and where in the event the server computer program product detects new data being received from device 1 to device n, such new input data is stored in the server non-transitory storage medium and indexed such that each such new input data is associated to the originating user account of device 1 to device n of the system, and wherein;

[0134] (i) a third computer program product in server adapted to include “n” subprogram products, each sub-program product adapted to “n” different parts of the head of a person to process new data in parallel for each new input data, resulting in “n” outputs referred to as “encoding data 1 to n”, and

[0135] (ii) wherein each such output “encoding data 1 to n” is used as input to a fourth computer program product in server which compares all the “encoding data 1 to n” and ignores all such outputs that are inconclusive (e.g. in which a match is not produced) and generates a new modified output, referred to as an “auto modified encoding data”, and stores such “auto modified encoding data” in the second non- transitory storage medium indexed to the corresponding user account of device 1 to device n of the system, and

[0136] (iii) wherein, in the event of all outputs being inconclusive (e.g. in which in all a match is not produced), then the input data becomes the “auto modified encoding data” to be stored in the second non-transitory storage medium indexed to the corresponding user account of device 1 to device n of the system.

[0137] In a different example of the present invention, a system includes at least one or more internet enabled devices (wireless devices, smartphones, portable / desktop computer), with a built-in camera and at least one or more internet enabled server, wherein;

[0138] The first internet enabled device 1 (IED1) includes a first non-transitory storage medium, and a first computer program product embodied on the first non-transitory storage medium, the first computer program product executable on the device 1 when executed communicates with the server, and with a second non-transitory storage medium, and a second computer program product in the second non-transitory storage medium, the second computer program product executable on server when executed communicates with device 1, and wherein the first computer program product when executed on IED1 uses a built-in camera to extract the “digital data from the camera” of device 1, and transmits such data through the communication channel between the device and the server, and is processed by the server computer program product and where in the event the server computer program product detects new data being received from device 1 to device n, such new input data is stored in the server non- transitory storage medium and indexed such that each such new input data is associated to the originating user account of device 1 to device n of the system, and wherein;

[0139] (i) a third computer program product in the server adapted to include “n” subprogram products, each sub-program product adapted to “n” different angles between the camera and the head of a person to process new data in parallel for each new input data, resulting in “n” outputs referred to as “encoding data 1 to n”, and

[0140] (ii) wherein each such output “encoding data 1 to n” is used as input to a fourth computer program product in the server which compares all the “encoding data 1 to n” and ignores all such outputs that are inconclusive (e.g. in which a match is not produced) and calculates an interpolation between those conclusive (e.g. in which a match is produced) outputs that are different to each other and generates a new modified output, referred to as an “auto modified encoding data”, and stores such “auto modified encoding data” in the second non-transitory storage medium indexed to the corresponding user account of device 1 to device n of the system, and

[0141] (iii) wherein, in the event of only one output being conclusive (e.g. in which a match is produced) and all other outputs are inconclusive (e.g. in which a match is not produced), then no interpolation is performed and the one conclusive (e.g. in which a match is produced) output becomes the “auto modified encoding data” to be stored in the second non-transitory storage medium indexed to the corresponding user account of device 1 to device n of the system.

[0142] In a different example of the present invention, a system is defined wherein the “digital data from the camera” of device 1, consists or comprises of a single image, consecutive images or a short video and extracts images from the video from the “face or head of a person” of from “an object” and converts such data into a two- dimensional digital matrix array x, y where “x, y” are the coordinates of each array in the matrix versus a reference point and becomes the new data, replacing the original data, and transmits the new data through the communication channel between the device and the server for processing. In a different example of the present invention, a system is defined wherein the “digital data from the camera” of device 1, consists or comprises of a single image, consecutive images or a short video and extracts images from the video from the “face or head of a person” of from “an object” and converts such data into a three- dimensional digital matrix array (x, y, d where “x, y” are the coordinates of each array in the matrix and the 3rddimension “d” corresponds to the distance from the camera to each such array in the matrix) and becomes the new data, replacing the original data, and transmits the new data through the communication channel between device 1 and the server for processing.

[0143] In a further example of the present invention, a system is defined wherein the “digital data from the camera” of device 1, consists or comprises of a single image, consecutive images or a short video and extracts images from the video from the “face or head of a person” of from “an object” and converts such data into a three- dimensional digital matrix array (x, y, c where “x, y” are the coordinates of each array in the matrix and the 3rddimension “c” corresponds to the colour code of each such array in the matrix) and becomes the new data, replacing the original data, and transmits the new data through the communication channel between the device 1 and the server for processing.

[0144] In a different example of the present invention, a system is defined wherein the “digital data from the camera” of device 1, consists or comprises of a single image, consecutive images or a short video and extracts images from the video from the “face or head of a person” of from “an object” and converts such data into a fourdimensional digital matrix array (x, y, c, d where “x, y” are the coordinates of each array in the matrix and the 3rddimension “c” corresponds to the colour code of each such array in the matrix and the 4thdimension “d” corresponds to the distance from the camera to each such array in the matrix) and becomes the new data, replacing the original data, and transmits the new data through the communication channel between the device 1 and the server for processing.

[0145] In a further example of the present invention, a system is defined wherein the “sub- program products”, each sub-program product adapted to “n” different ethnic groups, are “pretrained neural network program products”, wherein each such pretrained neural network program product is pretrained for a specific ethnic or sub-ethnic-group of an ethnic group (e.g. Hispanic-Spain, Hispanic descent-Mexico, Hispanic descent- Peru, etc., African descent-North America, African descent-United Kingdom, African descent-Nigeria, African descent-Kenya, African descent South- Africa, etc., Asian descent-China, Asian descent-Japan, Asian descent-Indonesia, Asian descent- Singapore, Caucasian descent-North America, Caucasian descent-United Kingdom, Caucasian descent-Germany, Caucasian descent-South Africa etc.) generating each a mathematical representation of the face of a person resulting in “n” outputs referred to as “encoding data 1 to n”.

[0146] In a different example of the present invention, a system is defined wherein the “subprogram products”, each sub-program product adapted to “n” different parts of the head of a person, are “pretrained neural network program products”, wherein each such pretrained neural network program product is pretrained for a specific area or sub-area of a head of a person or a face of a person (e.g. area enclosing both eyes, or the sub-area of each enclosure of each eye separately, the area above the eyes / forehead or sub-area of the eye-brows separately from the area above the eyebrows, the area below the eyes but above the lips or the sub-areas of the nose separately to the sub-area of each cheek, the area of the nose to include the nostril and bridge of the nose, the area mouth or the sub-area of the upper and lower lip separately to the area left and right of the lip endings towards each lower cheek area, the area of the chin, the area of the neck, the area of each ear or sub-area of each earlobe separately) generating in combination of all the available data (some may not be available) a mathematical representation of the face or head of a person resulting in “n” outputs referred to as “encoding data 1 to n”.

[0147] In a different example of the present invention, a system is defined wherein the “subprogram products”, each sub-program product adapted to “n” different angles between the camera and the head of a person, are “pretrained neural network program products, wherein each such pretrained neural network program product is pretrained for a specific angle between a reference on the face of a person and the camera, wherein; in a two-dimensional data (x,y) such reference on the face to calculate the angle to the camera is the position of iris or centre point of the pupil versus the background sclera of the eye, where the angle is defined as;

[0148] -90 degrees LEFT when the “iris or centre point of the pupil” (IRCPP) is on the far left of the sclera of the eye, +0 degrees RIGHT / LEFT when the “iris or centre point of the pupil” (IRCPP) is in the horizontal centre of the sclera of the eye, +90 degrees RIGHT when the “iris or centre point of the pupil” (IRCPP) is on the far right of the sclera of the eye, and in between, the degrees are a linear interpolation of the mathematical calculation between the previous, and

[0149] -90 degrees DOWN when the “iris or centre point of the pupil” (IRCPP) is on the lower end of the sclera of the eye, +0 degrees UP / DOWN when the “iris or centre point of the pupil” (IRCPP) is in the vertical centre of the sclera of the eye, +90 degrees UP when the “iris or centre point of the pupil” (IRCPP) is on the upper end of the sclera of the eye, and in between, the degrees are a linear interpolation of the mathematical calculation between the previous, and resulting in a two-dimensional horizontal angle (left / right) and a two-dimensional vertical angle (up / down), which in combination of all the available data (one may not be available) form a mathematical representation of the angle(s) resulting in “n” outputs referred to as “encoding data 1 to n”.

[0150] In a different example of the present invention, a system is defined wherein the “subprogram products”, each sub-program product adapted to “n” different angles between the BICI and the head of a person, are “pretrained neural network program products”, wherein each such pretrained neural network program product is pretrained for a specific angle between a reference on the face of a person and the camera BICI, wherein; in a three-dimensional data (x,y, z) such reference on the face to calculate the angle to the camera is the distance “z” from the camera to each those points x, y, where the angle is defined as; the linear interpolation between the delta of the distance “zl” of a reference point on the upper half of the face and the distance “z2” of a mirror reference point on the lower half of the face as the UP / DOWN angle, and the linear interpolation between the delta of the distance “z3” of a reference point on the left half of the face and the distance “z4” of a mirror reference point on the right half of the face as the LEFT / RIGHT angle, and resulting in three dimensional horizontal and vertical angle representation, which form a mathematical representation of the angle(s) resulting in “n” outputs referred to as “encoding data 1 to n”.

