Method and system for using visual analysis to validate the identity of a remote user

Abstract

A computer-implemented method for using visual analysis to validate the identity of a remote user, comprising: retrieving a digital copy of the user's identification document comprising a photograph of the user and a signature of the user; requesting a location specification of the user; requesting a set of photographs taken in real time, wherein the set of photographs taken in real time comprises a first photograph showing the user and the associated background of the specified location, and a second photograph showing only the background of the specified location of the first photograph; requesting a signature taken in real time together with each photograph of the requested set of photographs taken in real time; comparing the requested signatures taken in real time with the user's signature on the associated digital identification document;Comparing the background of the first photo of the real-time set of photos with the background of the second photo of the real-time set of photos and verifying the user's stated location; comparing an image of the user in the first photo of the real-time set of photos with the user's photo on the associated digital ID; and validating the user's identity based on the results of the signature comparison, the background comparison of the photos, and the user image comparison, each with regard to reaching a threshold for each comparison result.

Description

BACKGROUND OF THE INVENTION

[0001] The present disclosure relates to a universal architecture (high level architecture) and in particular to image analysis methods for validating the identity of a user.

[0002] Know Your Customer (KYC) is a process for validating the identity of users. The KYC process is typically a personalized process with strict rules for identifying and validating customer identities. Consequently, written programs for customer identification must be created and maintained to ensure compliance with identity verification standards. Long-standing limitations prevent users from validating their identity; this is a critical step in the Know Your Customer process. Current technology only inadequately prevents fraudulent remote account verification. SUMMARY

[0003] Aspects of the present disclosure relate to a method for using image analysis techniques to verify the identity of a remote user. The method may include retrieving a photograph of the user and a signature of the user from a digitally readable identification document and simultaneously requesting a set of real-time photographs. The set of real-time photographs includes a first photograph and a second photograph. The method may further include requesting a signature rendered in real time along with the requested set of real-time photographs. The method may also include comparing the signature rendered in real time in the first photograph with the signature of the user on the digitally readable identification document. Additionally, the method may include comparing a background in the first photograph of the set of real-time photographs with a background in the second photograph of the set of real-time photographs.Furthermore, the procedure may involve comparing an image of the user in the first photo of the set of real-time photos with the photo of the user on the digitally readable identification document and beginning to validate the user's identity in response to the finding that the signature provided in real time in the first photo of the set, compared with the signature of the user retrieved from the digitally readable identification document, reaches a first threshold, that the background in the first photo of the set of real-time photos, compared with the background in the second photo of the set of real-time photos, reaches a second threshold, and that the image of the user in the first photo of the set of real-time photos, compared with the photo of the user on the digitally readable identification document, reaches a third threshold.According to some selectable embodiments, the method further includes refusing to verify the user's identity in response to the detection of at least one of the following: a number of signature matches falling below the first threshold, a number of background matches falling below the second threshold, or a number of user matches falling below the third threshold, with initiating an online video session being one of the grounds for refusing to verify the user's identity.

[0004] Aspects of the present disclosure relate to a system comprising a computer-readable storage medium for storing a data set, a user interface for inputting values ​​and outputting values, and a processor connected to the computer-readable storage medium and the user interface for exchanging data. The processor contains a memory containing instructions. The instructions may be used to employ image analysis techniques to verify the identity of a remote user. The instructions may also be used to select a photograph of the user and a signature of the user from an associated digital ID and to request a set of real-time photographs. The set of real-time photographs includes a first photograph and a second photograph. The instructions may also be used to request a signature provided in real time, along with the requested set of real-time photographs.The instructions can also be used to perform a signature comparison between the signature made in real time on the first photo and the user's signature on the associated digital ID. The instructions can also be used to perform a background comparison between the background of the first photo in the set of real-time photos and the background of the second photo in the set of real-time photos.Furthermore, the instructions can be used to perform a user comparison between an image of the user in the first photo of the set of real-time photos with the photo of the user on the associated digital ID and to confirm the user's identity in response to finding that the signature made in real time on the first photo of the set of real-time photos, compared with the signature of the user on the retrieved associated digital ID, reaches a first threshold, that the background on the first photo of the set of real-time photos, compared with the background on the second photo of the set of real-time photos, reaches a second threshold, and that the image of the user on the first photo of the set of real-time photos, compared with the photo of the user on the associated digital ID, reaches a third threshold.According to some selectable embodiments, the instructions may further serve to decline validation of the user's identity in response to the finding that: a number of signature matches falls below the first threshold, and / or a number of background matches falls below the second threshold, and / or a number of user matches falls below the third threshold, with the decline to validate the user's identity including initiating an online video session.

[0005] Aspects of the present disclosure relate to a computer program product for using image analysis methods to validate the identity of a remote user in a mobile communication application, wherein the computer program product comprises a computer-readable storage medium containing program instructions embodied therein, which are executable by a processor. The program instructions can cause the processor to select a photograph of the user and a signature of the user from an associated digital identity document and to request a set of real-time photographs. The set of real-time photographs includes a first photograph and a second photograph. The program instructions can cause the processor to request a signature provided in real time along with the requested set of real-time photographs.The program instructions can further instruct the processor to perform a signature comparison between the signature made in real time on the first photo and the user's signature on the associated digital ID. The program instructions can further instruct the processor to perform a background comparison between the background on the first photo of the set of real-time photos and the background on the second photo of the set of real-time photos.Furthermore, the program instructions can cause the processor to perform a user comparison between an image of the user in the first photo of the set of real-time photos and the photo of the user on the associated digital ID, and to validate the user's identity in response to finding that the signature made in real time on the first photo of the set of real-time photos, compared with the signature of the user on the retrieved associated digital ID, reaches a first threshold, that the background on the first photo of the set of real-time photos, compared with the background on the second photo of the set of real-time photos, reaches a second threshold, and that the image of the user on the first photo of the set of real-time photos, compared with the photo of the user on the associated digital ID, reaches a third threshold.According to some selectable embodiments, the program instructions may further cause the processor to reject validation of the user's identity in response to: a number of signature matches falling below the first threshold, and / or a number of background matches falling below the second threshold, and / or a number of user matches falling below the third threshold.

[0006] Aspects of the present disclosure relate to a method for using image analysis techniques to validate the identity of a remote user. The method may include receiving a request to open an account for a user and, in response to receiving the request, retrieving a digital copy of the user's identification document, wherein the identification document contains a photograph of the user and the user's signature. The method may further include providing a request to obtain an initial real-time photograph of the user along with the user's environment. The method may further include providing a request to obtain a second real-time photograph of the user's environment without the user.The procedure may further include verifying the user's current location in response to finding that the user's environment in the first real-time photo matches the user's environment in the second real-time photo, and validating the user's identity at least partially based on verifying the user's current location.

[0007] Aspects of the present disclosure relate to a system for using image analysis techniques to confirm a user's identity. The system includes a user interface for inputting values ​​and outputting values, and a processor connected to the user interface for exchanging data. The processor can be used to receive, via the user interface, a request to open an account for a user. The processor can further be used, in response to receiving the request, to retrieve a digital copy of the user's identification document, wherein the identification document contains a photograph of the user and the user's signature. The processor can also be used to provide a prompt to request an initial real-time photograph of the user along with a user environment.The processor can also be used to request a second real-time photo of the user's surroundings without the user present, and to request the user's current location. Furthermore, the processor can be used to verify the user's current location in response to a finding that the user's surroundings in the first real-time photo match the user's surroundings in the second real-time photo, and to confirm the user's identity, at least partially, based on proof of the user's current location.

