Web-based identification card copy determination system

Abstract

A web-based identification card copy determination system according to an embodiment of the disclosed invention may comprise a first artificial intelligence model, and a processor configured to determine whether an identification card is authentic, wherein: the first artificial intelligence model and the processor are configured to be downloaded in a manner of being transmitted from a web server to a user terminal when the user terminal accesses an identification card authenticity determination website; the first artificial intelligence model and the processor which are downloaded by the user terminal are stored in a cache memory that is a temporary storage for storing data of a web page accessed by the user terminal; and the processor stored in the cache memory is configured to, on the basis of an inspection target image which includes an identification card type object and is acquired by a camera of the user terminal, determine whether the identification card type object included in the inspection target image is a real identification card by using the first artificial intelligence model stored in the cache memory.

Description

Web-based ID copy verification system

[0001] The present invention relates to a web-based identification card copy determination system that can check whether an identification card is a copy or not on a web basis.

[0002] Technology that determines whether an ID is a copy verifies the validity of documents containing ID or personal identification information and detects forged or altered copies. This copy detection technology plays a crucial role in various fields, including digital security, identity verification, and authentication at airlines and financial institutions.

[0003] Recent technologies for determining whether an ID is a copy apply machine learning and deep learning techniques to learn and identify patterns in the document, allowing for more accurate and faster detection of forged copies.

[0004] While these technologies continue to evolve to enhance security and privacy and prevent fraud and forgery, they also face ethical and legal issues related to privacy.

[0005] The present invention provides a web-based ID copy determination system capable of performing copy determination of an ID card on a web page through a terminal's website access, and a control method of the web-based ID copy determination system.

[0006] In addition, the present invention provides a web-based ID card copy determination system and a control method of the web-based ID card copy determination system that can perform copy determination on an ID card without the need for a terminal to download an application or transmit an image to be analyzed to a server.

[0007] In addition, the present invention provides a web-based ID copy verification system and a control method of the web-based ID copy verification system that eliminates the user's resistance to downloading an application and thus causes no unnecessary waste of terminal memory, as the user does not need to download an application.

[0008] In addition, the present invention provides a web-based ID card copy determination system and a control method of the web-based ID card copy determination system that can utilize an automatic shooting function to automatically capture a real-time image that matches the ID card form by having the terminal's camera capture the image of the target of copy determination without installing an application.

[0009] In addition, the present invention provides a web-based ID card copy determination system and a control method of the web-based ID card copy determination system that can improve system performance, such as increasing data processing speed or reducing required memory capacity, compared to a conventional copy determination system.

[0010] A web-based ID copy verification system according to one aspect of the disclosed invention comprises: a first artificial intelligence model; and a processor configured to determine whether an ID is genuine, wherein the first artificial intelligence model and the processor are configured to be downloaded from a web server to a user terminal when the user terminal accesses an ID authenticity judgment website, and the first artificial intelligence model and the processor downloaded by the user terminal are stored in a cache memory, which is a temporary storage for storing data of a web page accessed by the user terminal, and the processor stored in the cache memory can be configured to determine whether an ID format object included in the inspection target image is a real ID, using the first artificial intelligence model stored in the cache memory, based on an inspection target image including an ID format object and acquired by a camera of the user terminal.

[0011] In addition, the system further includes a machine learning module configured to train the first artificial intelligence model through a machine learning method by setting a learning image containing an ID card-format object as an input variable and setting learning image classification information, which is information on whether the ID card-format object contained in the learning image is the actual ID card, as an output variable, wherein each of the learning images may be one of a learning real ID card image containing the actual ID card, a learning screen display image in which the ID card is displayed on a screen, and a learning ID card copy image which is an image in which a copy of the ID card is photographed.

[0012] In addition, the processor stored in the cache memory can: receive a real-time image acquired by a camera of the user terminal; determine whether the real-time image is an image including the ID card-type object; and if the real-time image is determined to be an image including the ID card-type object, control the camera to perform image capturing to acquire the real-time image as the inspection target image.

[0013] In addition, the processor stored in the cache memory may be configured to: extract text included in the real-time image through a second artificial intelligence model that is pre-trained to perform optical character recognition (OCR) based on the real-time image; determine whether the extracted text is text that satisfies the identification card component based on the content of the extracted text and the location in the real-time image; and determine whether the real-time image is the inspection target image based on whether the extracted text is text that satisfies the identification card component.

[0014] Additionally, the first artificial intelligence model and processor stored in the cache memory may be configured to be removed and deleted from the cache memory when it is determined by the processor whether the ID card-type object included in the inspection target image is a real ID card.

[0015] In addition, the server first artificial intelligence model configured to operate identically to the first artificial intelligence model and provided on the web server; and a server processor provided on the web server and configured to determine whether the identification card is authentic, wherein the server processor determines whether to perform the identification card authenticity determination on the web server based on authentication performing subject information transmitted from the user terminal to the communication module of the web server, and the authentication performing subject information may be information on which of the processor stored in the cache memory and the server processor is the subject for determining the identification card's authenticity.

[0016] In addition, the server processor may be configured to: when receiving authentication performing entity information including information indicating that the web server has decided to perform a determination on the authenticity of the ID card, control the communication module of the web server to receive an inspection target image acquired by the camera of the user terminal from the user terminal; and, based on the inspection target image, determine, using the server first artificial intelligence model, whether an ID card-type object included in the inspection target image is a real ID card.

[0017] In addition, the server processor: when it is determined that the determination of authenticity of the ID card is to be performed on the web server, generates information of a shooting auxiliary screen including information to directly perform shooting while including a real-time image acquired in real time by a camera of the user terminal; controls the communication module to transmit the information of the shooting auxiliary screen to the user terminal so that the shooting auxiliary screen is displayed on the display of the user terminal; and when the camera of the user terminal performs image shooting according to a user's shooting command input to the user terminal, controls the communication module to receive the captured real-time image from the user terminal.

[0018] In addition, the server processor may be configured to: receive a real-time image captured by a camera of the user terminal from the communication module; extract text included in the real-time image through a second artificial intelligence model that is pre-trained to perform optical character recognition (OCR) based on the real-time image; determine whether the extracted text is text that satisfies the elements of an identification card based on the content of the extracted text and the location in the real-time image; and determine the real-time image as an inspection target image acquired by the camera of the user terminal if the extracted text is text that satisfies the elements of an identification card.