[0151] In a different example of the present invention, a system includes at least two or more internet enabled devices (e.g. wireless devices, smartphones, portable / desktop computer), each with a respective built-in camera, and at least one or more internet enabled server, wherein;

[0152] The first internet enabled device 1 includes a first non-transitory storage medium, and a first computer program product embodied on the first non-transitory storage medium, the first computer program product executable on the device 1 when executed communicates with the server, and the second internet enabled device, a second non-transitory storage medium, and a second computer program product embodied on the second non-transitory storage medium, the second computer program product executable on the device 2 when executed communicates with the server, and both the first- and the second- computer program product communicate with a third non-transitory storage medium, and a third computer program product in the third non-transitory storage medium, the third computer program product executable on SI when executed communicates with device 1 and device 2, and wherein, the first computer program product when executed on device 1 uses a built-in camera to extract the “digital data from the camera” of the device 1, and transmit such data through the communication channel between the device 1 and the server, and the second computer program product when executed on device 2 uses a built-in camera to extract the “digital data from the camera” of device 2, and transmit such data through the communication channel between the device 2 and the server, and wherein data from device 1 and data from device 2 are processed by the server computer program product and where in the event the server computer program product detects new data being received from device 1 or device 2, such new input data is stored in the server non-transitory storage medium and indexed such that each such new input data is associated to the originating user account of device 1 or device 2 of the system, and wherein;

[0153] (i) the fourth computer program product in server is adapted to include “n” subprogram products, each sub-program product adapted to process new data in parallel for each new input data, resulting in “n” outputs referred to as “encoding data 1, 2 to n”, and

[0154] (ii) wherein each such output “encoding data 1, 2 to n” is used as input to a fifth computer program product in the server which compares the new received “encoding data” with the corresponding historical past “encoding data” of the device 1 or device 2 user account stored in the server non-transitory storage medium indexed to the corresponding user account, AND

[0155] (iii) wherein, in the event of a mismatch (meaning the match is below a predefined threshold “x”) between encoding data and historical encoding data OR a mismatch between encoding data and historical encoding data then the transaction of the user account corresponding to the mismatch will be denied by the system (if encoding data shows a mismatch it means it is not the same person as the authorised person of the user account of device 1 and this user account will be denied the transaction that it was intending to do with device 2, if encoding data 2 shows a mismatch it means it is not the same person as the authorised person of the user account of device 2 and this user account will be denied the transaction that it was intending to do with device 1), AND

[0156] (iv) wherein, in the event there is a match above a predefined threshold “y” between encoding data 1 and historical encoding data 1 AND between encoding data 2 and historical encoding data 2 then the transaction between the user of device 1 and 2 is allowed to proceed by the system, allowing devices 1 and 2 to instruct any such transaction between their corresponding user, respectively through the communications channels 1 and 2 for execution by the computer program product of the server. Wherein the user of the account corresponding to devices 1 and 2 could be two end-users, two merchants, an end-user and a merchant, etc. and wherein the transaction requested for execution could be a bank transfer, an online payment, a physical point of sale payment, a government benefits request, an insurance claim, a mortgage application, a loan application, etc. In a different example of the present invention, a system is defined wherein only the check of a match or mismatch is performed on the encoding data corresponding to an end-user of a device 1, comparing encoding data 1 with historical encoding data and in the event of a match, the system will allow the transaction request to proceed between device 1 and device 2 through the server.

[0157] In a different example of the present invention, a system is defined wherein the system does not do any checks but instead sends the encoding data of the end-users of the device 1 to the screen of the merchant device 2 and it is the merchant of device 2 that checks visually, and if the merchant confirms the user in front of him (in example in a physical point of sale use-case) he matches with the image corresponding with the encoding data of the user in front of him (historical encoding data a past successful transaction or with historical encoding data the original image encoding data when end-user of device 1 opened his account), if merchant confirms manually by clicking on his device 2 then through communication channel it notifies the computer program product in the server and the transaction can proceed, otherwise the transaction is denied and communicated by the server computer program product to both user devices 1 and 2 respectively through communication channel.

[0158] In a different example of the present invention, a system is defined wherein the check of a match or mismatch is performed on the encoding data corresponding to an enduser of a device 1, comparing encoding data 1 with historical encoding data 1 and wherein; i. in the event there is a match above a predefined threshold “y” the system will allow the transaction request to proceed between device 1 and device 2 through the server. ii. in the event of a mismatch (meaning the match is below a predefined threshold “x”) between encoding data 1 and historical data 1 the system will deny the transaction request. iii. in the event there is a match above a predefined threshold “x” BUT below a predefined threshold “y” the system will send the encoding data of the end-users of the device 1 to the screen of the merchant device 2 and it is the merchant of device 2 that checks visually, and if the merchant confirms the user in front of him (in example in a physical point of sale use-case) he matches with the image corresponding with the encoding data of the user in front of him, if merchant confirms manually by clicking on his device 2 then through communication channel it notifies the computer program product in the server and the transaction can proceed, otherwise the transaction is denied and communicated by the server computer program product to both user devices 1 and 2 respectively through communication channel.

[0159] Examples of this invention apply protective measure methods in the execution of any transfer on the system, among other things, with methods to identify the actual natural person who is executing a transaction or initiate it on the system. The steps to be carried out by the method(s) are decided automatically by the method depending to the different use-cases in which the steps of method(s) are follow a next step or a different next step, depending, among other things, on the configuration of the system in which the method(s) is used in, wherein the method(s) take into account the historical data stored by the system itself are the triggers for the method(s) to auto select consecutive steps depending on such historical truncation(s) data and / or current transaction data.

[0160] The different security methods described below can be implemented independently, in combination or all together:

[0161] 1. The user who wants to carry out a new current transaction takes a selfie with his device wireless device / smartphone or portable device / PC or other such device) and the system and / or methods automatically compares the identity of the person involved in the current transaction (with the “encoding data” generated from the selfie or consecutive selfies or short video) against the “encoding data” of the actual owner of the account from which the current transaction is to be executed from. Only if there is a match above a threshold selected by the method (such threshold may be different depending on the historical data accessible by the method from the system database), the transaction is approved for execution; on the other hand, if there is no match (below the method auto selected threshold depending on historical data in the system database) such current transaction is Rejected. The method then in a next step, of such rejected transaction, if the user involved in the current transaction had logged into the account successfully as a prior step of the method before initiating the transaction, then both the user of the current transaction “encoding data” obtained from the new selfie / consecutive selfies / short video and the owner of that account “encoding data” are stored in a blacklist pending review not allowing any transactions from such account. If, however the rejected transaction did not involve a login at all as a prior step of the method of the current transaction, in example a Card payment at a merchant or an online card payment, then the next method step is the following: put the “encoding data” of the user that attempted that transaction in Blacklist and alert the actual real user owner of such account and optional in a further step of the method block any outgoing transactions of such account related to that rejected transaction.

[0162] 2. The transaction, as an additional step of the method, cannot be executed without a visual confirmation from the merchant confirming (by a click of a key on the merchant’s device in the event of a point of sale) that the person (user) involved in the current transaction is the same as the person registered in the system. For this purpose, the merchant’s device, method in an additional step, will display the image stored in the system of the actual account user of that transaction (in example owner of the paying Card account or owner of the account if paying account to account), meaning the owner of the account from which the transaction is to be executed, either the image taken during the onboarding process or from one of the previous successful transactions. The transaction will not be approved by the method for execution until the merchant confirms that the person in front of him and the one shown on the merchant's device is the same, otherwise the transaction will be immediately rejected and such user image “encoding data” in a follow in step blacklisted and not allow to do any transaction through the system of examples of this invention.

[0163] 3. The method wherein an additional step, it's the merchant who takes a photo of the person who wants to make the transaction and is uploaded into the system and the corresponding “encoding data” is generated. This last “encoding data” is in a next step of method automatically compared with the selfie “encoding data” of the person of that associated account registered in the system and is either a match if above the required threshold (such threshold decided by the method based on historical data if available otherwise a default pre-set threshold is used) or not. If they match, then the method approves the current transaction for execution, otherwise the method in a next step automatically rejects such current transaction. If rejected then in a further step of the method the current transaction user photo “encoding data” is put in the “blacklist” database of the system, prohibiting any future transaction in the system by the “meta data” user. If approved for execution, then in a further step the method, the “encoding data” of the user photo / short video taken by the merchant of the person who wanted to perform the transaction is compared with the existing blacklisted “encoding data” of users of previous rejected transactions or detected or confirmed potential fraud cases. In case the “encoding data” of the person of the current transaction already exists on this blacklist, then in a further step of the method, the new transaction data is added to the historical data of the matched user profile and the user account of that transaction.

[0164] If the person / user of the current rejected transaction does not match any profile on the blacklist (no “encoding data” match and no historical rejected transactions correlation), then in a further step of method the “encoding data” of the user of the current transaction is added to a blacklist.

[0165] If, however the rejected transaction did not involve a login at all as a prior step of the method of the current transaction, in example a Card payment at a merchant or an online card payment, then the next method step is the following: put the “encoding data” of the user that attempted that transaction in Blacklist and alert the actual real user owner of such account and optional in a further step of the method block any outgoing transactions of such account related to that rejected transaction.

[0166] If however the rejected transaction did involve in a prior step of the method of the current transaction a user account login which was passed successfully and the method detects a mismatch between the transaction user “encoding data” and that of the actual account user owner “encoding data”, in example when paying account to account, then both the current transaction user “encoding data” and the account owner “encoding data” are put in blacklist and in a further step of the method the account of this rejected transaction is blocked not allowing any outgoing transactions and / or reverse login access.

[0167] 4. The method wherein an additional step, a person responsible for transactions compliance of a platform where this system herein is implemented reviews the blacklist pending review, in case he confirms that the facial points have been calculated correctly, such data, whether from an already registered user, a person unknown to the system database who has impersonated a lawful user and attempted to perform a transaction or a merchant person involved in the transaction, is marked as a final blacklist. If, on the other hand, the generation of the embedding “encoding data” is not correct, the compliance officer corrects it on his screen by moving the point to represent the real facial representation and in a further step the method recalculates the “encoding data” and stores it in the corresponding database of the system, marking such “manually confirmed meta data” which the method in any future transaction involving such “manually confirmed corrected encoding data” in the method steps does not go through the manual review step again, as it becomes an input of the method as historical data for future transactions involving such user or merchant “encoding data” accompanied in the database with the “manually confirmed corrected encoding data”. The method in an optional further step after generating the “manually confirmed corrected encoding data” repeats the previous step matching process that has been performed in the first instance, that had resulted in a blacklist, if it still does not match then this “manually confirmed corrected encoding data” is added to the data of the corresponding user of the transaction or in the user account or merchant involved in the current transaction.

[0168] 5. In the particular use-case method of point 2, where the merchant confirms the identity of the user and approves the execution of the transaction, combined with the method of point 3, where the method / system does not find a match in the compared “encoding data”, then in a further step of the method in the system, automatically applies the further steps of the method of point 4, to ensure that the potential fraud does not come from an improper / illicit act by the person using the merchant device.

[0169] 6. In a another method variant of previous points 2 and / or 3, in a further step of the method wherein the user involved in an online payment transaction (card payment or account to account payment) or an online bank transfer transaction AND the merchant or bank is automated with no person involved at that side of the transaction through the system, in this case the method step requires the user to confirm to the system to allow to take a photo or short video of the user of the transaction (through allowing, for example access to his smartphone or PC camera) and in a further step convert that photo or short video into “encoding data”. In a further step the “encoding data” will be compared with the historical “encoding data” of the account associated to that current transaction and then follow the corresponding further steps as any of the corresponding methods and steps already described in the previous points above.