[0008] The above summary is not intended to describe every illustrated embodiment or every implementation of the present disclosure. List of characters

[0009] The drawings included in this application are incorporated into the description and form an integral part thereof. They illustrate embodiments of the present disclosure and serve to explain the basic concepts of the disclosure. The drawings serve only to illustrate certain embodiments and are not intended to limit the disclosure. Fig. Figure 1 illustrates a flowchart of an exemplary procedure for verifying a remote user using image analysis techniques according to some embodiments of the present disclosure. Fig. Figure 2 illustrates a flowchart of an exemplary procedure for quality control of multiple images according to some embodiments of the present disclosure. Fig. Figure 3 illustrates a flowchart of an exemplary procedure for comparing signatures according to some embodiments of the present disclosure. Fig. Figure 4 illustrates a flowchart of an exemplary procedure for comparing backgrounds according to some embodiments of the present disclosure. Fig. Figure 5 illustrates a flowchart of an exemplary procedure for comparing a user according to some embodiments of the present disclosure. Fig. Figure 6 illustrates a flowchart of an exemplary error correction procedure according to some embodiments of the present disclosure. Fig. Figure 7 illustrates a block diagram of an exemplary system for mobile communication applications according to some embodiments of the present disclosure. Fig. Figure 8 illustrates a block diagram of an exemplary data processing environment according to some embodiments of the present disclosure.

[0010] Although the invention is open to various modifications and alternative forms, exemplary features are shown in the drawings and described in detail. However, it should be clear that the invention is not limited to the described embodiments. Rather, the invention is intended to include all modifications, equivalents, and alternatives that fall within the essence and scope of protection of the invention. DETAILED DESCRIPTION

[0011] Aspects of the present disclosure relate to a universal architecture. Specific aspects relate to improving the Know Your Customer (KYC) process by using image analysis techniques in a software application to confirm a user's identity. On the one hand, the present disclosure is not necessarily limited to such applications; on the other hand, various aspects of the disclosure are acceptable that arise from a discussion of different examples in this context.

[0012] The software application can be executed through a network of wireless physical units, taking into account the requirements and limitations of such units. Eligible wireless physical units include, but are not limited to, mobile phones, tablets, portable (non-stationary) units, wearables, and other units with electronics, hardware, software, sensors, actuators, and / or network connectivity. Software applications can also be internet-based, native, and / or hybrid applications.

[0013] According to embodiments, the software application operates on the basis of a simulation system (High Level Architecture, HLA). HLA is a universal architecture for distributed computer simulation systems. Using HLA, computer simulations can interact with other computer simulations regardless of the data processing platform (e.g., exchange data, synchronize actions). According to embodiments, the HLA operates on a mobile platform that has an input and an output control component; however, the mobile platform is provided only as an example for illustrative purposes and is not to be understood as a limitation. The input side has a user interface of a user device (e.g., a mobile phone) and contains a set of processes and code (e.g., HTML, CSS, JavaScript) that are executed on the user's side.The output side incorporates technology required to process the information sent from the input side and return a response to the user. Some components of the technology used in the output control component include, but are not limited to, a server, the software application, and a database.

[0014] Some organizations that might be vulnerable to malicious and fraudulent attacks (e.g., hacking, phishing, forgery) attempt to mitigate fraud-related risks by implementing a Know Your Customer (KYC) process when a customer requests access to open an account. KYC has strict rules for identifying and verifying customer identities and therefore requires the creation and maintenance of written customer identification programs that enshrine the verification standards. For example, if a customer wants to open an account, all actions must be initiated and confirmed at a local physical location, along with supplementary identification documents (e.g., national identity card, passport) to prove that the customer is indeed that person. The KYC approach prevents remote access for account opening and only allows customers within a reachable radius of a local physical location.Aspects of the present disclosure provide a solution to this problem.

[0015] Aspects of the present disclosure relate to image analysis methods and, in particular, comparing a user's signature affixed in real time, based on local and general features, with the signature of the holder of a presented identification document of the user, in order to validate the identity of a remote user. Real time can be defined as the temporal condition during which a user action was initiated. In the present context, an initial handwritten signature of an initial input (e.g., a photograph) is compared, based on an initial data-driven threshold, with the signature of a retrieved input (e.g., a driver's license) to determine whether the signatures are identical. According to embodiments, the signature of an identification document (e.g., an electronic Kartu Tanda Penduduk (e-KTP)) can be retrieved from a database located in an official system (e.g.,Dukcapil) is included and compared to a threshold to verify a user's signature in real time. According to embodiments, a threshold is derived from stored data, and by comparing it to the threshold, the account's authorization is determined.

[0016] Aspects of the present disclosure relate to image analysis methods and, in particular, comparing the background of a set of images that include a user and a user environment (hereinafter referred to as the background). In this context, background comparison means comparing the background of a first user photo, based on a data-driven threshold, with the background of a second photo to determine background equality. Images are captured using capabilities of the user unit, initiated by the software application. Background comparison may include various methods, including, but not limited to, scale-independent feature transformation and comparison of image sections.Various methods can be optimized using neural networks through deep learning or implemented using machine learning applications installed on the user unit. According to some embodiments, a background is compared to a threshold to verify the location of a remote user. The threshold is derived from stored data, and account authorization is determined by comparing the user's location to this threshold.

[0017] Aspects of the present disclosure relate to image analysis methods and, in particular, image comparison of a user's image, based on a threshold value, with the photograph of the holder of an identification document presented by the user, in order to confirm the identity of a remote user. In the present context, a photograph of the user in an image received from the user is compared, based on a third data-driven threshold value, with a photograph of the user contained in the retrieved digital identification document (e.g., a driver's license) to determine its identity. According to embodiments, user images are captured using capabilities of the user unit, which are initiated by the software application. Biometric methods and facial recognition can be used for user comparison to determine a degree of match.Various methods can be optimized using neural networks through deep learning or implemented using machine learning applications installed on the user device. According to some implementations, a threshold is derived from stored data, and account eligibility is determined by comparing it to this threshold.

[0018] Aspects of this disclosure may also relate to a video messaging system utilizing the capabilities of the user unit, which is initiated after account access authorization has been denied. Account access authorization may be denied, for example, if the data-driven threshold is not met when comparing the signature, background, and / or user.

[0019] The advantage of this disclosure is that it enables the implementation of a KYC process via remote access, bridging the gap between digitally active users and potential users in rural areas with difficult or no access to a local physical location for user verification. By verifying a user's signature, face, and a specified location, the in-person manual process for opening an account can be eliminated. Furthermore, aspects of this disclosure provide electronic safeguards to reduce or eliminate potential sources of fraud when granting authorization remotely, which precluded traditional systems for implementing secure remote account authorization.

[0020] The advantages mentioned are examples only, and there are embodiments that include all, some, or none of the advantages and yet remain within the scope and scope of protection of the present disclosure.