[0019] In addition, the processor stored in the cache memory: generates the authentication performing subject information; when the authentication performing subject information including information on the intent to perform the determination of the authenticity of the ID card in the cache memory is generated, the user terminal is controlled to receive the first artificial intelligence model from the web server and store it in the cache memory; when the authentication performing subject information including information on the intent to perform the determination of the authenticity of the ID card in the cache memory is generated, the processor is controlled to determine, based on the inspection target image, whether an ID card-type object included in the inspection target image is a real ID card using the first artificial intelligence model stored in the cache memory; and when the authentication performing subject information including information on the intent to perform the determination of the authenticity of the ID card in the web server is generated, the user terminal is controlled to transmit the generated authentication performing subject information to a communication module of the web server.

[0020] Additionally, the processor stored in the cache memory can generate authentication performing subject information corresponding to a user subject selection command input to the user terminal.

[0021] In addition, the processor stored in the cache memory: measures the server operation time, which is the time elapsed from the time when the authentication performing entity information including information indicating that the web server has decided to perform the determination of the authenticity of the ID card, is transmitted to the communication module of the web server; and if the server operation time exceeds the first reference time, controls the user terminal to receive the first artificial intelligence model from the web server and store it in the cache memory; and if the server operation time exceeds the first reference time, based on the inspection target image, the first artificial intelligence model stored in the cache memory can be used to determine whether the ID card-type object included in the inspection target image is a real ID card.

[0022] In addition, the processor stored in the cache memory: measures a cache operation time, which is a time elapsed from the time of generating authentication performing subject information including information on the intent to perform the determination of authenticity of the ID card in the cache memory; when the cache operation time exceeds a second reference time, generates authentication performing subject information including information on the intent to perform the determination of authenticity of the ID card in the web server; and when the cache operation time exceeds the second reference time, controls the user terminal to transmit authentication performing subject information including information on the intent to perform the determination of authenticity of the ID card in the web server to a communication module of the web server.

[0023] In addition, the processor stored in the cache memory may be configured to: determine whether the first artificial intelligence model can be stored in the cache memory based on the capacity of the cache memory; and, if it is determined that it is impossible to store the first artificial intelligence model in the cache memory, generate authentication performing entity information including information indicating that the web server has decided to perform the determination of authenticity of the identification card.

[0024] In addition, the server processor: when a user terminal accesses an ID card authenticity determination website, generates information on an application download selection screen that includes a question asking the user whether to download an application provided by an operator of the ID card authenticity determination website; controls the communication module to transmit information on the application download selection screen to the user terminal so that the application download selection screen is displayed on a display of the user terminal; and the processor stored in the cache memory: when the user terminal receives an application installation command from a user, controls the user terminal to download data of the application from the web server; when the user terminal receives an installation rejection command from a user, controls the user terminal to receive the first artificial intelligence model from the web server and store it in the cache memory; And when the user terminal receives an installation rejection command from the user, based on the inspection target image, the first artificial intelligence model stored in the cache memory is used to determine whether the ID card-type object included in the inspection target image is a real ID card, and the data of the application may include an application first artificial intelligence model configured to operate identically to the first artificial intelligence model and an application processor configured to determine whether the ID card is genuine.

[0025] In addition, the processor stored in the cache memory may be configured to: determine whether the application can be installed based on the memory capacity of the user terminal; if it is determined that downloading the application to the user terminal is impossible, control the user terminal to receive the first artificial intelligence model from the web server and store it in the cache memory; and if it is determined that downloading the application to the user terminal is impossible, determine whether the ID card-type object included in the inspection target image is a real ID card using the first artificial intelligence model stored in the cache memory.

[0026] In addition, the processor stored in the cache memory may be configured to: calculate a data transmission / reception speed between the user terminal and the communication module of the web server; if the data transmission / reception speed is equal to or greater than a first reference speed, generate authentication performing entity information including information indicating that the web server has decided to perform the determination of authenticity of the ID card, and control the user terminal to transmit the generated authentication performing entity information to the communication module of the web server; if the data transmission / reception speed is less than the first reference speed and equal to or greater than a second reference speed, control the user terminal to download data of the application from the web server; if the data transmission / reception speed is less than the second reference speed, control the user terminal to receive the first artificial intelligence model from the web server and store it in the cache memory; and if the data transmission / reception speed is less than the second reference speed, determine whether an ID card-type object included in the inspection target image is a real ID card using the first artificial intelligence model stored in the cache memory based on the inspection target image.

[0027] Additionally, the first artificial intelligence model and the processor may be configured to be implemented based on web assembly so that the function of the application can be implemented on the ID authenticity determination website.

[0028] A control method of a web-based ID copy verification system according to an aspect of the disclosed invention may include the steps of: when a user terminal accesses an ID authenticity judgment website, a first artificial intelligence model and a processor are downloaded to the user terminal in a manner transmitted from a web server to the user terminal; storing the first artificial intelligence model and processor downloaded by the user terminal in a cache memory, which is a temporary storage for storing data of a web page accessed by the user terminal; receiving, by the processor stored in the cache memory, a real-time image acquired by a camera of the user terminal; determining, by the processor stored in the cache memory, whether the real-time image is an image including an ID card-style object; controlling the camera to perform image capturing so as to acquire the real-time image as an inspection target image including an ID card-style object, if the real-time image is determined by the processor stored in the cache memory to be an image including the ID card-style object; and determining, by the processor stored in the cache memory, whether the ID card-style object included in the inspection target image is a real ID card, using the first artificial intelligence model stored in the cache memory, based on the inspection target image acquired by the camera of the user terminal.

[0029] A non-transitory recording medium according to one aspect of the disclosed invention may store a computer-readable computer program for executing a control method of a web-based ID card copy verification system.

[0030] According to one aspect of the disclosed invention, copy verification of an identification card can be performed on a web page by accessing a website of a terminal.

[0031] Additionally, according to an embodiment of the present invention, a terminal can perform copy determination on an ID card without having to download an application or transmit an image to be analyzed to a server.

[0032] In addition, according to an embodiment of the present invention, since the user does not need to download an application, unnecessary waste of terminal memory is not caused and the user's aversion to downloading an application can be eliminated.

[0033] In addition, according to an embodiment of the present invention, an automatic shooting function can be used to automatically capture a real-time image that matches the ID card form by having the terminal's camera capture the image of the target for copy determination without installing an application.

[0034] In addition, according to an embodiment of the present invention, system performance, such as an increase in data processing speed or a decrease in required memory capacity, can be improved compared to a conventional copy determination system.

[0035] FIG. 1 is a control block diagram of a web-based ID card copy verification system according to one embodiment.

[0036] FIG. 2 is a drawing illustrating a shooting assistance screen according to one embodiment.

[0037] FIG. 3 is a drawing illustrating a screen displayed in a step of determining whether an identification card is authentic according to one embodiment.