[0170] 7. In a different method of an example of the present invention the additional steps are as follows: in an initial step the method herein uses the image data (be it a selfie, consecutive selfies or one or multiple images extracted for a short video) as input into multiple parallel proprietary artificial intelligence engines (wherein each of those artificial intelligence engines are adapted to specifically operate and trained for each different ethnic group or sub-group). In a further step the method herein then checks the output “encoding data” of each of those adapted artificial intelligence engines and discards those outputs that provide a non-conclusive (e.g. in which a match is not produced) or standard “encoding data” output and uses only that output that is conclusive (e.g. in which a match is produced) or non-standard. In the event more than one of the parallel adapted artificial intelligence engines provides a conclusive (e.g. in which a match is produced) or non-standard “encoding data” output, then in a further step, the method herein will create an “auto modified encoding data” as the mathematical interpolation between the different conclusive (e.g. in which a match is produced) or non-standard outputs “encoding data” (be they in three-dimensional x,y,z or fourdimensional c,x,y,z).

[0171] The vast majority of these use-cases are related to mitigating against groups or individuals who aim to carry out fraudulent transactions or illicit activities, by resolving the mentioned shortcomings of the prior art. Consequently, the system and methods, described here in examples of this invention, rejects the above described potentially fraudulent or illicit transactions and stores the “auto modified encoding data” and / or if any “manually confirmed corrected encoding data” of the person(s) involved in such rejected or blocked transactions, be they an unknown or known user or merchant and stores those in a blacklist database as historical data to prevent those individuals from performing any future transactions on the system or block any such accounts on the system as suspected or confirmed attempt in compliance with antifraud, anti-money laundering and anti -terrorist financing rules and regulations.

[0172] In yet another different example of the present invention the addition of a different additional method to the system, there is a method that will act as a transactions firewall. In this use-case, the method will use as decision inputs the system weights generated to obtain a characterisation of the transaction with different priority decisions based on the higher the weight assigned the higher the risk and when certain threshold is reached the method in a further step will trigger a transaction rejection.

[0173] In the event that the transaction is not approved due to potential fraud risk as a consequence of the weight level of certain parameters, the data associated to such transaction is sent, as a further step of the method to the system neural decision network, to be used to train and optimise the characterisation of the system neural network and consequently, improve the parameters weighting assignment of certain parameters used by the method for future transactions.

[0174] In a different example, an example of this invention will display on any adapted device (e.g. wireless device, smartphone or portable / desktop PC), the different facial points that compose the encoding obtained during the facial recognition process. Said points will be represented on the corresponding image, in which case if the generation of the encoding data has been carried out correctly then no action is required, but if not correct in some or all the points, it would worsen the reliability of the system and methods. Once the validity of the stored encoding data has been confirmed or corrected and then confirmed, the automated checks of the future verifications of that user will be much more reliable, eliminating the possibility of false positives or erroneous detections. In a different example, a contingency method is applied for cases where the generation of the encodings, obtained during user facial recognition, are corrupted or false facial features have been detected. This means that the identification of the user is faulty, therefore, any automated actions of the system and methods will lose effectiveness and the security of the system is severely affected.

[0175] This previous mentioned issue is much more common in the prior art than is commonly believed, due to the large deviation of the data during the training of the artificial intelligence networks used for the generation of encodings, causing, for example, but not exclusively, that the characteristic features of ethnic groups such as African, Asian, Hispanic or other descendants are not well recognized, due to the lower amount of data in the western world of this ethnicity used on the training data. Even if such data of these or other ethnic groups would have been available, training an artificial intelligence algorithm would be optimal for one and not for another ethnic group, thus the reliability of the prior art would improve marginally and would still require the system and method(s) herein to obtain a reasonably higher reliability to be effective in use cases for security transactions or financial transactions where the absolute figures of fraud, money laundering or anti -terrorist financing are crucial to be reduced in a reliable manner.

[0176] The contingency method for this case is a method with an additional step allowing for manual intervention before initiating a next step, to directly edit the different facial points referred to in the encoding that alters the originally detected “encoding data” and results in an “auto modified encoding data” or, if any, “manually confirmed corrected encoding data”, with the aforementioned system and / or method(s) herein, visually displaying and correcting these points, modifying those that have deviated from their correct position, and correcting those different facial points into their correct position, regenerating the encoding data from the encoding.

[0177] Therefore, with the prior art described herein, all prior art shortcomings listed herein have been overcome with the examples of the present invention. Many modifications and variations or different examples of this present invention are possible in view of the disclosures herein of examples of this invention text, figures, drawings, flow-charts, explanations and claims. It is to be understood that, within the scope of the appended claims, the invention can be practiced other than as specifically described in the claims of this invention and new claims can be extracted as new claims or as a divisional patent. The invention which is intended to be protected should not, however, be construed as limited to the particular forms disclosed in the claims, or implementation examples outlined, as these are to be regarded as illustrative rather than restrictive. Variations and changes could be made by those skilled in the art without deviating from the novelty of the invention. Accordingly, the detailed descriptions and figures of this invention should be considered exemplary in nature and not limited to the novelties of the invention as set forth in the claims.

[0178] In a different example of the present invention a method, wherein the additional steps are as follows: instead of a single artificial intelligence engine that generates the “encoding data” to instead execute the step of the image data (be it a selfie, consecutive selfies or one or multiple images extracted for a short video) being fed to multiple parallel proprietary artificial intelligence engines (wherein each of those artificial intelligence engines are adapted to specifically operate for each pre-defined and pre-programmed ethnic group or sub-group). In a further step the method then checks the “encoding data” of each of those adapted artificial intelligence engines and discards those outputs that provide a non-conclusive (e.g. in which a match is not produced) or standard output “encoding data” and uses only that output data that is conclusive (e.g. in which a match is produced) or non-standard. In the event more than one of the parallel adapted artificial intelligence engines provides a conclusive (e.g. in which a match is produced) or non-standard “encoding data” output, then in a further step, the method will create an “auto modified encoding data” as the mathematical interpolation between the different conclusive (e.g. in which a match is produced) or non-standard outputs “encoding data”. This method resolves a big shortcoming of the prior art, in that all inconclusive (e.g. in which a match is not produced) or standard outputs (some or most also referred to as “false positives”) of the prior art's artificial intelligence “encoding data” have to be reviewed manually by an authorised person of the system wherein the prior art is used, whereas in an example of this invention, by using the novel method in an adapted system the “false positives” are dramatically reduced to a negligible number, if any at all.

[0179] Our findings confirm that there is a correlation between an illicit transaction and a person making different transactions on different accounts, where he is not authorized and correlates with one or more of the following: fraud, money laundering or terrorist financing.

[0180] Specifically, Figure 1 represents a diagram relating to a typical example of the present invention. The system of Figure 1 comprises:

[0181] One or more internet enabled devices (e.g. wireless devices, smartphones, portable / desktop computer) lEDn, with a respective first built-in camera (BICI), and respective first non-transitory storage medium (NTSM1), with subprogram product stored on NTSM1 of each lEDn that extract the digital data from camera (DDC1); each lEDn is connected to the cloud (Internet) through a communication channel (CCHn). One or more internet enabled Server (Sn) with first Computer program product (Sn_l), an optional second computer program product (Sn_2) that has ‘n’ sub program products (Sn_2 product n), and an optional third computer program product (Sn_3) stored in a non-transitory memory (NTSM2_Sn), which are connected to the Cloud (Internet) through a communication channel (CCH Sn).

[0182] Specifically, Figure 2 depicts a flow-chart illustrating a prior art example. It showcases the method employed by different prior art systems to create a model encoding from a digital data.

[0183] Specifically, Figure 3 depicts a flow-chart illustrating an example of the present invention. The device extracts the digital data (DD), which is processed by a computer program, and then sent to different ethnic sub-program products for further processing. Each of the subprograms receive the data, and generates independently an encoding Data (EDO). Once the EDO is generated by each of the subprograms, the system checks if the EDO is a default generated EDO, so the system ignores it, considering it invalid. In case there is more than one valid EDO, all the valid result are interpolated to create a new one (EDO INTER); otherwise, if there is only one EDO, this single EDO (EDO NEW) is the final result.

[0184] Specifically, Figure 4 shows a flow-chart illustrating an example of the present invention. The device extracts the digital data (DD), after which the DD is divided into different head parts by the program and sent to a subprogram product for each head part. Each subprogram product encodes the head part into a new encoded data output (EDO), all encoded data outputs are summed to a final result, to generate a reliable encoding data output.

[0185] Specifically, Figure 5 depicts a flow-chart illustrating an example of the present invention. The device extracts digital data (DD) from the camera. The computer program processes the DD from the camera and divides it into one or more angles, and each of the different angles is sent to each product subprogram, which is responsible for generating an encoding data output for that angle. Once the EDOs are generated from the sub-programs, they are compared if they are encoded and generated by default, in this case ignoring the EDO, considering it invalid. When the system has a valid EDO(s) to consider, then if there is more than one EDO an interpolation of all the results is performed, creating a new one (EDO INTER), otherwise the only valid EDO (EDO NEW) is used, and the final EDO will be the final output (EDO NEW-OUT).

[0186] Specifically, Figure 6 depicts a flow chart illustrating an example of the present invention, wherein it depicts a functional block diagram of the flow chart of the method of comparing the n results obtained from the processing of the digital data and the final encoding based on the above comparison results. The device extracts digital data (DD) from the camera, which is processed by a computer program, obtaining a matrix of multiple areas with respect to a reference point (NDDC). The NDDC is processed by each subprogram, generating a new encoding data output (EDO), ignoring all EDOs generated by default, considering it invalid.

[0187] In case there is more than one EDO considered valid, an interpolation is performed creating a new EDO (EDO INTER), on the other hand, if there is only a single EDO in the considered valid EDO(s) then this EDO will be the final EDO output (EDO NEW).

[0188] Specifically, Figure 7 depicts a flow-chart illustrating an example of the present invention. The device extracts digital data (DD) from camera, a computer program product processes DD from camera and gets an array of multiple areas with a reference point (NDDC), the NDDC is divided into different parts of the head and sent to a subprogram product for every part of the head, responsible to generate an encoding data output (EDO), once the EDO(s) are generated, the valid EDO(s) are added to the final result, resulting in a reliable encoding data output.

[0189] Specifically, Figure 8 depicts a flow diagram illustrating an example of the present invention. The device extracts digital data (DD), a software product processes the DD from the camera and obtains a multi-area matrix with a reference point (NDDC), the NDDC is divided into different parts of the head and sent to a sub-program product for each part of the head, responsible for generating an encoding data output (EDO), once the EDOs are generated, the valid EDOs are added to the final result, resulting in a reliable encoding data output.