[0021] Thus, aspects of the present disclosure are necessarily rooted in computer technology, at least insofar as they enable the establishment of a KYC process in a software application environment, which was previously unavailable. Furthermore, aspects of the present disclosure provide alternative improvements to an already established KYC process. Specifically, aspects of the present disclosure offer an improved remote identification method, enhanced accuracy, and improved computing power compared to known solutions, without affecting the current KYC process.

[0022] Fig. Figure 1 illustrates a flowchart of an example procedure. 100 for verifying a remote user by comparison using image analysis methods according to some embodiments of the present disclosure. The method 100 This can be carried out, for example, by one or more processors, a mobile application, or another hardware configuration. According to various embodiments, the process begins... 100 with step 110 The procedure is explained below. 100 described how this is carried out in a mobile user unit (e.g., a mobile phone) by a software application that has a user-side and a software-side processing architecture. The method 100However, it can be accomplished using other hardware components or combinations of hardware components. According to some embodiments, a mobile application can, for example, be used to perform the following steps: 100 Requesting data from a remote user and then transmitting the requested data to a server via remote access to enable analysis and testing processes of the procedure. 100 to be processed on the basis of the data retrieved by the mobile application.

[0023] In step 110The mobile application retrieves a digital copy of the user's identification document. Depending on the implementation, the identification document may be officially issued and managed by an official system. Eligible identification documents include, but are not limited to, physical identity cards and digital identity cards (e.g., e-KTP). Eligible official systems include any government agency that stores information relating to the identification documents of its citizens and / or residents, for example, Dukcapil, the Ministry of Internal Affairs. Identification documents such as the e-KTP may contain public information about the holder, such as the holder's personal identification number (NIK), full name, place and date of birth, gender, marital status, religion, blood type, address, occupation, nationality, photograph of the holder, expiry date, place and date of issue, the holder's signature, and / or the name and signature of the issuing official.

[0024] In step 120 The mobile application can receive input data from the user interface of a user device (e.g., a mobile phone). According to embodiments, the user uploads input data (e.g., text) relating to their identity to the mobile application. This input data includes, but is not limited to, the user's personal identification number (NIK), full name, place and date of birth, gender, marital status, religion, blood type, address, occupation, nationality, and / or the place and date of issue. The mobile application then verifies the user by comparing the syntax data (e.g., parts of speech, functions, underlying grammar, etymology, language divisions, etc.) of the input data with the data from step [number missing in original text]. 110The retrieved digital copy of the user's ID card is then sent back to the system software for further processing. According to embodiments, this step involves... 120 the first physical interaction between the surface of the mobile application and the user.

[0025] Syntax comparison is the formal analysis of the input text, performed using string comparison and syntax analysis algorithms. This results in concrete syntax trees that represent the syntactic structure of each word. Through syntax analysis, for example of the input data and the digital copy of the officially issued ID card, strings, symbols, and weighting attributes can be assigned to each word. The mobile application can then evaluate the majority of the included attributes in comparison with the input data and check for matches in the syntax data.

[0026] In step130 The mobile application can provide a notification (e.g., a notice, a message, a prompt) to indicate a request for real-time photos from the user. Using the existing capabilities of the user device, the mobile application can be granted access to, for example, a camera integrated into the user device. The notification can include text, graphics, sound signals, and / or other information. In some embodiments, the notification includes instructions for taking pictures. In step 130The mobile application prompts the user to specify their location and also to take a series of pictures. For example, the mobile application can prompt the user to repeatedly take pictures at different times throughout the day (e.g., 11:00 a.m., 2:00 p.m., 6:00 p.m.). A set of pictures comprises a set of at least two clearly distinct images (e.g., a first photo, a second photo) taken within a specified time period. According to embodiments, a first photo shows the user and a specified known background of the location, and a second photo shows only the specified known background of the location of the first photo. According to embodiments, each photo in the set of pictures can be taken either by the user (e.g., as a selfie) or by someone else in a landscape or portrait format.Each photo in the set of images is processed for quality control, and after successful image capture, the mobile application prompts the user to sign directly through the user interface. 130 will be referred to below in connection with Fig. 2 described in detail.

[0027] In step 140 Can the mobile application verify the holder's signature on the retrieved digital copy of an officially issued user ID card with the majority of signatures provided by step 130The signatures are compared to confirm the user's identity. When comparing the signatures, general characteristics (e.g., writing style (e.g., italics, block capitals), writing speed (e.g., fast, slow), spacing, size, proportions, orientation (e.g., slanted, upright), spelling, punctuation) and local characteristics (e.g., letter spacing, letter size, letter proportions, curves, arcs, overlaps, pen lift, strokes) of both sources of the user signature are checked. Step 140 will be referred to below in connection with Fig. 3 described in detail.

[0028] In step 150 Can the mobile application compare the user's location to where the user was during the majority of sets of captured images from step to step? 130allegedly stops to confirm the user's identity. According to some embodiments, the mobile application can use various techniques to compare backgrounds between a corresponding first photo and a corresponding second photo to confirm this identity. These background comparison techniques include, but are not limited to, scale-invariant feature transform (SIFT) and comparisons based on image cropping. Step 150 will be referred to below in connection with Fig. 4 described in detail.

[0029] In step 160 Can the mobile application use the photo of the holder, which is the retrieved digital copy of an officially issued ID card from step 110 belonging to, with the plural in step 130The application compares the first photos taken to confirm a user's identity. According to some implementations, the mobile application uses facial recognition methods to confirm identity. Facial recognition is a biometric method for identifying a person by comparing a recently taken image or a digital image with data points in a stored dataset. Step 160 will be referred to below in connection with Fig. 5 described in detail.

[0030] In step 170With a view to a positive comparison, the user profile is verified via remote access. While not discussed in detail above, according to some embodiments, the comparison during syntax checking, signature comparison, background comparison, and user comparison is performed using a threshold. If a threshold is not reached, authorization can be denied. According to embodiments, the denial of authorization can be combined with... Fig. 6 will be described in more detail.

[0031] Fig. Figure 1 is intended to present the exemplary steps of an exemplary method for validating a remote user according to some embodiments of the present disclosure. However, according to some embodiments, individual steps may be more or less complex than in Figure 1. Fig. 1 shown, and in addition to in Fig. Further steps may be added to (or replace) the steps shown in point 1. According to some embodiments, various other steps may also be included. Fig. 1 illustrated steps have more, fewer, or different functionalities than in Fig. 1 shown. According to some embodiments, various in Fig. The illustrated steps may occur in a different order, if at all.

[0032] In Fig. 2 is a flowchart of an exemplary procedure 200 for quality control of a set of images and for access control according to some embodiments of the present disclosure. The method 200 This can be accomplished, for example, by one or more processors, a mobile application, or another hardware configuration. To illustrate the process, 200 similar to the discussion of the procedure 100described below in connection with a mobile application. Just like the procedure. 100 can aspects of the procedure 200 can be achieved through other hardware components or combinations of hardware components. Aspects of Fig. 2. They feature a user-side system between the user and the mobile application. According to some embodiments, the method is 200 to perform a sub-process of the step 130 from Fig. 1.

[0033] In step 210The mobile application begins by requesting a real-time photo of the user and their background. A message may contain a notification, a message, and / or a request to access the camera capabilities of the mobile device. A request can be verified by receiving user input or feedback, such as, but not limited to, a swipe or tap on a surface of the mobile device. After verification, the camera capabilities of the mobile device are activated.