[0038] FIG. 4 is a diagram illustrating a screen displayed so that a user can input a subject selection command according to one embodiment.

[0039] FIG. 5 is a drawing illustrating a photographing assistance screen displayed when determining the authenticity of an ID card according to one embodiment.

[0040] FIG. 6 is a drawing illustrating a screen displayed when the determination of authenticity of an identification card is completed according to one embodiment.

[0041] FIG. 7 is a flowchart of a control method of a web-based ID copy verification system according to one embodiment.

[0042] Like reference numerals refer to like elements throughout the specification. This specification does not describe all elements of the embodiments, and general information within the technical field to which the disclosed invention pertains or information that overlaps between the embodiments is omitted.

[0043] Additionally, when a part is said to "include" a component, this does not mean that it excludes other components, but rather that it may include other components, unless otherwise specifically stated.

[0044] As used herein, the term "~unit" refers to a unit that processes at least one function or operation, and may refer to, for example, software, an FPGA, or a hardware component. The function provided by the "~unit" may be performed separately by multiple components, or may be integrated with other additional components. The "~unit" of this specification is not necessarily limited to software or hardware, and may be configured to be located in an addressable storage medium, or may be configured to play one or more processors. According to embodiments, multiple "~units" may be implemented as a single component, or a single "~unit" may include multiple components.

[0045] The terms first, second, etc. are used to distinguish one component from another, and the components are not limited by the aforementioned terms.

[0046] Singular expressions include plural expressions unless the context clearly indicates otherwise.

[0047] The identification codes for each step are used for convenience of explanation and do not describe the order of each step. Each step may be performed in a different order than specified unless the context clearly indicates a specific order.

[0048] The operating principle and embodiments of the disclosed invention will be described with reference to the attached drawings below.

[0049] FIG. 1 is a control block diagram of a web-based ID card copy verification system according to one embodiment.

[0050] A web-based ID copy determination system (1) may be a system configured to perform copy determination of an ID card on a website using data received from a web server (200) that provides information on the website and stored in a cache memory (110) by a user terminal (100) accessing the website. In this case, the ID card to be determined to be a copy or original may be, but is not limited to, a resident registration card, driver's license, passport, alien registration card, credit card, etc.

[0051] Referring to FIG. 1, a web-based ID copy verification system (1) may include a processor, a first artificial intelligence model, a second artificial intelligence model, and a machine learning module (300).

[0052] The processor can determine the authenticity of an ID. That is, the processor can determine whether the ID-shaped object contained in the image being inspected is a genuine original ID or a copy of the ID, such as a photo, a fake, or an image displayed on a screen based on the genuine ID.

[0053] A user may manipulate a user terminal (100) to access an ID card authenticity determination website. The user terminal (100) may be a smartphone used by the user. The ID card authenticity determination website may be a website operated by an operator that provides services requiring user authentication through the website. In this case, the operator of the ID card authenticity determination website may be, but is not limited to, a bank, a credit card company, an insurance company, a securities company, a delivery service provider, etc. If a user needs to authenticate his / her ID card while using the website to use a service provided by the operator, he / she may want to authenticate that his / her ID card is an original ID card and not a copy through the ID card authentication webpage provided by the operator.

[0054] At this time, in the past, in order to authenticate an ID card, there was a method of downloading an application provided by the operating entity onto a smartphone and using an artificial intelligence model included in the downloaded application to determine the authenticity of the ID card. However, this method had the problem that the user had to download the application and caused unnecessary waste of memory on the terminal, which was unpleasant to the user. In addition, there was another method of sending a photo of the ID card to a server operated by the operating entity and using an artificial intelligence model provided on the server of the operating entity to determine the authenticity of the ID card. However, this method inevitably caused data to be transmitted and received to the server, and therefore, there was a problem that the automatic shooting function, in which the camera (120) of the smartphone automatically takes a picture of the real-time image that matches the ID card format and acquires an image of the subject of inspection, could not be used.

[0055] A web-based ID copy verification system (1) may be a system that allows a user to easily analyze an ID copy using an artificial intelligence model through a web page without having to download an application or transmit an image to a server.

[0056] The first artificial intelligence model and processor according to the present invention can be downloaded to the user terminal (100) by being transmitted from the web server (200) to the user terminal (100) when the user terminal (100) accesses the ID card authenticity determination website.

[0057] At this time, the first artificial intelligence model and processor may be downloaded to the user terminal (100) by generating a first artificial intelligence model and processor configured identically to the server first artificial intelligence model (202) and the server processor (201) based on the data of the server first artificial intelligence model (202) and the data of the server processor (201) provided in the web server (200).

[0058] A web server (200) may be a server managed by an operating entity that operates a website. This web server (200) may be a single, large server, or it may be distributed across multiple servers in a large data center. Operating entities that provide large-scale web services may own such data centers or build servers through cloud services to provide services. The web server (200) is managed by the operating entity providing the service. In this case, the operating entity may operate its own data center and maintain a large server farm to provide services, or it may rent necessary server resources through cloud services to operate the services. The web server (200) may be comprised of high-performance computer hardware. Within a data center, hundreds to thousands of servers are installed in racks and connected to a high-bandwidth network. In other words, the web server (200) may store data necessary to operate a web service and provide information upon request from users of the services provided by the operating entity.

[0059] A web server (200) may be provided with a server first artificial intelligence model (202) and a server processor (201). The server first artificial intelligence model (202) may be configured to operate in the same manner as the first artificial intelligence model. The server processor (201) may be provided in the web server (200) and configured to determine the authenticity of an identification card.

[0060] The first artificial intelligence model and processor downloaded by the user terminal (100) can be stored in the cache memory (110).

[0061] The cache memory (110) of the user terminal (100) may be a temporary storage that stores data of a web page accessed by the user terminal (100).

[0062] Specifically, when a user terminal (100) accesses the web, the browser performs various tasks to download and display data of the webpage, and one of the entities performing the role at this time may be a cache memory. The cache memory (110) may be a temporary storage that stores part or all of the data of a previously visited webpage. When the browser of the user terminal (100) visits a webpage, it may store images, style sheets, scripts, etc. of the page in the cache memory (110). At this time, when the same webpage is visited again through the user terminal (100), the browser can load the page quickly using the cached data rather than downloading all new data. The cache memory (110) can usually be managed in the browser settings. If the user has not changed the settings, the browser retains the cached data for a certain period of time, thereby loading the webpage more quickly.