[0190] Specifically, Figure 9 depicts a flow-chart illustrating an example of the present invention. The device extracts the digital data (DD), which is processed by a computer program, and then sent to different pretrained neural network program products (PTNNPP) for different ethnic group, for further processing. Each of the PTNNPP receive the data, and generates independently an encoding Data (EDO). Once the EDO is generated by each of the subprograms, the system checks if the EDO is a default generated EDO, so the system ignores it, considering it invalid. In case there is more than one valid EDO, all the valid result are interpolated to create a new one (EDO INTER); otherwise, if there is only one EDO, this single EDO (EDO NEW) is the final result.

[0191] Specifically, Figure 10 shows a flow-chart illustrating an example of the present invention. The device extracts the digital data (DD), after which the DD is divided into different head parts by the program and sent to a pretrained neural network program product for each head part. Each subprogram product encodes the head part into a new encoded data output (EDO), all encoded data outputs are summed to a final result, to generate a reliable encoding data output.

[0192] Specifically, Figure 11 depicts a flow-chart illustrating an example of the present invention. The device extracts digital data (DD) from the camera. The computer program processes the DD and divides it into one or more angles, and each of the different angles is sent to each pretrained neural network program product (PTNNPP), which is responsible for generating an encoding data output for that angle. Once the EDOs are generated from the sub-programs, they are compared if they are encoded and generated by default, in this case ignoring the EDO, considering it invalid. When the system has a valid EDO(s) to consider, then if there is more than one EDO an interpolation of all the results is performed, creating a new one (EDO INTER), otherwise the only valid EDO (EDO NEW) is used, and the final EDO will be the final output (EDO NEW-OUT).

[0193] Specifically, Figure 12 represents a diagram relating to an example of the present invention. The system of Figure 12 comprises:

[0194] One or more internet enabled devices (e.g. wireless devices, smartphones, portable / desktop computer) lEDn, with a respective first built-in camera (BICI), and first non-transitory storage medium (NTSM1), with subprogram product stored on NT SMI of each lEDn that extract the digital data from camera (DDC1) each lEDn is connected to the cloud (Internet) through a communication channel (CCHn).

[0195] One or more internet enabled Server (Sn) with first Computer program product (Sn_l), an optional second computer program product (Sn_2) that has ‘n’ sub program products (Sn_2 product n), and an optional third computer program product (Sn_3) stored in a non-transitory memory (NTSM2_Sn), which are connected to the Cloud (Internet) through a communication channel (CCH Sn).

[0196] Specifically, Figure 13 depicts a flow-chart illustrating an example of the present invention. The digital data (DD) is extracted from at least one device and sent to a program containing a subprogram for each of the different data extracted, where it is processed and as many EDs are obtained as there are digital data (DDn).

[0197] These are compared with the historical EDs stored for that user, obtaining a numerical value, representing the similarity between the different data.

[0198] If the RESULT is below the threshold, the system detects the result as a mismatch and denies the transaction.

[0199] If the RESULT is above the threshold, the system detects the result as a match and approves the transaction. In this case, the ED is stored in the history of that user, for future comparison.

[0200] Figure 14 depicts a flow-chart illustrating an example of the present invention.

[0201] The digital data (DD) is extracted from at least one device and sent to a program containing a subprogram for each of the different data extracted, where it is processed and as many EDs are obtained as there are digital data (DDn).

[0202] Once at least one ED is obtained, the method makes the decision whether to approve the transaction to be executed or not.

[0203] One option, which is not exclusive, is to follow the process described in Figure 13. Another option, not exclusive, the user's ED history is sent to the merchant’s device, who visually checks that they are the same person.

[0204] If the merchant’s response is positive, the transaction is approved, provided that:

[0205] The option of automatic comparison of the ED with the PED has not been performed and the transaction has been denied.

[0206] In any case where the transaction is approved, the ED obtained at the beginning of the process is stored in the user's PED.

[0207] Specifically Figure 15 depicts a flow-chart illustrating an example of the present invention.

[0208] The digital data (DD) is extracted from at least one device and sent to a program containing a subprogram for each of the different data extracted, where it is processed and as many EDs are obtained as there are digital data (DDn).

[0209] Once at least one ED is obtained, the method makes the decision whether to approve the transaction to be executed or not.

[0210] One option, which is not exclusive, the ED is compared with the historical EDs stored for that user, obtaining a numerical value, representing the similarity between the different data.

[0211] - If the RESULT is below the threshold X, the system detects the result as a mismatch and denies the transaction.

[0212] - If the RESULT is above the threshold X and below the threshold Y, the user’s ED history is sent to the merchant's device, who visually checks that they are the same person.

[0213] If the merchant's response is positive, the transaction is approved, provided that:

[0214] ■ The option of automatic comparison of the ED with the PED has not been performed and the transaction has been denied.

[0215] ■ In any case where the transaction is approved, the ED obtained at the beginning of the process is stored in the user's PED.

[0216] - If the RESULT is above the threshold Y, the system detects the result as a match and approves the transaction. In this case, the ED is stored in the history of that user, for future comparison.

[0217] Another option, not exclusive, the user's ED history is sent to the merchant's device, who visually checks that they are the same person. If the merchant's response is positive, the transaction is approved, provided that:

[0218] The option of automatic comparison of the ED with the PED has not been performed and the transaction has been denied.

[0219] In any case where the transaction is approved, the ED obtained at the beginning of the process is stored in the user's PED.

[0220] Specifically Figure 16, represents a use-case of an example of the invention wherein a merchant and a user are involved in a transaction process.

[0221] The merchant must introduce the amount of the transaction on the merchant’s device and the user authenticate on the systems, the transaction begins once the user interacts with the merchant’s device. So, a PED (Historical Encoding Data of The User) Is sent to the merchant device, and the merchant is the responsible to confirm that the PED of the user is the same as the present user. If the merchant does not confirm that the user is the same, the transaction is denied and finished. By other hand, in the case that the merchant confirms that they are the same person. The transaction is sent to the server and the server must validate it.

[0222] Specifically Figure 17, depicts a use-case of an example of the invention on a physical payment, wherein, we have two different devices, the merchant device and the user’s device. In a first case the merchant Introduces the amount in the devices, then the Use authenticated, pays with the device. The transaction is sent to the server, and the server retrieves to the merchant’s device a picture of the user, the merchant must accept or rejects, once the merchant makes one of the actions, the server processes and validates or declines the transaction.

[0223] Specifically Figure 18, showcase a flow-chart of the user case of an example of the invention system in an online payment case once the user starts the transaction and selects the desired payment method. A selfie is required, so the user’s device extracts the digital data (DD) from camera and generates an encoding data (ED), then the ED is compared with the Historical (past) Encoding Data of users accounts (PED) and if the comparison exceed a Z threshold meaning that the user’s picture and the ED matches, the transaction is sent to payment server to be validated and the ED is sent to the PED for future transactions, by other hand if the comparison results in a mismatch, the transactions is denied by the system. Specifically Figure 19, depicts a use-case for an online payment, wherein the user goes into the payment process and selects the desired payment method, then the user must confirm the transaction, and a selfie is required. Once the user device’s camera takes a selfie from the user, sends it to the server and the server compares with the user’s digital data previously registered in the server, once the servers processed the pictures, then the server processes and validates or declines the transaction, and return to the User’s device a confirmation of the user transaction.

[0224] Specifically Figure 20, showcase a representation of the different parts of the head, wherein different areas are taken depending on the user representation.

[0225] A system and a method of operating an anti-fraud, AML and ATF system, through the use of adapted wireless internet devices, using a “selfie” or a picture of the head / face of a user. The system of an example of the invention for operating in an anti-fraud, AML and ATF systems uses set of different methods to generate a unique correct / optimise Encoding Data of the given picture;

[0226] The set of applied methods are in charge of automating the optimisation of the generated Encoding, improving the authentication system for certain anti-fraud, antimoney laundering (AML) or anti -terrorist financing (ATF) aspects through the use of fixed or wireless devices or portable computers or desktop computers, with at least one built in camera and application software in the device as per examples of this invention, such devices adapted as per examples of this invention, whereas the system server is adapted to include a local processing unit and / or a remote server processing unit adapted as per examples of this invention.

[0227] According to the state-of-the-art methods of the system of an example of the invention, there is provided improvement in the final generation of the picture encoded data, grouping in different image characteristics, such as ethnicity, the angle of the photo, or the different areas composing the face for the processing thereof.

[0228] In addition, examples of the invention increase the security of the system it applies to the execution of transactions by offering an alternative way in which the merchant confirms the user's identity through his device. CONCEPTS

[0229] 1. A system including at least one or more internet enabled devices (wireless devices, smartphones, portable / desktop computer), with a built-in camera and at least one or more internet enabled server (SI to Sn), wherein;

[0230] The first internet enabled device (IED1), a first non-transitory storage medium (NTSM1), and a first computer program product embodied on the first non-transitory storage medium, the first computer program product executable on the IED1 when executed communicates with the server (SI), and with a second non-transitory storage medium (NTSM2), and a second computer program product in the second non- transitory storage medium, the second computer program product executable on SI when executed communicates with IED1, and wherein the first computer program product when executed on IED1 uses a built-in camera (BICI) to extract the “digital data from the camera” (DDC1) of IEDl,and transmit such (DDCl)data through the communication channel (CCH1) between IED1 and SI, and is processed by SI computer program product and where in the event SI computer program product detects new data being received from IED1 to lEDn, such new input data is stored in the SI non-transitory storage medium and indexed such that each such new input data is associated to the originating user account of IED1 to lEDn of the system, and wherein;

[0231] (i) a third computer program product in SI to Sn adapted to include “n” subprogram products, each sub-program product adapted to “n” different ethnic groups to process new data (DDC1 to n) in parallel for each new input data, resulting in “n” outputs referred to as “encoding data 1 to n” (EDI to n), and

[0232] (ii) wherein each such output “encoding data 1 to n” is used as input to a fourth computer program product in SI to Sn which compares all the “encoding data 1 to n” and ignores all such outputs that are identical and calculates an interpolation between those outputs that are different to each other and generates a new modified output, referred to a “auto modified encoding data” (AMED), and stores such “auto modified encoding data” in the second non-transitory storage medium indexed to the corresponding user account of IED1 to lEDn of the system, and

[0233] (iii) wherein, in the event of only one output being different and all other outputs are the same, then no interpolation is performed and the one different output becomes the “auto modified encoding data” (AMED) to be stored in the second non-transitory storage medium indexed to the corresponding user account of IED1 to lEDn of the system.