[0034] In step 220A message (e.g., a notice, a message, a request) is sent to a user to take the first photo of a set of images, showing the user and their current background. According to some embodiments, the mobile application requires that the first photo be taken in either a portrait format (e.g., vertical) or a landscape format (e.g., horizontal) and in real time, showing an unobstructed frontal view of the user's upper body. According to some embodiments, the mobile application advises the user that the background should ideally not be white, not a solid color, and suitable for feature recognition. According to embodiments, a feature-recognizing background may include, but is not limited to, a landscape, a building, or a street environment.

[0035] In step 230The first photo in the set of images is processed by checking its image quality and generating a watermark. Checking image quality (e.g., resolution) can involve measuring specific types of image degradation (e.g., saturation, coverage, ringing) or considering all possible signal distortions, that is, multiple factors affecting image quality. These factors can include, but are not limited to, image sharpness (detail accuracy), image noise (changes in image density), hue (luminance relative to brightness), contrast (steepness of the double-logarithmic hue-ratio curve), color (color saturation), distortion (curved lines), vignetting (edge ​​darkening), lateral chromatic aberration (CLA) (color fringing), and stray light (scattered light). More precisely, according to embodiments, in step 230The quality of the first photo is checked for overexposure and image blur. Based on the image's intensity gradations, the mobile application further determines blur by measuring the number of edges in the image. Edge measurements can be performed by measuring blur (caused by a finite depth-of-field function and a finite point spread function), penumbral blur (caused by shadows cast by light sources with a non-zero radius), and shadowing. Using a machine learning-based approach, in step 230Facial features are detected to identify accessories (e.g., glasses, hat) that visibly alter (e.g., conceal, enhance) the user's natural appearance. The image quality is determined using the machine learning system integrated into the mobile application. Images of the user and background that reach a data-driven threshold are stored in a network database, sent to an application program for further processing, and output to the user interface to display a release message (e.g., a notice, message, or prompt). Images of the user and background that do not reach the threshold result in the output of a retry message (e.g., a notice, message, or prompt) to the user interface. 220 will be repeated.

[0036] After receiving a release message from step 230The mobile application sends in step 240 A second message (e.g., a notice, a message, a request) to the user to take the second photo in the set of images. According to some embodiments, the second photo must be taken within a specified time period (e.g., two minutes) after the first photo and should only include the background captured in the first image. According to some embodiments, the mobile application requires that the second photo be taken in the same format (e.g., landscape, portrait) as the first photo in the set of images and have the same angle of view of the background with identical characteristics.

[0037] In step 250The second photo in the set of images is processed by checking the image quality and generating a watermark. Checking the image quality (e.g., resolution) can involve measuring specific types of image degradation (e.g., saturation, masking, ringing) or considering all possible signal distortions, that is, multiple factors affecting image quality. These factors can include, but are not limited to, image sharpness (detail accuracy), image noise (changes in image density), hue (luminance relative to brightness), contrast (steepness of the double-logarithmic hue-ratio curve), color (color saturation), distortion (curved lines), vignetting (edge ​​darkening), lateral chromatic aberration (CLA) (color fringing), and stray light (scattered light). More precisely, according to embodiments, in step 230The quality of the second photo is checked for overexposure and image blur. According to embodiments, overexposure is detected by determining the luminance distribution of the image. Based on the intensity gradations of the image, the mobile application further determines blur by measuring the number of edges contained in the image. Edge measurements can be performed by measuring blur (caused by a finite depth-of-field function and a finite point spread function), penumbral blur (caused by shadows cast by light sources with a non-zero radius), and shadowing. Image quality is determined using a machine learning-based system integrated into the mobile application.Pure background images that reach a data-driven threshold are stored in a network database, sent to an application program for further use, and the mobile application issues a release message (e.g., a notice, message, or prompt) to the user interface. Images of the user and background that do not reach the threshold cause the mobile application to send a retry message (e.g., a notice, message, or prompt) to the user interface, and then proceed. 240 will be repeated.

[0038] After receiving two release messages, the mobile application proceeds in step 260A message is displayed to the user, prompting them to enter their alleged location and provide their signature to verify the set of images. The mobile application collects and stores each signature in a database for further processing. According to some embodiments, entering the user's alleged location may, for example, involve using a GPS (Global Positioning System) sensor included in the user unit. According to some embodiments, in step 260 It is additionally required that the signature be transcribed on the user interface using a writing instrument (e.g., finger, stylus). According to embodiments, general and local properties (e.g., the general and local properties in step 140 from Fig. 1) The signature provided is verified, stored in a database on the network and sent to the application program for further processing.

[0039] According to some embodiments, the process 200 repeated several times throughout the day to input different backgrounds and cope with different lighting conditions. Fig. Section 2 is intended to present the exemplary steps of an exemplary method for checking image quality according to some embodiments of the present disclosure. According to some embodiments, individual steps may be more or less complex than in Fig. 2 will be shown, and in addition to those in Fig. In addition to the two steps shown (or in their place), further steps may be present. According to some embodiments, various other steps may also be included. Fig. 2 illustrated steps to achieve greater, smaller, or different functionality than in Fig. 2 have shown. According to some embodiments, various other embodiments can also be found in Fig. The two illustrated steps may occur in a different order, if at all.

[0040] In Fig. 3 is a flowchart of an exemplary procedure 300 for comparing signatures according to some embodiments of the present disclosure. The method 300 This can be accomplished, for example, by one or more processors, a mobile application, or another hardware configuration. For clarity, the procedure will be described below. 300 The following describes how this is done using the mobile application. The process is the same as described above in the procedures. 100 and 200 Aspects of the procedure can be discussed. 300 by other hardware components or components of hardware components. Aspects of Fig. Three exhibit an application program of the mobile application. According to some aspects, the procedure represents 300 a sub-procedure of step 140 in Fig. 1 dar.

[0041] In step 310 The mobile application retrieves the signature from the corresponding ID card in step 110 from Fig. 1 from. In step 320 will be the majority of the proceedings 200 The system retrieves assigned signatures from the network database to which the entered data belongs and compares them. According to some embodiments, after creating an average assigned signature based on the data from the process, the system... 200 assigned majority signatures the average assigned signature with the one in step 110 The retrieved signature was compared. According to further embodiments, each assigned signature can be compared with the one in step 110The retrieved signatures are compared to obtain an average value of the results.

[0042] The collected data from the majority of assigned signatures is analyzed to generate an average assigned signature. Any general and local characteristic can be mapped and used to identify recognizable aspects and similarities between the assigned signatures and the signature on the user's identification document. For example, by measuring the unavoidable time variations when signing, the mobile application establishes a basis for determining a user's habit of providing their legally binding signature. If users are indeed the same person as the identified individual, the time variation in each signature is significantly smaller than for individuals attempting to forge the signature.Furthermore, statistical averages can be determined from an average signature, allowing both signatures to be quantifiably measured based on distances, proportions, and angles of inclination, without being limited to these factors. Various methods can be used to compare each assigned signature with the signature on the identification document.