[0063] When a user terminal (100) visits a web page, it requests necessary data from a web server (200), and the web server (200) can provide the data in response. At this time, the data may include HTML documents, images, style sheets, script files, etc. During the process of visiting a web page, the user terminal (100) sends an HTTP request to the web server (200), and the web server (200) transmits data for the request as an HTTP response. At this time, the transmitted data is interpreted and displayed by the browser of the smartphone, and a necessary portion thereof may be stored in the cache memory (110). The data that the user terminal stores in the cache memory (110) may be composed of various resources of the web page received from the web server (200). These resources may include images, scripts, styles, etc. necessary for composing the content of the web page.

[0064] The first artificial intelligence model and processor downloaded by the user terminal (100) may be cached data stored in the cache memory (110).

[0065] The processor (111) stored in the cache memory can determine whether the ID card type object (401) included in the inspection target image (400) is a real ID card, based on the inspection target image (400) and using the first artificial intelligence model (112) stored in the cache memory.

[0066] The ID format object (401) may be an object conforming to the format of a specific ID. For example, if the ID subject to inspection is a driver's license, the ID format object (401) may be a rectangular object in which ID components such as a photo, name, date of birth, and expiration date, which appear on the driver's license, are displayed in positions conforming to a preset license format, and which has a curve formed near the vertex.

[0067] The inspection target image (400) may be an image including an identification card type object (401). In this case, the inspection target image (400) may be an image acquired by a camera (120) of a user terminal (100).

[0068] Meanwhile, the first artificial intelligence model and the second artificial intelligence model may be provided in the cache memory (110) of the user terminal (100), the memory of the web server (200), or the data (220) of the application.

[0069] Each AI model, such as the first AI model and the second AI model, is not necessarily a distinct AI model. For example, the first AI model and the second AI model may be completely identical. Alternatively, each AI model may be expressed differently, but there is only one AI model according to the present invention. A single AI model may perform all of the deep learning operations of the present invention.

[0070] The machine learning module (300) sets a plurality of training images containing ID card format objects (401) as input variables, and sets training image classification information, which is information on whether the ID card format object (401) contained in the training images is a real ID card, as an output variable, so that the first artificial intelligence model can be trained through a machine learning method.

[0071] Each learning image may be one of a learning physical ID image containing the actual ID, a learning screen display image showing the ID displayed on a screen, and a learning ID copy image which is an image of a photographed copy of the ID.

[0072] At this time, the image of the real ID card for learning may have learning image classification information set to indicate that the image is a real ID card image, the image displayed on the learning screen may have learning image classification information set to indicate that the image is an image displayed on the screen or a copy image, and the image of the copy of the ID card for learning may have learning image classification information set to indicate that the image is a copy image.

[0073] The machine learning module (300) can create a second artificial intelligence model through a machine learning method by using multiple learning ID card images as input variables and setting the content of text and location information of text displayed in each learning ID card image as output variables.

[0074] Machine learning utilizes models composed of multiple parameters and can mean optimizing those parameters based on given data. Depending on the type of learning problem, machine learning can include supervised learning, unsupervised learning, and reinforcement learning. Supervised learning learns mappings between inputs and outputs and is applicable when input-output pairs are given as data. Unsupervised learning is applicable when there are only inputs and no outputs, and can identify patterns between inputs, etc.

[0075] The machine learning module (300) can generate an artificial intelligence model in various ways. For example, the machine learning module (300) can learn features extracted from training data using a deep learning-based learning method. At this time, a CNN (Convolutional Neural Networks) structure that stacks multiple stages of convolution layers can be utilized to learn a method for extracting features from training data. However, the learning method of the machine learning module (300) is not necessarily limited to a method utilizing the CNN structure. For example, the learning method of the machine learning module (300) can be a method through a machine learning algorithm including an artificial neural network (ANN) or a recurrent neural network (RNN).

[0076] FIG. 2 is a drawing illustrating a shooting assistance screen according to one embodiment, and FIG. 3 is a drawing illustrating a screen displayed in a step of determining whether an identification card is authentic according to one embodiment.

[0077] Referring to FIGS. 1, 2 and 3, the web-based ID copy determination system (1) can automatically acquire one real-time image as an analysis target image when an object included in an image being captured in real time by a camera (120) is an ID shape object that conforms to the ID form.

[0078] The processor (111) stored in the cache memory can receive real-time images acquired by the camera (120) of the user terminal (100) from the camera (120).

[0079] The processor (111) stored in the cache memory can determine whether the real-time image is an image containing an ID card type object (401).

[0080] The processor (111) stored in the cache memory can control the camera (120) to perform image capturing to acquire the real-time image as an inspection target image (400) if the real-time image is determined to be an image including an ID card type object (401).

[0081] Specifically, the processor (111) stored in the cache memory can extract text included in the real-time image through the second artificial intelligence model (113) stored in the cache memory that has been pre-trained to perform optical character recognition (OCR) based on the real-time image.

[0082] The processor (111) stored in the cache memory can determine whether the extracted text is text that satisfies the identification component based on the content of the extracted text and its location in the real-time image.

[0083] The processor (111) stored in the cache memory can determine whether the real-time image is an image to be inspected (400) based on whether the extracted text is text that satisfies the identification component.

[0084] Optical Character Recognition (OCR), utilized by the second AI model, may be a technology that detects and recognizes characters in images or scanned documents and converts them into text. The OCR process may consist of the following stages: image acquisition, preprocessing, character detection, character recognition, and postprocessing.

[0085] In the image acquisition stage, the target real-time image can be captured in digital form. In the preprocessing stage, image resizing, rotation correction, noise removal, and contrast enhancement can be performed. In the character detection stage, the location of characters can be detected in the preprocessed image. This process primarily uses computer vision technology to identify and segment character boundaries at the pixel level. In the character recognition stage, actual text can be extracted from each character region. This stage can be converted from image to text using artificial neural networks, machine learning algorithms, or rule-based methods included in the second artificial intelligence model. In the postprocessing stage, the OCR results can be refined and errors corrected.

[0086] FIG. 4 is a diagram illustrating a screen displayed so that a user can input a subject selection command according to one embodiment.

[0087] Referring to FIG. 4, the user may input a command to determine the type of identification card to be inspected according to his / her choice, and may also input a command to determine whether the entity performing the identification card copy determination will be a processor and artificial intelligence model downloaded to the cache or a processor and artificial intelligence model provided on the web server.

[0088] For example, a user can input an input to select the entity that automatically performs identification copy verification by determining the performance of the mobile phone by checking “Use automatic switching” displayed on the screen, from the terminal cache or the web server (200).

[0089] Additionally, the user can check “Use manual switching” and additionally input the subject to be selected from among the terminal cache or web server (200) to perform the copy verification of the ID card as a direct subject selection command.