[0234] 2. A system including at least one or more internet enabled devices (wireless devices, smartphones, portable / desktop computer), with a built-in camera and at least one or more internet enabled server (SI to Sn), wherein;

[0235] The first internet enabled device (IED1), a first non-transitory storage medium (NTSM1), and a first computer program product embodied on the first non-transitory storage medium, the first computer program product executable on the IED1 when executed communicates with the server (SI), and with a second non-transitory storage medium (NTSM2), and a second computer program product in the second non- transitory storage medium, the second computer program product executable on SI when executed communicates with IED1, and wherein the first computer program product when executed on IED1 uses a built-in camera (BICI) to extract the “digital data from the camera” (DDC1) of IEDl,and transmit such (DDCl)data through the communication channel (CCH1) between IED1 and SI, and is processed by SI computer program product and where in the event SI computer program product detects new data being received from IED1 to lEDn, such new input data is stored in the SI non-transitory storage medium and indexed such that each such new input data is associated to the originating user account of IED1 to lEDn of the system, and wherein;

[0236] (i) a third computer program product in SI to Sn adapted to include “n” subprogram products, each sub-program product adapted to “n” different parts of the head of a person to process new data (DDC1 to n) in parallel for each new input data, resulting in “n” outputs referred to as “encoding data 1 to n” (EDI to n), and

[0237] (ii) wherein each such output “encoding data 1 to n” is used as input to a fourth computer program product in SI to Sn which compares all the “encoding data 1 to n” and ignores all such outputs that are inconclusive (e.g. in which a match is not produced) and generates a new modified output, referred to an “auto modified encoding data” (AMED), and stores such “auto modified encoding data” in the second non-transitory storage medium indexed to the corresponding user account of IED1 to lEDn of the system, and (iii) wherein, in the event of all outputs being inconclusive (e.g. in which in all a match is not produced), then the input data (DDC1) becomes the “auto modified encoding data” (AMED) to be stored in the second non-transitory storage medium indexed to the corresponding user account of IED1 to lEDn of the system.

[0238] 3. A system including at least one or more internet enabled devices (wireless devices, smartphones, portable / desktop computer), with a built-in camera and at least one or more internet enabled server (SI to Sn), wherein;

[0239] The first internet enabled device (IED1), a first non-transitory storage medium (NTSM1), and a first computer program product embodied on the first non-transitory storage medium, the first computer program product executable on the IED1 when executed communicates with the server (SI), and with a second non-transitory storage medium (NTSM2), and a second computer program product in the second non- transitory storage medium, the second computer program product executable on SI when executed communicates with IED1, and wherein the first computer program product when executed on IED1 uses a built-in camera (BICI) to extract the “digital data from the camera” (DDC1) of IED1, and transmit such (DDCl)data through the communication channel (CCH1) between IED1 and SI, and is processed by SI computer program product and where in the event SI computer program product detects new data being received from IED1 to lEDn, such new input data is stored in the SI non-transitory storage medium and indexed such that each such new input data is associated to the originating user account of IED1 to lEDn of the system, and wherein;

[0240] (i) a third computer program product in SI to Sn adapted to include “n” subprogram products, each sub-program product adapted to “n” different angles between the BICI and the head of a person to process new data (DDC1 to n) in parallel for each new input data, resulting in “n” outputs referred to as “encoding data 1 to n” (EDI to n), and

[0241] (ii) wherein each such output “encoding data 1 to n” is used as input to a fourth computer program product in SI to Sn which compares all the “encoding data 1 to n” and ignores all such outputs that are inconclusive (e.g. in which a match is not produced) and calculates an interpolation between those conclusive (e.g. in which a match is produced) outputs that are different to each other and generates a new modified output, referred to a “auto modified encoding data” (AMED), and stores such “auto modified encoding data” in the second non-transitory storage medium indexed to the corresponding user account of IED1 to lEDn of the system, and

[0242] (iii) wherein, in the event of only one output being conclusive (e.g. in which a match is produced) and all other outputs are inconclusive (e.g. in which a match is not produced), then no interpolation is performed and the one conclusive (e.g. in which a match is produced) output becomes the “auto modified encoding data” (AMED) to be stored in the second non-transitory storage medium indexed to the corresponding user account of IED1 to lEDn of the system.

[0243] 4. A system of Concept 1 and / or Concept 2 and / or Concept 3 wherein the “digital data from the camera” (DDC1) of IED1, consists or comprises of a single image, consecutive images or a short video and extracts images from the video from the “face or head of a person” of from “an object” and converts such data into a 2-dimensional digital matrix array x, y where “x, y” are the coordinates of each array in the matrix versus a reference point and becomes the new data (NDDC1), replacing the original DDC1, and transmits the NDDC1 through the communication channel (CCH1) between IED1 and SI for processing.

[0244] 5. A system of Concept 1 and / or Concept 2 and / or Concept 3 wherein the “digital data from the camera” (DDC1) of IED1, consists or comprises of a single image, consecutive images or a short video and extracts images from the video from the “face or head of a person” of from “an object” and converts such data into a 3 -dimensional digital matrix array (x, y, d where “x, y” are the coordinates of each array in the matrix and the 3rddimension “d” corresponds to the distance from BICI to each such array in the matrix) and becomes the new data (NDDC1), replacing the original DDC1, and transmits the NDDC1 through the communication channel (CCH1) between IED1 and SI for processing.

[0245] 6. A system of Concept 1 and / or Concept 2 and / or Concept 3 wherein the “digital data from the camera” (DDC1) of IED1, consists or comprises of a single image, consecutive images or a short video and extracts images from the video from the “face or head of a person” of from “an object” and converts such data into a 3 -dimensional digital matrix array (x, y, c where “x, y” are the coordinates of each array in the matrix and the 3rddimension “c” corresponds to the colour code of each such array in the matrix) and becomes the new data (NDDC1), replacing the original DDC1, and transmits the NDDC1 through the communication channel (CCH1) between IED1 and SI for processing.

[0246] 7. A system of Concept 1 and / or Concept 2 and / or Concept 3 wherein the “digital data from the camera” (DDC1) of IED1, consists or comprises of a single image, consecutive images or a short video and extracts images from the video from the “face or head of a person” of from “an object” and converts such data into a 4-dimensional digital matrix array (x, y, c, d where “x, y” are the coordinates of each array in the matrix and the 3rddimension “c” corresponds to the colour code of each such array in the matrix and the 4thdimension “d” corresponds to the distance from BICI to each such array in the matrix) and becomes the new data (NDDC1), replacing the original DDC1, and transmits the NDDC1 through the communication channel (CCH1) between IED1 and SI for processing.

[0247] 8. A system of Concept 1 wherein the “sub-program products”, each sub-program product adapted to “n” different ethnic groups, are “pretrained neural network program products” (PTNNPP1 to n), wherein each such PTNNPP is pretrained for a specific ethnic or sub-ethnic-group of an ethnic group (Hispanic-Spain, Hispanic descent-Mexico, Hispanic descent-Peru, etc., African descent-North America, African descent-United Kingdom, African descent-Nigeria, African descent-Kenya, African descent South-Africa, etc., Asian descent-China, Asian descent-Japan, Asian descent- Indonesia, Asian descent- Singapore, Caucasian descent-North America, Caucasian descent-United Kingdom, Caucasian descent-Germany, Caucasian descent-South Africa etc.) generating each a mathematical representation of the face of a person resulting in “n” outputs referred to as “encoding data 1 to n” (EDI to n)in Concept 1.

[0248] 9. A system of Concept 2 wherein the “sub-program products”, each sub-program product adapted to “n” different parts of the head of a person, are “pretrained neural network program products” (PTNNPP 1 to n), wherein each such PTNNPP is pretrained for a specific area or sub-area of a head of a person or a face of a person (area enclosing both eyes, or the sub-area of each enclosure of each eye separately, the are above the eyes / forehead or sub-area of the eye-brows separately from the area above the eye-brows, the area below the eyes but above the lips or the sub-areas of the nose separately to the sub-area of each cheek, the area of the nose to include the nostril and bridge of the nose, the area mouth or the sub-area of the upper and lower lip separately to the area left and right of the lip endings towards each lower cheek area, the area of the chin, the area of the neck, the area of each ear or sub-area of each earlobe separately) generating in combination of all the available data (some may not be available) a mathematical representation of the face or head of a person resulting in “n” outputs referred to as “encoding data 1 to n” (EDI to n)in Concept 2.

[0249] 10. A system of Concept 3 wherein the “sub-program products”, each sub-program product adapted to “n” different angles between the BICI and the head of a person, are “pretrained neural network program products” (PTNNPP1 to n), wherein each such PTNNPP is pretrained for a specific angle between a reference on the face of a person and the camera BICI, wherein; in a 2-dimensional data (x,y) such reference on the face to calculate the angle to the camera BICI is the position of iris or centre point of the pupil versus the background sclera of the eye, where the angle is defined as;

[0250] -90 degrees LEFT when the “iris or centre point of the pupil” (IRCPP) is on the far left of the sclera of the eye, +0 degrees RIGHT / LEFT when the “iris or centre point of the pupil” (IRCPP) is in the horizontal centre of the sclera of the eye, +90 degrees RIGHT when the “iris or centre point of the pupil” (IRCPP) is on the far right of the sclera of the eye, and in between, the degrees are a linear interpolation of the mathematical calculation between the previous, and

[0251] -90 degrees DOWN when the “iris or centre point of the pupil” (IRCPP) is on the lower end of the sclera of the eye, +0 degrees UP / DOWN when the “iris or centre point of the pupil” (IRCPP) is in the vertical centre of the sclera of the eye, +90 degrees UP when the “iris or centre point of the pupil” (IRCPP) is on the upper end of the sclera of the eye, and in between, the degrees are a linear interpolation of the mathematical calculation between the previous, and resulting in a 2-dimensional horizontal angle (left / right) and a 2-dimensional vertical angle (up / down), which in combination of all the available data (one may not be available) form a mathematical representation of the angle(s) resulting in “n” outputs referred to as “encoding data 1 to n” (EDI to n) in Concept 3.

[0252] 11. A system of Concept 3 wherein the “sub-program products”, each sub-program product adapted to “n” different angles between the BICI and the head of a person, are “pretrained neural network program products” (PTNNPP1 to n), wherein each such PTNNPP is pretrained for a specific angle between a reference on the face of a person and the camera BICI, wherein; in a 3 -dimensional data (x,y, z) such reference on the face to calculate the angle to the camera BICI is the distance “z” from BICI to each those points x, y, where the angle is defined as; the linear interpolation between the delta of the distance “zl” of a reference point on the upper half of the face and the distance “z2” of a mirror reference point on the lower half of the face as the UP / DOWN angle, and the linear interpolation between the delta of the distance “z3” of a reference point on the left half of the face and the distance “z4” of a mirror reference point on the right half of the face as the LEFT / RIGHT angle, and resulting in 3 dimensional horizontal and vertical angle representation, which form a mathematical representation of the angle(s) resulting in “n” outputs referred to as “encoding data 1 to n” (EDI to n) in Concept 3.