[0043] According to some embodiments, the mobile application can use a neural network approach to filter out a set of features representative of each signature (e.g., length, height, duration) and learn the relationship between each class (e.g., genuine, forged). Once a feature has been identified, the mobile application can use features from the outline, direction, and data volume of a signature to generate a score for comparison.

[0044] According to some embodiments, the mobile application can use a pattern matching approach to take one-dimensional positional changes (e.g., letter orientation) from the signature on the ID card to compare them with the captured data of the associated signatures and physically map each signature to it in terms of characteristic matches.

[0045] According to some embodiments, the mobile application can use a statistical approach to determine correlation coefficients for each signature by analyzing the relationship between two or more captured data points (e.g., general characteristics, local characteristics, letter slant angles) and their deviations. Each signature correlation coefficient can then be weighted to identify deviations from agreement.

[0046] In step 330Based on the stored data of the signature associated with the ID card, it is determined whether the number of matches between the assigned signatures and the signature on the ID card reaches the threshold. If the threshold is reached, the signature is verified in block form. 340 This concludes the process. For example, in block... 340 The processing in the application program continued with the following processing step, and a release message (e.g., a notice, a message, a request) was sent to the user interface for signature comparison. In block 350 If assigned signatures do not meet the signature threshold for the ID card, an error message (e.g., a warning, a notification, a request) is sent to the user interface. The authorization is then rejected, and processing in the application program is terminated.

[0047] Fig. Figure 3 presents exemplary steps of an exemplary method for comparing signatures according to some embodiments of the present disclosure. However, according to some embodiments, individual steps may be more or less complex than shown. Fig. 3 shown and in addition to those in Fig. The 3 steps shown (or their place) may be present. According to some embodiments, various elements may also be present. Fig. 3 illustrated steps to achieve greater, smaller, or different functionality than in Fig. 3 have shown. According to some embodiments, various other embodiments can also be used. Fig. The 3 illustrated steps may occur in a different order, if at all.

[0048] Fig. Figure 4 illustrates a flowchart of an example procedure. 400 for background comparison according to some embodiments of the present disclosure. The method400 This can be done, for example, by one or more processors, the mobile application, or another hardware configuration. For clarity, the procedure will be described below. 400 The following procedures are similar to those described above. 100 until 300 described how it is executed through the mobile application. The procedure 400 However, as with the procedures 100 until 300 be carried out by other hardware components or combinations of hardware components. Aspects of Fig. 4. The application program of the mobile application is shown. According to some embodiments, the method is 400 to perform a sub-process of the step 150 from Fig. 1.

[0049] In step 410 The mobile application retrieves the multiple sets of photos from the captured images (e.g., the sets of captured images in step 1). 130 from Fig. 1) from the user's system. In step 420 Each individual image in each corresponding set of photos is then subjected to a security check. To rule out forgeries and digitally manipulated images, the mobile application uses an interchangeable image data format (EXIF) (e.g., capture time, with or without flash, exposure time, focal length, location, aperture) and extracts the corresponding geocode and timestamp. The mobile application then checks whether the geocode of both images in the set of images falls within a predefined coordinate range for the step described in [the previous step]. 260 from Fig. The specified location is located at the set of images with identical and nearly identical areas (depending on the area) are forwarded for watermark verification and timestamp processing. Each image captured by the user unit is assigned a watermark indicating the time the photo was taken. Each timestamp is converted from a string to a date value and compared. Images within a given set of captured images must meet a predefined time threshold within which the first and second images were captured. For example, the user must capture the first image (e.g., the photo of the user along with the background) and the second image (e.g., the photo of just the background) within two minutes of each other.By setting time parameters, the mobile application reduces the possibility of fraud by forgers attempting to gain access to an account from a location other than the one specified. Image sets that have passed the security check are then processed in step [number missing]. 430 forwarded.

[0050] In step 430 The image sets that have passed the security check will be used in step 420The second image is received, and only the background of the second image is compared with the background of the second image. According to embodiments, various approaches to background comparison can be used. One embodiment may employ scale-invariant feature transformation (SIFT). Using SIFT, local feature descriptors (e.g., shapes, objects) are extracted from the background of the two images contained in the image sets and compared with identical reference objects. A reference object in the second image is identified by comparing each feature from the first image to find matching features based on the Euclidean distances of feature vectors to the object. Reference objects are filtered from the entire set of matches; objects with identical location, identical size, and identical orientation indicate a successful match.The identification of consistent clusters is performed quickly using a powerful implementation of the generalized Hough transform's hash table. Clusters of three or more features matching an object undergo background checks. Finally, the probability that a set of feature objects indicates the presence of an object is computationally compared to a threshold value.

[0051] Alternative implementations can involve comparisons based on image sections. Using alternative facial recognition methods, the mobile application detects the user's face in the first image and predicts image sections that contain both the user's face and body. Image sections represent a group of pixels in an image. For other aspects of the image, pixel positions differing from the extracted facial section are found in the second image, and similarity ratings are assigned to these positions.

[0052] In step 440The background pixels are compared against a predefined, data-driven threshold. When the number of matches between the background pixels in the first and second photos reaches the threshold, the background check is complete, and processing of the released image set continues in step... 450 A message (e.g., a notice, a message, a prompt) confirming background matching is displayed to the user interface of the user unit. Sets of captured images that do not meet the background matching threshold cause the mobile application to display an error message (e.g., a notice, a message, a prompt) to the user interface of a user unit, and the permission is revoked in step 460Rejected. According to some implementations, processing is terminated in the system's application program.

[0053] Fig. Figure 4 presents exemplary steps of an exemplary background comparison method according to some embodiments of the present disclosure. However, according to some embodiments, individual steps may be more or less complex than described in Figure 4. Fig. 4 shown, and in addition to those in Fig. The four steps shown (or their place) may be supplemented by further steps. According to some embodiments, various steps may be included. Fig. The four illustrated steps have greater, lesser, or different functionality than in the original text. Fig. 4 shown. According to some embodiments, various in Fig. The 4 illustrated steps may occur in a different order, if at all.

[0054] In Fig. 5 is a flowchart of an exemplary procedure 500 to illustrate the comparison of users according to some embodiments of the present disclosure. The method 500 This can be accomplished, for example, by one or more processors, a mobile application, or another hardware configuration. For clarity, the procedure will be described below. 500 The following describes how this is done using the mobile application. As above, with reference to the procedures. 100 until 400 Discussed, aspects of the procedure 500 However, this can be achieved through other hardware components or combinations of hardware components. Aspects of Fig. 5 show the application program of the mobile application. According to some embodiments, the method is 500 to perform a sub-process of the step 160 from Fig. 1.

[0055] In step 510The mobile application retrieves the shared set of images from step 450 in Fig. 4 and extracts the first photo that shows the user and the current background. In step 520 The mobile application retrieves the holder's photo from the associated ID card and starts the user's facial recognition.

[0056] In step 530 The user's identity will be verified on both sides in step 510 and 520 The retrieved photos were compared. Different methods can be used to compare each user.