[0090] For example, if a user selects "WASM", an inspection target image (400) can be acquired using the automatic shooting function without having to press the shooting button, and at this time, the processor of the cache memory (110) can perform the OCR function. On the other hand, if a user selects "Server", the shooting button is activated, the user can directly take a picture, and the image thus acquired can be transmitted to the web server (200), so that the web server (200) can perform the OCR function.

[0091] When the user selects “Use Forced Completion UI,” the manual shooting button is activated, and when the user presses the shooting button, the processor of the cache memory (110) can perform the OCR function by utilizing the photos taken up to that point.

[0092] Information on the authentication performing entity may be generated based on the user's input or the user terminal (100)'s own judgment. The authentication performing entity information may be information on which of the processor (111) and the server processor (201) stored in the cache memory is the entity that determines the authenticity of the ID card. The user terminal (100) may transmit the generated authentication performing entity information to the communication module (210) of the web server (200).

[0093] The server processor (201) can determine whether to perform a determination of authenticity of an identification card on the basis of the authentication performing entity information transmitted from the user terminal (100) to the communication module (210) of the web server (200).

[0094] When receiving authentication performing subject information including information indicating that the web server (200) has decided to perform a determination on the authenticity of an ID card, the server processor (201) can control the communication module (210) of the web server (200) to receive an inspection target image (400) acquired by the camera (120) of the user terminal (100) from the user terminal (100).

[0095] The server processor (201) can determine, based on the inspection target image (400), whether the ID card-type object (401) included in the inspection target image (400) is a real ID card using the server first artificial intelligence model (202).

[0096] If the server processor (201) determines that the determination of the authenticity of an ID card is to be performed on the web server (200), it can generate information on a shooting auxiliary screen (131) that includes information indicating that shooting should be performed directly, including real-time images acquired in real time by the camera (120) of the user terminal (100). The shooting auxiliary screen (131) may be a signal or data required for the display (130) of the user terminal (100) to display the shooting auxiliary screen (131).

[0097] The server processor (201) can control the communication module (210) to transmit information of the shooting auxiliary screen (131) to the user terminal (100) so that the shooting auxiliary screen (131) is displayed on the display (130) of the user terminal (100).

[0098] The user can take a picture while checking the shooting auxiliary screen (131) to ensure that the entire configuration of the ID card is properly included. At this time, the user can complete the image capture by inputting a shooting command into the user terminal (100) by pressing the shooting button.

[0099] When the camera (120) of the user terminal (100) performs image capturing, the server processor (201) can control the communication module (210) to receive the captured real-time image from the user terminal (100).

[0100] The server processor (201) can receive real-time images captured by the camera (120) of the user terminal (100) from the communication module (210).

[0101] The server processor (201) can extract text contained in a real-time image using a second artificial intelligence model pre-trained to perform optical character recognition (OCR) based on the real-time image. In this case, the second artificial intelligence model may be an artificial intelligence model provided in the web server (200).

[0102] The server processor (201) can determine whether the extracted text satisfies the identification component based on the content of the extracted text and its location in the real-time image.

[0103] If the extracted text satisfies the identification component, the server processor (201) can determine the real-time image as the inspection target image (400) acquired by the camera (120) of the user terminal (100).

[0104] Meanwhile, the processor (111) stored in the cache memory can generate authentication performing subject information.

[0105] When authentication performing subject information including information on the purpose of deciding to perform a determination of authenticity of an ID card in the cache memory (110) is generated, the processor (111) stored in the cache memory can control the user terminal (100) to receive the first artificial intelligence model from the web server (200) and store it in the cache memory (110).

[0106] When information on the authentication performing entity that includes information on the purpose of deciding to perform a determination on the authenticity of an ID card in the cache memory (110) is generated, the processor (111) stored in the cache memory can determine, based on the inspection target image (400), whether the ID card format object (401) included in the inspection target image (400) is a real ID card using the first artificial intelligence model (112) stored in the cache memory.

[0107] When authentication performing subject information including information on the purpose of deciding to perform a determination on the authenticity of an ID card on a web server (200) is generated, the processor (111) stored in the cache memory can control the user terminal (100) to transmit the generated authentication performing subject information to the communication module (210) of the web server (200).

[0108] At this time, the processor (111) stored in the cache memory can generate authentication execution subject information corresponding to the user's subject selection command entered by the user into the user terminal (100).

[0109] Meanwhile, the authentication agent may be determined automatically rather than manually. In this case, the authentication agent may be determined automatically based on the memory capacity status of the user terminal (100) or the communication status between the user terminal (100) and the web server (200).

[0110] The processor (111) stored in the cache memory can measure the server operation time. The server operation time may be the time elapsed from the time when the authentication performing entity information, including information indicating the intent of the decision to perform the identification verification determination on the web server (200), is transmitted to the communication module (210) of the web server (200).

[0111] The processor (111) stored in the cache memory can control the user terminal (100) to receive the first artificial intelligence model from the web server (200) and store it in the cache memory (110) when the server operation time exceeds the first reference time. The first reference time may be a preset time that serves as a standard for determining whether the web server (200) is excessively delayed in determining the authenticity of an identification card.

[0112] If the server operation time exceeds the first reference time, the processor (111) stored in the cache memory can determine whether the ID card type object (401) included in the inspection target image (400) is a real ID card by using the first artificial intelligence model (112) stored in the cache memory based on the inspection target image (400).

[0113] The processor (111) stored in the cache memory can measure the cache operation time. The cache operation time may be the time elapsed from the time of generating the authentication performing entity information containing information indicating the purpose of deciding to perform the determination of the authenticity of an ID card in the cache memory (110).

[0114] The second reference time may be a preset time that serves as a reference for determining whether the determination of authenticity of an ID card in the cache memory (110) is excessively delayed.

[0115] If the cache operation time exceeds the second reference time, the processor (111) stored in the cache memory can generate authentication performing subject information including information indicating that the web server (200) has decided to perform a determination on the authenticity of the identification card.

[0116] If the cache operation time exceeds the second reference time, the processor (111) stored in the cache memory can control the user terminal (100) to transmit authentication performing subject information including information indicating that the web server (200) has decided to perform a determination on the authenticity of an ID card to the communication module (210) of the web server (200).

[0117] The processor (111) stored in the cache memory can determine whether the first artificial intelligence model can be stored in the cache memory (110) based on the capacity of the cache memory (110) that is empty and capable of storing additional data.

[0118] If it is determined that it is impossible to store the first artificial intelligence model in the cache memory (110), the processor (111) stored in the cache memory can generate authentication performing subject information including information indicating that the determination of authenticity of the identification card is to be performed on the web server (200).