[0253] 12. A system including at least two or more internet enabled devices (wireless devices, smartphones, portable / desktop computer), with a built-in camera and at least one or more internet enabled server (SI to Sn), wherein;

[0254] The first internet enabled device (IED1), a first non-transitory storage medium (NTSM1), and a first computer program product embodied on the first non-transitory storage medium, the first computer program product executable on the IED1 when executed communicates with the server (SI), and the second internet enabled device (IED2), a second non-transitory storage medium (NTSM2), and a second computer program product embodied on the second non- transitory storage medium, the second computer program product executable on the IED2 when executed communicates with the server (SI), and both the first- and the second- computer program product communicate with a third non-transitory storage medium (NTSM3), and a third computer program product in the third non-transitory storage medium, the third computer program product executable on SI when executed communicates with IED1 and IED2, and wherein, the first computer program product when executed on IED1 uses a built-in camera (BICI) to extract the “digital data from the camera” (DDC1) of IED1, and transmit such (DDC1) data through the communication channel (CCH1) between IED1 and SI, and the second computer program product when executed on IED2 uses a built-in camera (BIC2) to extract the “digital data from the camera” (DDC2) of IED2, and transmit such (DDC2) data through the communication channel (CCH2) between IED2 and SI, and wherein DDC1 and DDC2 is processed by SI computer program product and where in the event SI computer program product detects new data being received from IED1 or IED2, such new input data is stored in the SI non-transitory storage medium and indexed such that each such new input data is associated to the originating user account of IED1 or IED2 of the system, and wherein;

[0255] (i) the fourth computer program product in SI adapted to include “n” subprogram products, each sub-program product adapted to process new data (DDC1, DDC2 to DDCn) in parallel for each new input data, resulting in “n” outputs referred to as “encoding data 1, 2 to n” (EDI, ED2 to EDn), and

[0256] (ii) wherein each such output “encoding data EDI, ED2 to n” is used as input to a fifth computer program product in SI to Sn which compares the new received “encoding data” (EDI or ED2) with the corresponding historical past “encoding data” of the IED1 or IED2 user account stored in the SI to Sn non-transitory storage medium indexed to the corresponding user account (in example, compares EDI with ED1-1 to EDl-n, where EDI is the current new input data and ED1-1 is the previous historical input data and EDl-n is most past historical input data when that user account was opened, AND compares ED2 with ED2-1 to ED2-n, where ED2 is the current new input data and ED2-1 is the previous historical input data and ED2-n is most past historical input data when that user account was opened), AND

[0257] (iii) wherein, in the event of a mismatch (meaning the match is below a predefined threshold “x”) between EDI and ED1-1 to EDl-n OR a mismatch between ED2 and ED2-1 to ED2-n then the transaction of the user account corresponding to the mismatch will be denied by the system (if EDI shows a mismatch it means it is not the same person as the authorised person of the user account of device IED1 and this user account will be denied the transaction that it was intending to do with device IED2, if ED2 shows a mismatch it means it is not the same person as the authorised person of the user account of device IED2 and this user account will be denied the transaction that it was intending to do with device IED1), AND

[0258] (iv) wherein, in the event there is a match above a predefined threshold “y” between EDI and ED 1-1 to EDl-n AND between ED2 and ED2-1 to ED2-n then the transaction between the user of device IED1 and IED2 is allowed to proceed by the system, allowing IED1 and IED2 to instruct any such transaction between their corresponding user, respectively through the communications channels CCH1 and CCH2 for execution by the computer program product of SI. Wherein the user of the account corresponding to devices IED1 and IED2 could be two end-users, two merchants, an end-user and a merchant, etc. and wherein the transaction requested for execution could be a bank transfer, an online payment, a physical point of sale payment, a government benefits request, an insurance claim, a mortgage application, a loan application, etc.

[0259] 13. A system of Concept 12 wherein only the check of a match or mismatch is performed on the encoding data corresponding to an end-user of a device (IED1), comparing EDI with ED 1-1 to EDl-n and in the event of a match, the system will allow the transaction request to proceed between IED1 and IED2 through SI.

[0260] 14. A system of Concept 12 wherein the system does not do any checks but instead sends the encoding data of the end-users of the device (IED1) to the screen of the merchant device IED2 and it is the merchant of device IED2 that checks visually, and if the merchant confirms the user in front of him (in example in a physical point of sale use-case) he matches with the image corresponding with the encoding data of the user in front of him (ED1-1 of a past successful transaction or with EDl-n the original image encoding data when end-user of device IED1 opened his account), if merchant confirms manually by clicking on his device IED2 then through communication channel CCH2 it notifies the computer program product in SI and the transaction can proceed, otherwise the transaction is denied and communicated by SI computer program product to both user devices IED1 and IED2 respectively through communication channel CCH2

[0261] 15. A system of Concept 14 wherein the check of a match or mismatch is performed on the encoding data corresponding to an end-user of a device (IED1), comparing EDI with ED1-1 to EDl-n and wherein; i. in the event there is a match above a predefined threshold “y” the system will allow the transaction request to proceed between IED1 and IED2 through SI. ii. in the event of a mismatch (meaning the match is below a predefined threshold “x”) between EDI and ED1-1 to EDl-n the system will deny the transaction request. iii. in the event there is a match above a predefined threshold “x” BUT below a predefined threshold “y” the system will send the encoding data of the end-users of the device (IED1) to the screen of the merchant device IED2 and it is the merchant of device IED2 that checks visually, and if the merchant confirms the user in front of him (in example in a physical point of sale use-case) he matches with the image corresponding with the encoding data of the user in front of him (ED1-1 of a past successful transaction or with EDl-n the original image encoding data when end-user of device IED1 opened his account), if merchant confirms manually by clicking on his device IED2 then through communication channel CCH2 it notifies the computer program product in SI and the transaction can proceed, otherwise the transaction is denied and communicated by SI computer program product to both user devices IED1 and IED2 respectively through communication channel CCH2.

[0262] Note

[0263] It is to be understood that the above-referenced arrangements are only illustrative of the application for the principles of the present invention. Numerous modifications and alternative arrangements can be devised without departing from the spirit and scope of the present invention. While the present invention has been shown in the drawings and fully described above with particularity and detail in connection with what is presently deemed to be the most practical and preferred example(s) of the invention, it will be apparent to those of ordinary skill in the art that numerous modifications can be made without departing from the principles and concepts of the invention as set forth herein.

Claims

CLAIMS1. A system including at least one or more internet enabled devices (e.g. wireless devices, smartphones, portable computer, desktop computer), each internet enabled device including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executable on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, the at least one server including a respective second non-transitory storage medium (NTSM2), and a respective second computer program product embodied on the respective second non-transitory storage medium, the respective second computer program product executable on the at least one server to communicate with a first internet enabled device of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device is configured to use the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, wherein the respective second computer program product is executable on the at least one server to process the extracted digital image data, wherein the respective second computer program product is executable on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response to store the new digital image data in the respective second non-transitory storage medium of the at least one server and to index the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device, and wherein;(i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product adapted to analyse image data for a respective different ethnic group to process the new digital imagedata in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and(ii) wherein each resulting output “encoding data 1 to n” of the n outputs is used as input to a respective fourth computer program product in the at least one server which is executable on the at least one server to compare all the n outputs and to ignore all such outputs that are identical and to calculate an interpolation between outputs that are different to each other to generate a new modified output, referred to as an “auto modified encoding data” (AMED), and to store such “auto modified encoding data” in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device, and(iii) wherein, in the event of only one output of the n outputs being different and all other outputs of the n outputs being the same, then no interpolation is performed in (ii), and the one different output of the n outputs becomes the “auto modified encoding data” (AMED) which is stored in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device.

2. A system including at least one or more internet enabled devices (e.g. wireless devices, smartphones, portable computer, desktop computer), each internet enabled device including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executable on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, the at least one server including a respective second non-transitory storage medium (NTSM2), and a respective second computer program product embodied on the respective second non-transitory storage medium, the respective second computer program product executable on the at least one server to communicate with a first internet enabled device of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device is configured to use the respective built-in camera of the first internetenabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, wherein the respective second computer program product is executable on the at least one server to process the extracted digital image data, wherein the respective second computer program product is executable on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response to store the new digital image data in the respective second non-transitory storage medium of the at least one server and to index the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device, and wherein;(i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product adapted to analyse image data for respective different parts of a head of a person to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and(ii) wherein each resulting output “encoding data 1 to n” of the n outputs is used as input to a respective fourth computer program product in the at least one server which is executable on the at least one server to compare all the n outputs and to ignore all such outputs that are inconclusive (e.g. in which a match is not produced) and to generate a new modified output, referred to as an “auto modified encoding data” (AMED), and to store such “auto modified encoding data” in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device, and(iii) wherein, in the event of all outputs being inconclusive (e.g. in which in all a match is not produced) then the new digital image data becomes the “auto modified encoding data” (AMED) which is stored in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device.

3. A system including at least one or more internet enabled devices (e.g. wireless devices, smartphones, portable computer, desktop computer), each internet enableddevice including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executable on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, the at least one server including a respective second non-transitory storage medium (NTSM2), and a respective second computer program product embodied on the respective second non-transitory storage medium, the respective second computer program product executable on the at least one server to communicate with a first internet enabled device of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device is configured to use the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, wherein the respective second computer program product is executable on the at least one server to process the extracted digital image data, wherein the respective second computer program product is executable on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response to store the new digital image data in the respective second non-transitory storage medium of the at least one server and to index the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device, and wherein;(i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product adapted to analyse a respective image of a different respective viewing angle of a head when viewed from the respective built-in camera, to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and(ii) wherein each resulting output “encoding data 1 to n” of the n outputs is usedas input to a respective fourth computer program product in the at least one server which is executable on the at least one server to compare all the n outputs and to ignore all such outputs that are inconclusive (e.g. in which a match is not produced) and to calculate an interpolation between the conclusive (e.g. in which a match is produced) outputs that are different to each other, to generate a new modified output, referred to as an “auto modified encoding data” (AMED), and to store such “auto modified encoding data” in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device, and (iii) wherein, in the event of only one output being conclusive (e.g. in which a match is produced) and all other outputs are inconclusive (e.g. in which a match is not produced), then no interpolation is performed and the one conclusive (e.g. in which a match is produced) output becomes the “auto modified encoding data” (AMED) which is stored in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device.