[0057] According to some embodiments, neural networks can be used for deep learning. Neural networks represent a class of multilayered artificial neural networks that have been successfully applied to image analysis. They consist of artificial or neuron-like structures with learnable weights and tendencies. Each neuron receives input values ​​and forms a scalar product. In artificial learning, neurons possess the basic computing units of neural networks. According to embodiments, during the forward pass (e.g., when an image is sent through the network), each filter is convolved over the volume of the receive field of an input image. Then, a scalar product is formed from the filter's input values ​​and the input image, resulting in a two-dimensional feature mapping of that filter.Accordingly, the neural network learns the filters that are activated when it detects a specific feature at a particular spatial position in the input image. The feature mapping of the input user image can then be processed by further network layers to identify matching features between the two input images.

[0058] According to some embodiments, scale-invariant feature transformation (SIFT) can also be used here. Using SIFT, local feature vectors (e.g., eyebrows, nose, chin) are extracted from the photograph on the ID card and combined with identical reference vectors of facial features from the majority of the first photos in the set of images taken in step 1. 510 and 520 from Fig. Five matches are compared. From the entire set of matches, reference vectors of facial features are filtered out; vectors where identical locations, proportions, and orientations coincide show a positive match. The identification of consistent clusters is performed quickly using a powerful implementation of a hash table of the generalized Hough transform.

[0059] According to some embodiments, a more conventional approach to facial verification can be used. Using self-portraits, which are characteristic of the variations between faces of the same user, the first photos from the image set can be selected in step 1. 510 and 520 from Fig. Data from the network database is extracted, and a linear combination is calculated from it. Weights of the user's own face in each photo from the majority of the sets of images are used to determine whether the user matches the photo of the holder on the corresponding ID card.

[0060] In step 540 The matched user weights are compared to a predefined, data-driven threshold. If the facial weights contained in both the first photo and the photo on the ID card reach the similarity threshold, facial recognition and processing are complete, and the process proceeds to step 1. 550 A message (e.g., a notification, a message, a request) regarding the granted authorization is displayed on the user interface of the user unit. A message confirming successful verification may contain details about a newly created account for the user.

[0061] Sets of captured images that do not reach the user verification threshold may cause the mobile application to display an error message (e.g., a notice, message, or prompt) to the user interface, and the permission will be revoked in step 560 Rejected. According to the implementation, processing is terminated in the system's application program.

[0062] In Fig. Figure 5 shows exemplary steps of an exemplary method for comparing users according to some embodiments of the present disclosure. However, according to some embodiments, individual steps may be more or less complex than shown. Fig. 5 shown, and in addition to those shown in Fig. In addition to the 5 steps shown (or in place of them), further steps may be present. According to some embodiments, various other steps may also be included. Fig. 5 illustrated steps to have greater, lesser, or different functionality than in Fig. 5 shown. According to some embodiments, various in Fig. The 5 illustrated steps may occur in a different order, if at all.

[0063] In Fig. 6 is a flowchart of an exemplary procedure 600 The method for error correction according to some embodiments of the present disclosure is illustrated. 600 This can be accomplished, for example, by one or more processors, a mobile application, or another hardware configuration. For clarity, the procedure will be described below. 600 The following describes how this is done using the mobile application. As above, with reference to the procedures. 100 until 500 Discussed, aspects of the procedure 600However, they can be executed by other hardware components or combinations of hardware components. Aspects of Fig. 6 show the application program of the mobile application. According to some embodiments, the method represents 600 a subprogram of the step 330 from Fig. 3 of the step 440 from Fig. 4 and of the step 550 from Fig. 5 dar.

[0064] In step 610A message (e.g., a notification, a message, a request) is sent to a user to receive an error correction message. According to embodiments, an error correction message may contain the reason for the error correction, which may include, but is not limited to, failed signature verification, failed background check, and / or failed user verification. A message may be received by the user via a notification, for example, by clicking on the message on the user interface, but is not limited to.

[0065] After receiving the message, the mobile application accesses the capabilities contained in the user device based on mobile data. In step 620A real-time video call is initiated on the mobile user device to start an online video session between the user and a designated representative of the organization where the user wishes to open an account. During the video session, the user can open and set up an account. Through the video session, the user can open and set up the account by, but not limited to, providing the representative with documents, answering security questions, and following alternative procedures in a one-on-one conversation.

[0066] Fig. Section 6 presents exemplary steps of an exemplary method for predicting future typical signatures according to some embodiments of the present disclosure. However, according to some embodiments, individual steps may be more or less complex than described above. Fig. 6 shown, and in addition to those in Fig. The six steps shown (or their place) may be supplemented by further steps. According to some embodiments, various steps may be included. Fig. The 6 illustrated steps also have a more or less extensive or different functionality than in Fig. 6 shown. According to some embodiments, various in Fig. The 6 illustrated steps may also occur in a different order, if at all.

[0067] Fig. Figure 7 illustrates a block diagram of an example user unit. 700 according to some embodiments of the present disclosure. According to some embodiments, the user unit can 700 a mobile unit or mobile computer and for implementing embodiments of one or more methods 100 until 600 be set up, which above with reference to the Fig. 1 to Fig. 6 were discussed.

[0068] The embodiment of the in Fig. 7 user units shown 700 contains a working memory 725 , a mass storage device 730 , a connecting line (e.g. BUS) 720 , one or more CPUs 705 (also referred to here as processors) 705 (designated as), an I / O unit interface 710 , I / O units 712 and a network interface 715 The RAM 725 can instructions 726 exhibit the instructions 726 can the signature comparison 727 , the background comparison 728 and the user comparison 729 serve. The mass storage device 730 can issue an ID 760 and a set of recorded photos 765 exhibit. The user unit 700 can also be used with a network 735 be connected. Additionally, the user unit can 700on a camera included in the user unit (I / O units) 712 access.

[0069] Each CPU 705 calls into the working memory 725 or the mass storage 730 It retrieves stored program instructions and executes them. The connecting line 720 It is used to transport data, such as program instructions, between the CPUs. 705 , the I / O unit interface 710 , the mass storage 730 , the network interface 715 and the RAM 725 The connecting line 720 This can be implemented using one or more buses. Regarding CPUs... 705 It can be a single CPU, multiple CPUs, or a single CPU with multiple processor cores, depending on the specific embodiment. According to some embodiments, a CPU can be... 705This involves a digital signal processor (DSP). According to some embodiments, the CPU contains 705 one or more integrated 3D circuits (3DICs) (e.g., 3D encapsulation at the wafer level (3DWLP), 3D integration with interlayer, stacked 3D ICs (3D-SICs), monolithic 3D ICs, heterogeneous 3D integration, 3D system-in-package (3DSiP) and / or package-on-package CPU configurations (PoP)). The main memory 725 is always included and represents a random access memory (e.g., static random access memory (SRAM), dynamic random access memory (DRAM), or flash memory). Mass storage 730 is always included and represents the cloud storage or other units that connect via the I / O unit interface. 710 with the user unit 700 or via the network interface 715 with a network 735 are connected. According to various embodiments, the network can 735It should be set up as a database in which currently used data is stored.

[0070] According to some embodiments, the working memory 725 instructions 726 and in the mass storage 730 Data used on the input side is stored. The data used on the input side can be used to identify an individual. 760 and a set of recorded photos 765 exhibit. According to embodiments, the mass storage device can be 730 This involves a data repository located in multiple backed-up source storage systems that are geographically spread across multiple units 712 are distributed across the network. The data to be used on the input side is transmitted via the network. 435 the database 450 extracted. The instructions 726 start the comparison with threshold values ​​and in particular the signature comparison 727 , the background comparison 728and the user comparison 729 .

