WO2022097982A1

WO2022097982A1 - Method and server for providing face recognition-based digital signature service

Info

Publication number: WO2022097982A1
Application number: PCT/KR2021/015084
Authority: WO
Inventors: 유제훈
Original assignee: 주식회사 아이온커뮤니케이션즈
Priority date: 2020-11-06
Filing date: 2021-10-26
Publication date: 2022-05-12
Also published as: KR20220061919A

Abstract

A method for providing a face recognition-based digital signature service of the present invention comprises the steps of: receiving an identification image and a real-time face image of a signer; determining whether the identification image and the face image match each other, through deep leaning analysis thereof; when the identification image and the face image match, generating and issuing a face ID on the basis of the face image; and applying the face ID as a digital signature for a digital document. Accordingly, the reliability and efficiency of a digital signature can be increased by performing the digital signature on the basis of the recognition of the signer's own face.

Description

Method and server for providing facial recognition-based digital signature service

The present invention relates to a method and server for providing a face recognition-based electronic signature service, and in particular, a method for providing a face recognition-based electronic signature service capable of increasing the reliability and efficiency of the electronic signature by performing the electronic signature based on the signer's own facial recognition and to the server.

According to the revision of the Digital Signature Act, the superior status of public certificates was extinguished, and there was a situation in which a strong personal authentication method as strong as the public certificates was needed. Accordingly, in order to completely replace the signature in the offline environment in the online environment, a technical means for proving that the party to the signature (hereinafter referred to as a 'signer') is the person is important.

In preparation for such a situation, a private certificate (PASS, Kakao, etc.) can be used as an alternative to self-authentication and signature, but it is insufficient to completely relieve the signer's anxiety. In addition, technically, there is a need for a means to increase the reliability of digital signatures in consideration of compatibility in accordance with technical standards such as FIDO (Fast Identity Online) and digital signature.

The present invention is to solve the above problems, and to provide a face recognition-based electronic signature service providing method and server that can increase the reliability and efficiency of the electronic signature by performing the electronic signature based on the signer's own facial recognition will be.

In order to achieve the above object, a method for providing a face recognition-based electronic signature service according to an embodiment of the present invention includes the steps of receiving an ID image of a signer and a real-time face image; determining whether the identification image and the facial image are matched by deep learning analysis; generating and issuing a face ID based on the face image when the ID image and the face image match; and applying the Face ID as an electronic signature of the electronic document.

In addition, the step of receiving the ID image may include: extracting verification data from the ID image through optical character recognition; and, verifying the authenticity of the ID image by using the extracted verification data.

In addition, when the face ID is issued, deleting the extracted verification data; may include.

Furthermore, the step of determining whether the identification image and the facial image are matched by deep learning analysis may include: storing a training data set for the facial image; and performing authentication of the face image using a deep learning model through training based on the data set.

In addition, the generating and issuing of the face ID may include generating a hash value of an electronic signature to be given to the electronic document and issuing it together with the face ID.

In addition, generating and issuing the face ID may issue document identification information for the electronic document.

A face recognition-based digital signature service providing server according to another embodiment of the present invention includes: a contract verification unit for receiving a signer's ID image input; a signature management unit receiving a real-time facial image of the signer; an authentication unit that determines whether the identification image and the facial image are matched by deep learning analysis; an electronic signature issuing unit that generates and issues a face ID based on the face image when the ID image and the face image match; and an electronic document recording unit that applies the Face ID as an electronic signature of the electronic document.

In addition, the contract verification unit extracts verification data from the ID image through optical character recognition, and verifies the authenticity of the ID image using the extracted verification data, and the verification unit, when the ID image is verified , it may be determined whether the ID image and the face image match.

Also, the contract verification unit may delete the extracted verification data when the Face ID is issued.

In addition, the authentication unit, a data set storage unit for storing a training data set for the face image; and a deep learning unit that performs authentication of the facial image using a deep learning model through training based on the data set.

Also, the electronic signature issuing unit may generate a hash value of the electronic signature to be given to the electronic document and issue it together with the face ID.

In addition, the electronic signature issuing unit may issue document identification information for the electronic document.

The method and server for providing a face recognition-based electronic signature service according to the present invention as described above can increase the reliability and efficiency of the electronic signature by performing the electronic signature based on the signer's own facial recognition.

Furthermore, in completing the electronic signature by performing identity authentication through facial recognition, issuing Face ID, the hash value and document identification information given to the electronic contract according to it, and combining it with the electronic document, preventing forgery and falsification of the electronic document and the signer's It can enhance non-repudiation.

1 is a block diagram schematically showing the configuration of a face recognition-based digital signature service providing server according to an embodiment of the present invention.

2 is a block diagram schematically showing the configuration of an authenticator according to an embodiment of the present invention.

3 is a flowchart illustrating a method for providing a face recognition-based digital signature service according to an embodiment of the present invention.

4 is an exemplary diagram for explaining a Face ID issuance process according to an embodiment of the present invention.

5 is an exemplary diagram illustrating an electronic document combined with Face ID according to an embodiment of the present invention.

6 is a flowchart illustrating a method for providing a face recognition-based electronic signature service according to an embodiment of the present invention.

7 is an exemplary diagram illustrating an electronic signature field of a signature management unit according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings in order to clarify the technical spirit of the present invention. In describing the present invention, if it is determined that a detailed description of a related well-known function or component may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. Elements having substantially the same functional configuration in the drawings are given the same reference numbers and reference numerals as much as possible even though they are shown in different drawings. For convenience of explanation, if necessary, the device and method will be described together.