[0119] Meanwhile, the determination of authenticity of an ID card may be performed by a cache memory (110) or by a web server (200), and an application may be installed on the user terminal (100) and supplementary processing may be performed by the installed application.

[0120] When a user terminal (100) accesses an ID card authenticity determination website, the server processor (201) can generate information on a screen for selecting whether to download an application.

[0121] The screen for selecting whether to download an application may be a screen that includes a question asking the user whether to download an application provided by the operator of an ID verification website, and the information in the screen for selecting whether to download an application may be data or a signal required to display the screen for selecting whether to download an application on the display (130).

[0122] The server processor (201) can control the communication module (210) to transmit information on the application download selection screen to the user terminal (100) so that the application download selection screen is displayed on the display (130) of the user terminal (100).

[0123] The application data (220) may include an application first artificial intelligence model (222) and an application processor (221) configured to determine whether an identification is genuine.

[0124] Specifically, the application data (220) may include application packages. Each application package may include pre-written source codes that enable the operation of the first artificial intelligence model and the processor. The application package may be configured to include an APK (Android application package) file of the Android platform, but is not limited thereto, and may be a program package that includes files of various platforms. For example, the application package may also include files of the iOS platform.

[0125] The application first artificial intelligence model (222) is configured to operate identically to the first artificial intelligence model that can be stored in the cache memory (110), and the application processor (221) can be configured to perform copy determination identically to the processor that can be stored in the cache memory (110).

[0126] When a user terminal (100) receives an application installation command from a user, a processor (111) stored in a cache memory can control the user terminal (100) to download application data (220) from a web server (200).

[0127] When the user terminal (100) receives an installation rejection command from the user, the processor (111) stored in the cache memory can control the user terminal (100) to receive the first artificial intelligence model from the web server (200) and store it in the cache memory.

[0128] When the user terminal (100) receives an installation rejection command from the user, the processor (111) stored in the cache memory can determine whether the ID card type object (401) included in the inspection target image (400) is a real ID card by using the first artificial intelligence model (112) stored in the cache memory based on the inspection target image (400).

[0129] The processor (111) stored in the cache memory can determine whether an application can be installed based on the memory capacity of the user terminal (100).

[0130] If the processor (111) stored in the cache memory determines that it is impossible to download the application to the user terminal (100), it can control the user terminal (100) to receive the first artificial intelligence model from the web server (200) and store it in the cache memory (110).

[0131] If it is determined that it is impossible to download the application to the user terminal (100), the processor (111) stored in the cache memory can use the first artificial intelligence model (112) stored in the cache memory to determine whether the ID card type object (401) included in the inspection target image (400) is a real ID card.

[0132] The processor (111) stored in the cache memory can calculate the data transmission and reception speed between the user terminal (100) and the communication module (210) of the web server (200).

[0133] The processor (111) stored in the cache memory can generate authentication performing entity information including information indicating the intent of deciding to perform the identification card authenticity determination on the web server (200) if the data transmission / reception speed is higher than the first reference speed. The first reference speed may be the data transmission speed per unit time that serves as a standard for determining whether the identification card authenticity determination can be performed smoothly on the web server (200).

[0134] At this time, the processor (111) stored in the cache memory can control the user terminal (100) to transmit the generated authentication performing subject information to the communication module (210) of the web server (200).

[0135] If the data transmission / reception speed is lower than the first reference speed and higher than the second reference speed, the processor (111) stored in the cache memory can control the user terminal (100) to download application data (220) from the web server (200). The second reference speed may be a data transmission speed per unit time that serves as a standard for determining whether the application can be downloaded smoothly to verify the authenticity of an ID card.

[0136] If the data transmission / reception speed is lower than the second reference speed, the processor (111) stored in the cache memory can control the user terminal (100) to receive the first artificial intelligence model from the web server (200) and store it in the cache memory (110).

[0137] In this way, if the data transmission and reception speed is lower than the second reference speed, the processor (111) stored in the cache memory can determine whether the ID card-type object (401) included in the inspection target image (400) is a real ID card, based on the inspection target image (400), using the first artificial intelligence model (112) stored in the cache memory.

[0138] Meanwhile, in order for the processor and the first artificial intelligence model to be downloaded to the cache memory (110) to be run based on a web server (200) rather than being provided in an application or server, it is necessary to implement the processor and the first artificial intelligence model in a language used to write a program run on a web server (200).

[0139] The first artificial intelligence model and processor may be configured to be implemented based on Web Assembly so that the application's functions can be implemented on an ID authentication website.

[0140] WebAssembly is a binary source code that runs in a web browser and can be used to build web applications with enhanced speed and security. WebAssembly can be generated by compiling code written in languages ​​such as C, C++, and Rust. The generated WebAssembly code can then be executed in a browser.

[0141] The web-based ID copy verification system (1) may further include a code converter. The code converter may receive data of an application processor (221) and an application first artificial intelligence model (222). The code converter may automatically convert codes of application packages implementing the application processor (221) and the application first artificial intelligence model (222) implemented in the application to generate web assembly-based code. The code converter may generate a first artificial intelligence model and processor to be transmitted to a cache memory (110) based on the generated web assembly-based code. The code converter may transmit the generated first artificial intelligence model and processor to a communication module (210) so that they are transmitted to a user terminal (100).

[0142] FIG. 5 is a drawing illustrating a photographing assistance screen displayed when determining whether an ID is authentic according to one embodiment, and FIG. 6 is a drawing illustrating a screen displayed when determining whether an ID is authentic is completed according to one embodiment.

[0143] Referring to FIGS. 5 and 6, the user can adjust the position of the camera (120) so that the ID card type object (401) that the user wishes to photograph is photographed while viewing the real-time image displayed on the shooting auxiliary screen (131) displayed on the user's smartphone.

[0144] Optical character recognition (OCR) can be continuously performed on real-time images being acquired by the camera (120) in real time while the user adjusts the position of the camera (120).

[0145] At this time, when the ID card format object (401) is recognized based on the content and location of the texts extracted by optical character recognition (OCR), the processor of the cache memory (110) acquires a real-time image including the ID card format object (401) as an image to be analyzed.

[0146] When the image to be analyzed is acquired, the processor of the cache memory (110) analyzes whether the ID card type object (401) included in the image is a real ID card, and when the analysis is completed, the analysis result can be displayed on the screen of the display (130) as shown.

[0147] When it is determined by the processor whether the ID type object (401) included in the inspection target image (400) is a real ID, the first artificial intelligence model (112) stored in the cache memory and the processor can be configured to be removed and deleted from the cache memory (110).

[0148] At least one component may be added or deleted in response to the performance of the components described above. Furthermore, those skilled in the art will readily understand that the relative positions of the components may be altered in response to the performance or structure of the system.