4. A system of Claim 1 or Claim 2 or Claim 3, wherein the digital image data of the first internet enabled device, consists or comprises of a single image, or consecutive images, or a short video, wherein in the case of a short video, images are extracted from the short video corresponding to the face or head of a person, or to an object, and the extracted images are converted into respective two-dimensional digital matrix arrays x, y where “x, y” are the coordinates of each matrix array relative to a reference point, and wherein the single image, or the consecutive images, or the converted extracted images, are transmitted through the communication channel between the first internet enabled device and the at least one server, for processing.

5. A system of Claim 1 or Claim 2 or Claim 3, wherein the digital image data of the first internet enabled device, consists or comprises of a single image, or consecutive images, or a short video, wherein in the case of a short video, images are extracted from the short video corresponding to the face or head of a person, or to an object, and the wherein the image, or the consecutive images, or the extracted images are converted into respective three-dimensional digital matrix arrays (x, y, d where “x, y” are the coordinates of each array element in the matrix and the 3rddimension “d” corresponds to the respective distance from the respective built-in camera to eachrespective array image element) and wherein the respective three-dimensional digital matrix arrays are transmitted through the communication channel between the first internet enabled device and the at least one server, for processing.

6. A system of Claim 1 or Claim 2 or Claim 3, wherein the digital image data of the first internet enabled device, consists or comprises of a single image, or consecutive images, or a short video, wherein in the case of a short video, images are extracted from the short video corresponding to the face or head of a person, or to an object, and the wherein the image, or the consecutive images, or the extracted images are converted into respective three-dimensional digital matrix arrays (x, y, c where “x, y” are the coordinates of each array element in the matrix and the 3rddimension “c” corresponds to a respective colour code of each respective array element in the matrix) and wherein the respective three-dimensional digital matrix arrays are transmitted through the communication channel between the first internet enabled device and the at least one server, for processing.

7. A system of Claim 1 or Claim 2 or Claim 3, wherein the digital image data of the first internet enabled device, consists or comprises of a single image, or consecutive images, or a short video, wherein in the case of a short video, images are extracted from the short video corresponding to the face or head of a person, or to an object, and the wherein the image, or the consecutive images, or the extracted images are converted into respective four-dimensional digital matrix arrays (x, y, c, d where “x, y” are the coordinates of each array element in the matrix and the 3rddimension “c” corresponds to a respective colour code of each respective array element in the matrix and the 4thdimension “d” corresponds to the respective distance from the respective built-in camera to each respective array image element) and wherein the respective four-dimensional digital matrix arrays are transmitted through the communication channel between the first internet enabled device and the at least one server, for processing.

8. A system of Claim 1 wherein the n sub-program products, are pretrained neural network program products (PTNNPP1 to n), wherein each respective pretrained neural network program product is pretrained for a specific respective ethnic or sub-ethnic-group of an ethnic group (e.g. Hispanic-Spain, Hispanic descent-Mexico, Hispanic descent-Peru, etc., African descent-North America, African descent-United Kingdom, African descent-Nigeria, African descent-Kenya, African descent South- Africa, etc., Asian descent-China, Asian descent-Japan, Asian descent-Indonesia, Asian descent-Singapore, Caucasian descent-North America, Caucasian descent- United Kingdom, Caucasian descent-Germany, Caucasian descent-South Africa etc.) and is usable to generate a respective mathematical representation of the face of a person, to generate a respective output of the n outputs.

9. A system of Claim 2 wherein the n sub-program products, are pretrained neural network program products (PTNNPP I to n), wherein each respective pretrained neural network program product is pretrained for a specific respective area or sub-area of a head of a person or a face of a person (e.g. area enclosing both eyes, or the subarea of each enclosure of each eye separately, the area above the eyes / forehead or sub-area of the eye-brows separately from the area above the eye-brows, the area below the eyes but above the lips or the sub-areas of the nose separately to the subarea of each cheek, the area of the nose to include the nostril and bridge of the nose, the area mouth or the sub-area of the upper and lower lip separately to the area left and right of the lip endings towards each lower cheek area, the area of the chin, the area of the neck, the area of each ear or sub-area of each earlobe separately) and is usable to generate a respective mathematical representation of the face or head of a person, wherein the n sub-program products are usable to generate the n outputs.

10. A system of Claim 3 wherein the n sub-program products are pretrained neural network program products (PTNNPPI to n), wherein each respective pretrained neural network program product is pretrained to analyse an image of a respective viewing angle of a reference point on the face of a person when viewed from the respective built-in camera, wherein: for a two-dimensional image data (x,y) such reference point on the face to calculate an angle to the respective built-in camera is the position of iris or centre point of the pupil versus the background sclera of the eye, where the angle is defined as:-90 degrees LEFT when the “iris or centre point of the pupil” (IRCPP) is on the far left of the sclera of the eye, +0 degrees RIGHT / LEFT when the “iris or centre point ofthe pupil” (IRCPP) is in the horizontal centre of the sclera of the eye, +90 degrees RIGHT when the “iris or centre point of the pupil” (IRCPP) is on the far right of the sclera of the eye, and in between, the degrees are a linear interpolation of the mathematical calculation between the defined -90 degrees LEFT, +0 degrees RIGHT / LEFT and +90 degrees RIGHT, and-90 degrees DOWN when the “iris or centre point of the pupil” (IRCPP) is on the lower end of the sclera of the eye, +0 degrees UP / DOWN when the “iris or centre point of the pupil” (IRCPP) is in the vertical centre of the sclera of the eye, +90 degrees UP when the “iris or centre point of the pupil” (IRCPP) is on the upper end of the sclera of the eye, and in between, the degrees are a linear interpolation of the mathematical calculation between the between the defined -90 degrees DOWN, +0 degrees UP / DOWN and +90 degrees UP, and resulting in a two-dimensional horizontal angle (left / right) and a two-dimensional vertical angle (up / down), and is usable to generate a respective mathematical representation of the angle(s), wherein the n sub-program products are usable to generate the n outputs.

11. A system of Claim 3 wherein the n sub-program products are pretrained neural network program products (PTNNPP1 to n), wherein each respective pretrained neural network program product is pretrained to analyse an image of a respective viewing angle of a reference point on the face of a person when viewed from the respective built-in camera, wherein: in a three-dimensional image data (x,y, z) such reference point on the face to calculate the angle to the respective built-in camera is the distance z from the respective built-in camera to each of those points x, y, where the angle is defined as; the linear interpolation between the delta of the distance zl of a reference point on the upper half of the face and the distance z2 of a mirror reference point on the lower half of the face as the UP / DOWN angle, and the linear interpolation between the delta of the distance z3 of a reference point on the left half of the face and the distance z4 of a mirror reference point on the right half of the face as the LEFT / RIGHT angle, and resulting in 3 dimensional horizontal and vertical angle representation, and is usable to generate a respective mathematical representation of the horizontal and vertical angle representation, wherein the n sub-program products are usable to generate the noutputs.

12. A system including at least two or more internet enabled devices (e.g. wireless devices, smartphones, portable / desktop computer), each internet enabled device including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executable on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, wherein each respective first computer program product is configured to communicate with a second computer program product stored in a second non-transitory storage medium (NTSM2) of the at least one server, the second computer program product executable on the at least one server to communicate with a first internet enabled device (IED1) of the at least one or more internet enabled devices, and to communicate with a second internet enabled device (IED2) of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device is configured to use the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, and wherein the respective first computer program product when executing on the second internet enabled device is configured to use the respective built-in camera of the second internet enabled device to extract digital image data from the respective built-in camera of the second internet enabled device, and to transmit the extracted digital image data through a communication channel between the second internet enabled device and the at least one server, wherein the respective second computer program product is executable on the at least one server to process the extracted digital image data, wherein the respective second computer program product is executable on the at least one server to detect newdigital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response to store the new digital image data in the respective second non-transitory storage medium of the at least one server and to index the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device; wherein the respective second computer program product is executable on the at least one server to detect new digital image data being received from the second internet enabled device of the at least one or more internet enabled devices, and in response to store the new digital image data in the respective second non-transitory storage medium of the at least one server and to index the new digital image data such that the new digital image data is associated with an originating user account of the second internet enabled device, and wherein:(i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product adapted to analyse image data to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and(ii) wherein each such output of the n outputs is used as input to a fourth computer program product in the at least one server which is executable on the at least one server to compare the n outputs with corresponding historical past n outputs of the user account of the first internet enabled device and / or of the user account of the second internet enabled device, stored in the second non-transitory storage medium, indexed to the corresponding user account (in example, compares EDI with ED1-1 to EDl-n, where EDI is the current new input data and ED1-1 is the previous historical input data and EDl-n is most past historical input data when that user account was opened, AND compares ED2 with ED2-1 to ED2-n, where ED2 is the current new input data and ED2-1 is the previous historical input data and ED2-n is most past historical input data when that user account was opened), AND(iii) wherein, in the event of a mismatch (meaning the match is below a predefined threshold “x”) between the n outputs with the corresponding historical past n outputs of the user account of the first internet enabled device OR a mismatch between the n outputs with the corresponding historical past n outputs of the user account of the second internet enabled device, then the transaction of the user account correspondingto the mismatch is denied by the system (e.g. if EDI shows a mismatch it means it is not the same person as the authorised person of the user account of device IED1 and this user account will be denied the transaction that it was intending to do with device IED2, if ED2 shows a mismatch it means it is not the same person as the authorised person of the user account of device IED2 and this user account will be denied the transaction that it was intending to do with device IED1), AND(iv) wherein, in the event there is a match above a predefined threshold “y” between the n outputs with the corresponding historical past n outputs of the user account of the first internet enabled device AND a match between the n outputs with the corresponding historical past n outputs of the user account of the second internet enabled device, then the transaction between the users of the first internet enabled device and the second internet enabled device is allowed to proceed by the system, allowing the first internet enabled device and the second internet enabled device to instruct any such transaction between their corresponding users, respectively through the respective communications channels (CCH1 and CCH2) for execution by a fifth computer program product of the at least one server, wherein the users of the accounts corresponding to the first internet enabled device and to the second internet enabled device could be two end-users, two merchants, an end-user and a merchant, etc. and wherein the transaction requested for execution could be a bank transfer, an online payment, a physical point of sale payment, a government benefits request, an insurance claim, a mortgage application, a loan application, etc.

13. A system of Claim 12 wherein only the check of a match or mismatch is performed on the encoding data corresponding to an end-user of a device (IED1), comparing EDI with ED 1-1 to EDl-n and in the event of a match, the system will allow the transaction request to proceed between IED1 and IED2 through SI.