[0071] The comparison of signatures 727 can be in accordance with step 140 from Fig. 1. The comparison of the backgrounds 728 can be in accordance with step 150 from Fig. 1. The comparison of users 729 can be in accordance with step 160 from Fig. 1. Once these steps are completed, the CPU will execute a procedure according to the Fig. 1 to Fig. 6 initiated.

[0072] The database 750 It can display analysis programs and digital data from a multiple ID card databases.

[0073] According to various embodiments, the I / O units contain 712a surface for displaying data and receiving input data (e.g., a user interface, such as a screen like a touchscreen, pointing devices, speakers, etc.).

[0074] Fig. Section 7 presents exemplary components of an exemplary user unit 700 according to embodiments of the present disclosure. However, according to some embodiments, individual components may be more or less complex than in Fig. 7 shown, there may be others besides those shown. Fig. The 7 components shown, or additional components, may be present. Depending on the components, various options may be available. Fig. The 7 illustrated components have a more or less extensive or different functionality than in Fig. 7 shown.

[0075] Fig. Figure 8 shows a block diagram of an example data processing environment. 800, in which embodiments of the present disclosure can be realized. According to some embodiments, aspects of the data processing environment can be 800 in one or more of the Fig. 1 to Fig. The procedures described in section 6 can be carried out. According to the embodiments, the data processing environment can be configured as follows: 800 a remote system 802 and a host unit 812 contain.

[0076] According to embodiments, the host unit can be 812 and the distant system 802 These are data processing systems. The remote system 802 and the host unit 812 can include one or more processors 806 or 814 and one or more RAM modules 808 or 818 included. The remote system 802 and the distant unity 812can be configured to use an internal or external network interface 804 and one or more data transmission connections 820 (e.g., modems or interface cards) can exchange data with each other. The remote system 802 and / or the host unit 812 They can be equipped with a screen or monitor. Additionally, the remote system can be used. 802 and / or the host unit 812 Any input devices (e.g., a keyboard, a mouse, a scanner, or other input device) and / or any commercially available or specialized software (e.g., browser software, data transmission software, server software, speech processing software, search engine and / or internet search software, filter modules for filtering content based on predefined parameters, etc.). According to some embodiments, the remote system may be... 802and / or the host unit 812 This includes servers, desktop computers, laptop computers, or handheld units.

[0077] The remote system 802 and the host unit 812 can be geographically separated and connected via a network 840 exchanging data with each other. The network 840 can be used with the network 735 from Fig. 7 agree. According to embodiments, the host unit may be 812 This involves a data center from which a remote system 802 and other (not shown) remote units can establish a data transmission connection, for example in a client-server network model. According to some embodiments, the host unit can 812 and the distant system 802 be set up in any other suitable network relationship (e.g., in a peer-to-peer arrangement or using a different network topology).

[0078] According to certain embodiments, the network can 840 This can be achieved using any number of suitable data transmission media. For example, the network could be... 840 This could be a wide area network (WAN), a local area network (LAN), the internet, or an intranet. Depending on certain configurations, the remote system can 802 and the host unit 812 They must be spatially adjacent to each other and exchange data via any suitable local data transmission medium. For example, the remote system 802 and the host unit 812 Data is exchanged using a local area network (LAN), one or more connection lines, a wireless connection, a routing computer, or an intranet. According to some embodiments, the remote system, the host unit, can be... 812and any other units may be connected to each other using a combination of one or more networks and / or one or more local connections to exchange data. For example, the remote system 802 via a cable (e.g. via an Ethernet cable) to the host unit 812 be connected, and a second (not shown) unit can be connected using the network. 840 (e.g., via the Internet) exchange data with the host unit.

[0079] According to some embodiments, the network can 840This can be implemented within a cloud data processing environment or using one or more cloud data processing services. Depending on the specific implementation, a cloud data processing environment may include a network-based distributed data processing system that provides one or more cloud data processing services. Furthermore, a cloud data processing environment may comprise many computers (e.g., hundreds, thousands, or more) located within one or more data centers and sharing resources over the network. 840 are set up.

[0080] According to some embodiments, users can use the remote system 802 be enabled to access the host unit 812 to provide image analysis. According to some embodiments, the host unit can 812 directly one or more input units 824and one or more output units 826 included. The host unit 812 can include subcomponents such as a data processing environment 830 Included. The data processing environment 830 can a processing unit 814 , a graphics processing unit 816 and a RAM 818 included. The RAM 818 gives instructions 320 on. According to embodiments, the instructions can 820 with the instructions 726 from Fig. 7 agree. The data processing environment 830 can be set up so that they receive the data from the remote system 802 adopted content 810 processed.

[0081] The storage 828 can be set up to store training data in relation to weighted data and use the working memory 818 be connected. The storage 828can be used with the mass storage 730 from Fig. 7 match.

[0082] Fig. Figure 8 illustrates a data processing environment 800 with a single host unit 812 and a single remote system 802 However, suitable data processing environments for implementing embodiments of the present disclosure can include any number of remote units and host units. The various in Fig. The 8 illustrated models, modules, systems, instructions and components may be distributed across multiple host units and further units.

[0083] It is pointed out that in Fig. 8 representative components of an exemplary data processing environment 800 are to be represented. According to some embodiments, individual components can be more or less complex than in Fig. 8 shown, furthermore, other components than in Fig. 8 shown or additional components may be present, and the number, type and arrangement of such components may vary.

[0084] A computer-readable storage medium can be a physical unit capable of retaining and storing instructions for use by a system to execute instructions. For example, a computer-readable storage medium can be an electronic storage unit, a magnetic storage unit, an optical storage unit, an electromagnetic storage unit, a semiconductor storage unit, or any suitable combination thereof, without limitation. A non-exhaustive list of more specific examples of computer-readable storage media includes the following: a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), and erasable programmable read-only memory (EPROM).Flash memory), static random-access memory (SRAM), portable compact storage disk-read-only memory (CD-ROM), DVD (digital versatile disc), USB flash drive, floppy disk, a mechanically coded unit such as punched cards or raised structures in a groove on which instructions are stored, and any suitable combination thereof. A computer-readable storage medium shall not, in its use herein, be understood as volatile signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission medium (e.g., light pulses traveling through an optical fiber cable), or electrical signals transmitted by a wire.

[0085] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to the respective data processing units or, via a network such as the Internet, a local area network, a wide area network, and / or a wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission lines, wireless transmission, routing computers, firewalls, switching units, gateway computers, and / or edge servers. A network adapter card or network interface in each data processing unit receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage on a computer-readable storage medium within the respective data processing unit.

[0086] Computer-readable program instructions for executing the steps of the present invention can be assembly instructions, ISA (Instruction Set Architecture) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., as well as conventional procedural programming languages ​​such as C or similar languages. The computer-readable program instructions can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server.In the latter case, the remotely located computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be established with an external computer (for example, via the internet using an internet service provider). In some embodiments, electronic circuits, including, for example, programmable logic circuits, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), can execute the computer-readable program instructions by using state information from the computer-readable program instructions to personalize the electronic circuits to perform aspects of the present invention.