[0149] Figure 7 is a flowchart illustrating a control method for a web-based ID card copy verification system according to one embodiment. This is merely a preferred embodiment for achieving the purpose of the present invention, and it is understood that certain components may be added or deleted as needed.

[0150] Referring to FIG. 7, when a user terminal (100) accesses an ID card authenticity determination website, the first artificial intelligence model and processor can be downloaded to the user terminal (100) in a manner transmitted from the web server (200) to the user terminal (100) (1001).

[0151] The first artificial intelligence model and processor downloaded by the user terminal (100) can be stored in the cache memory (110), which is a temporary storage that stores data of the web page accessed by the user terminal (100) (1002).

[0152] The processor (111) stored in the cache memory can receive real-time images acquired by the camera (120) of the user terminal (100) (1003).

[0153] The processor (111) stored in the cache memory can determine whether the real-time image is an image containing an ID card type object (401) (1004).

[0154] The processor (111) stored in the cache memory can control the camera (120) to perform image capturing to acquire the real-time image as an inspection target image (400) including the ID card type object (401) if the real-time image is determined to be an image including the ID card type object (401) (1005).

[0155] By using the first artificial intelligence model (112) stored in the cache memory, based on the inspection target image (400) acquired by the camera (120) of the user terminal (100), by the processor (111) stored in the cache memory, it is possible to determine whether the ID card type object (401) included in the inspection target image (400) is a real ID card (1006).

[0156] The control method of the web-based ID copy verification system (1) according to the embodiments of the present invention described so far and the embodiments to be described in the future can be implemented in the form of a program that can be driven by a processor.

[0157] Here, the program may include program commands, data files, and data structures, either singly or in combination. The program may be designed and produced using machine language code or high-level language code. The program may be specifically designed to implement the control method of the aforementioned web-based ID copy verification system (1), or may be implemented using various functions or definitions that are known and available to those skilled in the art of computer software. The program for implementing the control method of the aforementioned web-based ID copy verification system (1) may be recorded on a recording medium readable by a processor. In this case, the recording medium may be a memory.

[0158] The memory can store a program that performs the operations described above and the operations described below, and the memory can execute the stored program. In the case where there are multiple processors and memories, they can be integrated into a single chip or provided in physically separate locations. The memory can include volatile memory such as static random access memory (S-RAM) and dynamic random access memory (DRAM) for temporarily storing data. In addition, the memory can include non-volatile memory such as read only memory (ROM), erasable programmable read only memory (EPROM), and electrically erasable programmable read only memory (EEPROM) for long-term storage of control programs and control data.

[0159] The processor may include various logic circuits and arithmetic circuits, process data according to a program provided from memory, and generate control signals according to the processing results.

[0160] The disclosed embodiments have been described with reference to the attached drawings as described above. Those skilled in the art will understand that the present invention can be implemented in forms other than the disclosed embodiments without altering the technical spirit or essential features of the present invention. The disclosed embodiments are illustrative and should not be construed as limiting.

Claims

1. First artificial intelligence model; and comprising a processor configured to determine whether an identification is authentic; The above first artificial intelligence model and the above processor, When a user terminal accesses a website for determining the authenticity of an ID card, it is configured to be downloaded in a manner that is transmitted from a web server to the user terminal. The first artificial intelligence model and processor downloaded by the above user terminal are, It is stored in the cache memory, which is a temporary storage that stores data of the web page accessed by the user terminal. The processor stored in the above cache memory, A web-based ID card copy determination system, configured to determine whether an ID card format object included in an inspection target image is a real ID card by using a first artificial intelligence model stored in the cache memory, based on an inspection target image acquired by a camera of the user terminal and including an ID card format object.

2. In paragraph 1, A machine learning module is further included, configured to learn the first artificial intelligence model through a machine learning method by setting a learning image containing an ID card format object as an input variable and setting learning image classification information, which is information on whether the ID card format object contained in the learning image is the actual ID card, as an output variable. Each of the above learning images, A web-based ID copy determination system, wherein the image is one of a real ID card image for learning, a screen display image for learning in which the ID card is displayed on a screen, and a copy image for learning of the ID card, which is an image of a photographed copy of the ID card.

3. In paragraph 1, The processor stored in the above cache memory: Receive real-time images acquired by the camera of the above user terminal; Determining whether the above real-time image is an image containing the above identification type object; and A web-based ID card copy determination system, which controls the camera to perform image capturing so as to acquire the real-time image as the inspection target image when the real-time image is determined to be an image including the ID card format object.

4. In paragraph 3, The processor stored in the above cache memory: Based on the above real-time image, text included in the real-time image is extracted through a second artificial intelligence model that has been pre-trained to perform optical character recognition (OCR); Based on the content of the extracted text and its location in the real-time image, determining whether the extracted text is a text that satisfies the identification component; and A web-based ID card copy determination system configured to determine whether the real-time image is the inspection target image based on whether the extracted text is text that satisfies the ID card component.

5. In paragraph 1, The first artificial intelligence model and processor stored in the above cache memory, A web-based ID copy determination system, wherein when it is determined by the processor whether an ID format object included in the inspection target image is a real ID card, it is configured to be removed and deleted from the cache memory.

6. In paragraph 1, A server first artificial intelligence model configured to operate identically to the first artificial intelligence model and provided on the web server; and Further comprising a server processor provided on the web server and configured to determine whether the identification is authentic; The above server processor, Based on the authentication execution subject information transmitted from the user terminal to the communication module of the web server, it is determined whether to perform a determination of the authenticity of the identification card on the web server. A web-based ID copy verification system, wherein the authentication performing subject information is information about which of the processor stored in the cache memory and the server processor is the subject that determines the authenticity of the ID.

7. In paragraph 6, The above server processor: When receiving information on the authentication performing entity including information on the intent of deciding to perform a determination on the authenticity of the above identification card on the above web server, controlling the communication module of the above web server to receive an inspection target image acquired by the camera of the above user terminal from the above user terminal; and A web-based ID copy determination system configured to determine whether an ID card format object included in the inspection target image is a real ID card using the server first artificial intelligence model based on the inspection target image.

8. In paragraph 7, The above server processor: If it is determined that the determination of authenticity of the above identification is to be performed on the web server, information on a shooting auxiliary screen containing information to directly perform shooting while including real-time images acquired in real time by the camera of the user terminal is generated; Controlling the communication module to transmit information of the shooting auxiliary screen to the user terminal so that the shooting auxiliary screen is displayed on the display of the user terminal; and A web-based ID card copy verification system that controls the communication module to receive the captured real-time image from the user terminal when the camera of the user terminal performs image capture according to the user's capture command input into the user terminal.