14. A system of Claim 12 wherein the system does not do any checks but instead sends the encoding data of the end-users of the device (IED1) to the screen of the merchant device IED2 and it is the merchant of device IED2 that checks visually, and if the merchant confirms the user in front of him (in example in a physical point of sale use-case) he matches with the image corresponding with the encoding data of the user in front of him (ED1-1 of a past successful transaction or with EDl-n the originalimage encoding data when end-user of device IED1 opened his account), if merchant confirms manually by clicking on his device IED2 then through communication channel CCH2 it notifies the computer program product in SI and the transaction can proceed, otherwise the transaction is denied and communicated by SI computer program product to both user devices IED1 and IED2 respectively through communication channel CCH2.

15. A system of Claim 14 wherein the check of a match or mismatch is performed on the encoding data corresponding to an end-user of a device (IED1), comparing EDI with ED1-1 to EDl-n and wherein; i. in the event there is a match above a predefined threshold “y” the system will allow the transaction request to proceed between IED1 and IED2 through SI. ii. in the event of a mismatch (meaning the match is below a predefined threshold “x”) between EDI and ED1-1 to EDl-n the system will deny the transaction request. iii. in the event there is a match above a predefined threshold “x” BUT below a predefined threshold “y” the system will send the encoding data of the end-users of the device (IED1) to the screen of the merchant device IED2 and it is the merchant of device IED2 that checks visually, and if the merchant confirms the user in front of him (in example in a physical point of sale use-case) he matches with the image corresponding with the encoding data of the user in front of him (ED1-1 of a past successful transaction or with EDl-n the original image encoding data when end-user of device IED1 opened his account), if merchant confirms manually by clicking on his device IED2 then through communication channel CCH2 it notifies the computer program product in SI and the transaction can proceed, otherwise the transaction is denied and communicated by SI computer program product to both user devices IED1 and IED2 respectively through communication channel CCH2.

16. A computer-implemented method of generating and storing a new modified output, the method using a system including at least one or more internet enabled devices (e.g. wireless devices, smartphones, portable computer, desktop computer), each internet enabled device including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium(NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executing on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, the at least one server including a respective second non-transitory storage medium (NTSM2), and a respective second computer program product embodied on the respective second non-transitory storage medium, the respective second computer program product executing on the at least one server to communicate with a first internet enabled device of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device uses the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and transmits the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, wherein the respective second computer program product executes on the at least one server to process the extracted digital image data, wherein the respective second computer program product executes on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response stores the new digital image data in the respective second non-transitory storage medium of the at least one server and indexes the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device, and wherein;(i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product executing to analyse image data for a respective different ethnic group to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and(ii) wherein each resulting output “encoding data 1 to n” of the n outputs is used as input to a respective fourth computer program product in the at least one server which executes on the at least one server to compare all the n outputs and to ignore all such outputs that are identical and to calculate an interpolation between outputs that are different to each other to generate the new modified output, referred to as an “automodified encoding data” (AMED), and to store such “auto modified encoding data” in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device, and(iii) wherein, in the event of only one output of the n outputs being different and all other outputs of the n outputs being the same, then no interpolation is performed in (ii), and the one different output of the n outputs becomes the “auto modified encoding data” (AMED) which is stored in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device.

17. A computer-implemented method of generating and storing a new modified output, the method using a system including at least one or more internet enabled devices (e.g. wireless devices, smartphones, portable computer, desktop computer), each internet enabled device including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executing on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, the at least one server including a respective second non-transitory storage medium (NTSM2), and a respective second computer program product embodied on the respective second non-transitory storage medium, the respective second computer program product executing on the at least one server to communicate with a first internet enabled device of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device uses the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, wherein the respective second computer program product executes on the at least one server to process the extracted digital image data, wherein the respective second computer program product executes on the at least one server to detect newdigital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response stores the new digital image data in the respective second non-transitory storage medium of the at least one server and indexes the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device, and wherein;(i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product analysing image data for respective different parts of a head of a person to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and(ii) wherein each resulting output “encoding data 1 to n” of the n outputs is used as input to a respective fourth computer program product in the at least one server which executes on the at least one server to compare all the n outputs and to ignore all such outputs that are inconclusive (e.g. in which a match is not produced) and to generate the new modified output, referred to as an “auto modified encoding data” (AMED), and to store such “auto modified encoding data” in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device, and(iii) wherein, in the event of all outputs being inconclusive (e.g. in which in all a match is not produced) then the new digital image data becomes the “auto modified encoding data” (AMED) which is stored in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device.

18. A computer-implemented method of generating and storing a new modified output, the method using a system including at least one or more internet enabled devices (e.g. wireless devices, smartphones, portable computer, desktop computer), each internet enabled device including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer programproduct executing on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, the at least one server including a respective second non-transitory storage medium (NTSM2), and a respective second computer program product embodied on the respective second non-transitory storage medium, the respective second computer program product executing on the at least one server to communicate with a first internet enabled device of the at least one or more internet enabled devices, and wherein the respective first computer program product when executing on the first internet enabled device uses the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server, wherein the respective second computer program product is executing on the at least one server to process the extracted digital image data, wherein the respective second computer program product is executing on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response to store the new digital image data in the respective second non-transitory storage medium of the at least one server and to index the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device, and wherein;(i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product analysing a respective image of a different respective viewing angle of a head when viewed from the respective built-in camera, to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and(ii) wherein each resulting output “encoding data 1 to n” of the n outputs is used as input to a respective fourth computer program product in the at least one server which is executing on the at least one server to compare all the n outputs and to ignore all such outputs that are inconclusive (e.g. in which a match is not produced) and to calculate an interpolation between the conclusive (e.g. in which a match is produced) outputs that are different to each other, to generate the new modified output, referredto as an “auto modified encoding data” (AMED), and to store such “auto modified encoding data” in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device, and(iii) wherein, in the event of only one output being conclusive (e.g. in which a match is produced) and all other outputs are inconclusive (e.g. in which a match is not produced) then no interpolation is performed and the one conclusive (e.g. in which a match is produced) output becomes the “auto modified encoding data” (AMED) which is stored in the respective second non-transitory storage medium indexed to the corresponding user account of the first internet enabled device.

19. A computer-implemented method of verifying a transaction, the method using a system including at least two or more internet enabled devices (e.g. wireless devices, smartphones, portable / desktop computer), each internet enabled device including a respective built-in camera, the system further including at least one or more internet enabled servers (SI to Sn), wherein: each internet enabled device includes a respective first non-transitory storage medium (NTSM1), and a respective first computer program product embodied on the respective first non-transitory storage medium, the respective first computer program product executing on the respective internet enabled device to communicate with at least one server (SI) of the at least one or more internet enabled servers, wherein each respective first computer program product communicates with a second computer program product stored in a second non-transitory storage medium (NTSM2) of the at least one server, the second computer program product executing on the at least one server to communicate with a first internet enabled device (IED1) of the at least one or more internet enabled devices, and to communicate with a second internet enabled device (IED2) of the at least one or more internet enabled devices, and wherein the respective first computer program product executes on the first internet enabled device to use the respective built-in camera of the first internet enabled device to extract digital image data from the respective built-in camera of the first internet enabled device, and to transmit the extracted digital image data through a communication channel between the first internet enabled device and the at least one server,and wherein the respective first computer program product executes on the second internet enabled device to use the respective built-in camera of the second internet enabled device to extract digital image data from the respective built-in camera of the second internet enabled device, and to transmit the extracted digital image data through a communication channel between the second internet enabled device and the at least one server, wherein the respective second computer program product executes on the at least one server to process the extracted digital image data, wherein the respective second computer program product executes on the at least one server to detect new digital image data being received from the first internet enabled device of the at least one or more internet enabled devices, and in response stores the new digital image data in the respective second non-transitory storage medium of the at least one server and indexes the new digital image data such that the new digital image data is associated with an originating user account of the first internet enabled device; wherein the respective second computer program product executes on the at least one server to detect new digital image data being received from the second internet enabled device of the at least one or more internet enabled devices, and in response stores the new digital image data in the respective second non-transitory storage medium of the at least one server and indexes the new digital image data such that the new digital image data is associated with an originating user account of the second internet enabled device, and wherein:(i) the at least one server includes a respective third computer program product which includes n sub-program products, each sub-program product analysing image data to process the new digital image data in parallel for each new input digital image data, resulting in n outputs referred to as “encoding data 1 to n” (EDI to n), and(ii) wherein each such output of the n outputs is used as input to a fourth computer program product in the at least one server which executes on the at least one server to compare the n outputs with corresponding historical past n outputs of the user account of the first internet enabled device and / or of the user account of the second internet enabled device, stored in the second non-transitory storage medium, indexed to the corresponding user account (in example, compares EDI with ED1-1 to EDl-n, where EDI is the current new input data and ED1-1 is the previous historical input data andEDl-n is most past historical input data when that user account was opened, AND compares ED2 with ED2-1 to ED2-n, where ED2 is the current new input data and ED2-1 is the previous historical input data and ED2-n is most past historical input data when that user account was opened), AND(iii) wherein, in the event of a mismatch (meaning the match is below a predefined threshold “x”) between the n outputs with the corresponding historical past n outputs of the user account of the first internet enabled device OR a mismatch between the n outputs with the corresponding historical past n outputs of the user account of the second internet enabled device, then the transaction of the user account corresponding to the mismatch is denied by the system (e.g. if EDI shows a mismatch it means it is not the same person as the authorised person of the user account of device IED1 and this user account will be denied the transaction that it was intending to do with device IED2, if ED2 shows a mismatch it means it is not the same person as the authorised person of the user account of device IED2 and this user account will be denied the transaction that it was intending to do with device IED1), AND(iv) wherein, in the event there is a match above a predefined threshold “y” between the n outputs with the corresponding historical past n outputs of the user account of the first internet enabled device AND a match between the n outputs with the corresponding historical past n outputs of the user account of the second internet enabled device, then the transaction between the users of the first internet enabled device and the second internet enabled device is allowed to proceed by the system, allowing the first internet enabled device and the second internet enabled device to instruct any such transaction between their corresponding users, respectively through the respective communications channels (CCH1 and CCH2) for execution by a fifth computer program product of the at least one server, wherein the users of the accounts corresponding to the first internet enabled device and to the second internet enabled device could be two end-users, two merchants, an end-user and a merchant, etc. and wherein the transaction requested for execution could be a bank transfer, an online payment, a physical point of sale payment, a government benefits request, an insurance claim, a mortgage application, a loan application, etc.

20. The method of any of Claims 16 to 19, the method using a system of any ofClaims 1 to 15.

Images