[0087] Aspects of the present invention are described herein with reference to flowcharts and / or block diagrams or charts of methods, devices (systems), and computer program products according to embodiments of the invention. It is pointed out that each block of the flowcharts and / or block diagrams or charts, as well as combinations of blocks in the flowcharts and / or block diagrams or charts, can be executed by means of computer-readable program instructions.

[0088] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a specialized computer, or another programmable data processing device to create a machine, such that the instructions executed via the processor of the computer or other programmable data processing device generate a means of implementing the functions / steps specified in the block(s) of the flowcharts and / or block diagrams or charts.These computer-readable program instructions may also be stored on a computer-readable storage medium capable of controlling a computer, programmable data processing device, and / or other units to function in a particular manner, such that the computer-readable storage medium on which instructions are stored has a manufactured product, including instructions that implement aspects of the function / step specified in the block(s) of the flowchart and / or block diagrams or charts.

[0089] The computer-readable program instructions can also be loaded onto a computer, other programmable data processing device, or other unit to cause the execution of a series of process steps on the computer or other programmable device or other unit in order to generate a process executed on a computer, such that the instructions executed on the computer, other programmable device, or other unit implement the functions / steps specified in the block(s) of the flowcharts and / or block diagrams or charts.

[0090] The flowcharts and block diagrams or charts in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this context, each block in the flowcharts or block diagrams or charts can represent a module, segment, or part of instructions that includes one or more executable instructions for performing the specific logical function(s). In some alternative embodiments, the functions specified in the block may occur in a different order than shown in the figures. For example, two blocks shown consecutively may in reality be executed essentially simultaneously, or the blocks may sometimes be executed in reverse order depending on the corresponding functionality.It should also be noted that each block of the block diagrams or charts and / or flowcharts, as well as combinations of blocks in the block diagrams or charts and / or flowcharts, can be implemented by special hardware-based systems that perform the specified functions or steps, or execute combinations of special hardware and computer instructions.

[0091] While it is clear that the process software can be installed on the client, server, and proxy computers by directly loading it from a storage medium such as a CD, DVD, etc., it can also be installed automatically or semi-automatically by sending the process software to a central server or group of central servers. The process software is then downloaded to the client computers, which run the process software. Alternatively, the process software can be sent directly to the client system via email. The process software is then either placed in a directory or loaded into a directory by executing a set of program instructions that place the process software in a directory. Another alternative is to send the process software directly to a directory on the client computer's hard drive.If proxy servers are available, the process selects the proxy server code, determines on which computers the proxy server code should be installed, transfers the proxy server code, and then installs the proxy server code on the proxy computer. The process software is transferred to the proxy server and then stored on the proxy server.

[0092] Embodiments of the present invention can also be provided as part of a service agreement with a client company, a non-profit organization, a government institution, an international organizational structure, or the like. In these embodiments, a computer system can be configured to run and install software, hardware, and internet services that implement some or all of the methods described herein.Within the scope of these embodiments, work processes at the customer's site can also be analyzed, recommendations developed in response to the analysis, systems set up to implement subsets of the recommendations, the systems integrated into existing processes and infrastructure, the systems used for consumption recording, costs distributed among users of the systems and billing carried out, invoices issued or payments otherwise received for the use of the systems.

Claims

[1] Computer-aided method for using image analysis techniques to validate the identity of a remote user, wherein the method comprises: Retrieving a photo of the user and a user's signature from an associated digital ID; Requesting a set of photos taken in real time, wherein the set of photos taken in real time includes a first photo and a second photo; Requesting a signature provided in real time along with the requested set of photos taken in real time; Comparison of the signature provided in real time on the first photo with the user's signature on the associated digital ID; Comparison of a background in the first photo of the real-time set of photos with a background in the second photo of the real-time set of photos; Comparison of an image of the user in the first photo of the real-time captured set of photos with the photo of the user on the associated digital ID; and Validation of the user's identity in response to the finding that the real-time signature on the first photo of the set of photos, compared to the user's signature on the retrieved associated digital ID, reaches a first threshold; that the background on the first photo of the real-time set of photos, compared to the background on the second photo of the real-time set of photos, reaches a second threshold; and that the user's image on the first photo of the real-time set of photos, compared to the user's image on the associated digital ID, reaches a third threshold. [2] The method of claim 1, further comprising rejecting the validation of the user's identity in response to the finding that: a number of signature matches is less than the first threshold; and / or a number of background matches is less than the second threshold; and / or a number of user matches is less than the third threshold. [3] Method according to claim 2, wherein the rejection of the validation of the user's identity is accompanied by the initiation of an online video session. [4] Method according to claim 1, wherein the first photo is an uploaded photo of the user together with the background and the second photo is an uploaded photo of the background alone. [5] Method according to claim 1, wherein the comparison of the background on the first photograph of the set of photographs taken in real time with the background on the second photograph of the set of photographs taken in real time further comprises a verification that a first timestamp and a first date of the image watermark date of the first photograph is within a threshold of a second timestamp and a second date of the image watermark date of the second photograph determined by the system. [6] The method of claim 1, further comprising: Analyzing the first and second photos of the real-time captured set of photos to determine whether they meet image quality requirements regarding blurriness, overexposure, and / or additions; and Issuing a prompt to the user indicating that a newly captured image should be uploaded, in response to the finding that the first photo and / or the second photo of the real-time captured set of photos does not meet the image quality requirements. [7] The method of claim 1, further comprising: Retrieving syntax data from the associated digital ID card; Receiving input data that contains a large amount of the user's personal information; and Verify that the syntax data matches the input data; wherein the syntax data and the majority of the user's personal information include at least an identification number, first and last name, place and date of birth, gender, marital status, religion, blood type, address, occupation and nationality of the user. [8] Computer-aided method for using image analysis techniques to validate the identity of a user, wherein the method comprises: Receiving a request to open an account for a user; in response to receiving the request, retrieving a digital copy of an identification document for the user, wherein the identification document contains a photograph of the user and a signature of the user; Providing a prompt to request an initial real-time photograph of the user along with a view of the user's surroundings; Providing a prompt to request a second real-time photograph of the user's surroundings without the user being present; Providing a prompt to request the user's current location; Verifying the user's current location in response to the finding that the user's surroundings in the first real-time photograph match the user's surroundings in the second real-time photograph; and Validating the user's identity, at least partially, based on verifying the user's current location. [9] Method according to claim 8, wherein the request to request the first photo taken in real time includes a request to request a first user signature, wherein the request to request the second photo taken in real time includes a request to request a second user signature. [10] The method of claim 9, further comprising: Comparing the user's photo on the digital copy of the ID card with the first photo of the user taken in real time; Comparing the user's signature on the digital copy of the ID card with the first user signature; Comparing the user's signature on the digital copy of the ID card with the second user signature; and where validating the user's identity is further based on the finding that the user's photograph on the digital copy of the ID card matches the first photograph of the user taken in real time, and on the finding that the user's signature on the digital copy of the ID card matches the first user signature and the second user signature. [11] System comprising means suitable for carrying out all of the steps of the method according to a preceding claim of the method. [12] Computer program comprising instructions for performing all of the steps of the method according to a preceding claim of the method when the computer program is executed on a computer system.