9. In paragraph 8, The above server processor: A real-time image captured by the camera of the user terminal is received from the communication module; Based on the above real-time image, text included in the real-time image is extracted through a second artificial intelligence model that has been pre-trained to perform optical character recognition (OCR); Based on the content of the extracted text and its location in the real-time image, determining whether the extracted text is a text that satisfies the identification component; and A web-based ID card copy determination system configured to determine the real-time image as an inspection target image acquired by a camera of the user terminal if the extracted text satisfies the ID card component.

10. In paragraph 6, The processor stored in the above cache memory: Generate the above authentication performing entity information; When the authentication performing entity information including information on the intent to perform the determination of authenticity of the above identification in the above cache memory is generated, the user terminal is controlled to receive the first artificial intelligence model from the web server and store it in the above cache memory; When the authentication performing subject information including information on the intent to perform the determination of authenticity of the above identification card in the cache memory is generated, based on the inspection target image, using the first artificial intelligence model stored in the cache memory, it is determined whether the identification card format object included in the inspection target image is a real identification card; and A web-based ID copy verification system that controls the user terminal to transmit the generated authentication performing subject information to the communication module of the web server when authentication performing subject information including information indicating that a decision to perform a determination on the authenticity of the ID is made on the web server is generated.

11. In paragraph 10, The processor stored in the above cache memory, A web-based ID card copy verification system that generates authentication execution subject information corresponding to a user subject selection command entered into the user terminal.

12. In paragraph 10, The processor stored in the above cache memory: Measure the server operation time, which is the time elapsed from the time the authentication performing entity information including information indicating that the web server has decided to perform a determination on the authenticity of the identification card; If the above server operation time exceeds the first reference time, control the user terminal to receive the first artificial intelligence model from the web server and store it in the cache memory; and A web-based ID copy determination system, which determines whether an ID card-type object included in the inspection target image is a real ID card by using a first artificial intelligence model stored in the cache memory based on the inspection target image when the above server operation time exceeds the first reference time.

13. In paragraph 10, The processor stored in the above cache memory: Measure the cache operation time, which is the time elapsed from the time of generating the authentication performing entity information containing information indicating the intent to perform the determination of authenticity of the above identification card in the cache memory; If the above cache operation time exceeds the second reference time, the authentication performing entity information including information indicating that the web server has decided to perform the determination of the authenticity of the ID card is generated; and A web-based ID copy verification system that controls the user terminal to transmit authentication performing entity information including information indicating that the web server has decided to perform a determination on the authenticity of the ID card if the cache operation time exceeds the second reference time, to the communication module of the web server.

14. In paragraph 10, The processor stored in the above cache memory: Based on the capacity of the cache memory, determine whether the first artificial intelligence model can be stored in the cache memory; and A web-based ID copy verification system configured to generate authentication performing entity information including information indicating that the determination of authenticity of the ID is to be performed on the web server when it is determined that it is impossible to store the first artificial intelligence model in the cache memory.

15. In paragraph 6, The above server processor: When a user terminal accesses an ID authenticity determination website, information on an application download selection screen containing a question asking the user whether to download an application provided by the operator of the ID authenticity determination website is generated; Controlling the communication module to transmit information on the application download selection screen to the user terminal so that the application download selection screen is displayed on the display of the user terminal; The processor stored in the above cache memory: When the user terminal receives an application installation command from a user, the user terminal is controlled to download data of the application from the web server; When the user terminal receives an installation rejection command from the user, the user terminal is controlled to receive the first artificial intelligence model from the web server and store it in the cache memory; and When the user terminal receives an installation rejection command from the user, based on the inspection target image, the first artificial intelligence model stored in the cache memory is used to determine whether the ID card type object included in the inspection target image is a real ID card. The data of the above application is, A web-based ID copy verification system, comprising an application first artificial intelligence model configured to operate identically to the first artificial intelligence model and an application processor configured to determine whether the ID is authentic.

16. In paragraph 15, The processor stored in the above cache memory: Based on the memory capacity of the user terminal, determine whether the application can be installed; If it is determined that it is impossible to download the application to the user terminal, control the user terminal to receive the first artificial intelligence model from the web server and store it in the cache memory; and A web-based ID card copy determination system configured to determine whether an ID card format object included in the inspection target image is a real ID card by using a first artificial intelligence model stored in the cache memory when it is determined that downloading the application to the user terminal is impossible.

17. In paragraph 15, The processor stored in the above cache memory: Calculate the data transmission and reception speed between the user terminal and the communication module of the web server; If the data transmission / reception speed is greater than or equal to the first reference speed, the user terminal is controlled to generate authentication performing entity information including information indicating that the web server has decided to perform a determination on the authenticity of the identification card, and to transmit the generated authentication performing entity information to the communication module of the web server; If the data transmission / reception speed is less than the first reference speed and greater than the second reference speed, control the user terminal to download data of the application from the web server; If the data transmission / reception speed is less than the second reference speed, control the user terminal to receive the first artificial intelligence model from the web server and store it in the cache memory; and A web-based ID copy determination system configured to determine whether an ID card-type object included in the inspection target image is a real ID card by using a first artificial intelligence model stored in the cache memory based on the inspection target image if the data transmission / reception speed is less than the second reference speed.

18. In paragraph 15, The above first artificial intelligence model and the above processor, A web-based ID copy verification system, configured to be implemented based on web assembly so that the functions of the above application can be implemented on the ID authenticity judgment website.

19. When a user terminal accesses an ID authenticity determination website, a step of downloading a first artificial intelligence model and processor to the user terminal in a manner transmitted from a web server to the user terminal; A step in which the first artificial intelligence model and processor downloaded by the user terminal are stored in a cache memory, which is a temporary storage that stores data of a web page accessed by the user terminal; A step of receiving a real-time image acquired by a camera of the user terminal by a processor stored in the cache memory; A step of determining, by a processor stored in the cache memory, whether the real-time image is an image including an identification card format object; A step of controlling the camera to perform image capturing so as to acquire the real-time image as an inspection target image including the ID card type object, if the real-time image is determined to be an image including the ID card type object by the processor stored in the cache memory; and A control method for a web-based ID copy identification system, comprising the step of determining, by a processor stored in the cache memory, whether the ID card format object included in the inspection target image is a real ID card, based on the inspection target image acquired by the camera of the user terminal, using the first artificial intelligence model stored in the cache memory.

20. A non-transitory storage medium having stored thereon a computer-readable computer program for executing the control method of the web-based identification card copy verification system of Article 19.

Images