1 is a block diagram schematically showing the configuration of a face recognition-based digital signature service providing server according to an embodiment of the present invention, and FIG. 2 is a block diagram schematically showing the configuration of an authentication unit according to an embodiment of the present invention. .

Referring to FIG. 1 , the server 100 includes an electronic document transmission unit 110 , a contract verification unit 120 , a signature management unit 130 , an authentication unit 140 , an electronic signature issuance unit 150 , an electronic document recording unit ( 160 ), a server control unit 170 , and a storage unit 180 .

The server 100 provides an electronic document that is the target of the electronic signature to a mobile terminal (not shown), analyzes an image input from the mobile terminal, issues an electronic signature, and combines the issued electronic signature with the electronic document Electronic documents for final distribution can be completed.

The electronic document transmission unit 110 may transmit the electronic document to the mobile terminal. In addition, the final completed electronic document combined with the electronic signature may be transmitted to the mobile terminal of the contracting party and interested parties. As a method of transmitting an electronic document to a mobile terminal, it can be transmitted using a network, and the network can be configured regardless of the communication mode such as wired or wireless, and includes a personal area network (PAN), a local area network. It may be composed of various communication networks such as a local area network (LAN), a metropolitan area network (MAN), and a wide area network (WAN).

Here, the mobile terminal may be a device used by the signer who is the subject of the electronic signature, that is, the contracting party. The signer's ID image and real-time facial image may be provided through the mobile terminal. Such a mobile terminal may be implemented as a smart device capable of network connection, such as a smart phone, a personal computer, a tablet, a mobile pad, a laptop computer, and the like.

In the mobile terminal, an application (application, APP) that performs a function for receiving a face recognition-based electronic signature service is installed, so that an electronic document can be provided using the application. In addition, an ID image and a real-time face image may be input through the application and provided to the server 100 . That is, the signer's ID image may be attached through the application or a real-time facial image of the signer may be acquired through the camera of the mobile terminal.

The contract verification unit 120 may receive the signer's ID image. That is, the contract certifying unit 120 may include various types of evidence that can prove that a party (ie, a signer) of a contract has concluded a contract by means of an electronic document. In this case, the data included in the contract proof may include one or more of a photo, a voice, and a fingerprint. Various other materials can be used as proof of contract if they can prove the conclusion of the contract. The verification data may be attached to the mobile terminal or may be downloaded from the storage unit 180 and attached to the electronic document. In this case, it may be reduced or enlarged to fit the form of the electronic document and combined at a predetermined position to be displayed.

The ID image input to the contract verification unit 120 is to prove the signer's identity by photographing the signer's identification card, such as a resident registration card, a car driver's license, or a passport.

The contract verification unit 120 may extract verification data from the received ID image through Optical Character Recognition (OCR). Here, the verification data may include information such as letters, numbers, and photos extracted from the ID image. Furthermore, the contract verification unit 120 may verify the authenticity of the ID image by using the extracted verification data.

The contract verification unit 120 may delete the extracted verification data when the Face ID is issued.

The signature management unit 130 may receive a real-time facial image captured through the signer's mobile terminal. Here, the real-time facial image may be an image used as an electronic signature for proving that the person himself/herself signed the contract by photographing the signer's face in order to prove that the contract has been concluded with the corresponding electronic document. Such a real-time facial image may mean an image captured in real time at a contract site with a camera of a mobile terminal. In this case, the face image may include at least one image including a front image of the face. Furthermore, the face image may include three images including front/left/right images of the face.

The authenticator 140 may determine whether the identity card image and the facial image are matched by deep learning analysis. When the ID image is verified, the authenticator 140 may determine whether the ID image and the face image match. That is, the authentication unit 140 is authenticated as the person used as the electronic signature when the ID image and the face image match each other. The authentication unit 140 may be implemented as a separate server to perform a function for authentication alone.

Referring to FIG. 2 , the authenticator 140 may include a data set storage unit 141 and a deep learning unit 143 .

The data set storage unit 141 may store a data set for the face image. Here, the training data set may include a source image including live-action figures of Asians including Koreans and a label in which the face of the source image is regionized. These data sets may be classified into a learning pair, a verification pair, and a test pair, which are data sets required for learning of the deep learning unit 143 , and may be input to the deep learning unit 143 as training data.

The deep learning unit 143 may sufficiently train the deep learning neural network by repeatedly comparing the source image and the label by inputting a data set to the deep learning neural network. That is, the deep learning unit 143 may mean learning a large amount of data using a dataset. Accordingly, the deep learning unit 143 may implement a process of increasing the accuracy of user authentication.

The deep learning unit 143 may perform deep learning analysis by reading the deep learning model and extracting features of the person from the ID image and facial image according to a neural network algorithm built up. To this end, the deep learning unit 143 is an inception module-based convolutional neural network (CNN), a deep neural network (DNN), a recurrent neural network having excellent performance in image recognition. , RNN), a neural network algorithm to which various deep learning models are applied, such as a restricted Boltzmann machine, a Deep Belief Network (DBN), and a Deep Q-Network can be used.

The deep learning unit 143 may perform authentication of a facial image using a deep learning model generated by learning a pre-stored data set based on a neural network algorithm. That is, the deep learning unit 143 may compare and analyze whether an identification card image and a real-time facial image are inputted using the deep learning model to determine whether they match.

The deep learning unit 143 extracts specific attribute spaces such as background, facial skin tone, color, and texture from the ID image, extracts specific attribute spaces such as background, facial skin tone, color, and texture from the facial image, and compares and analyzes the facial If the accuracy of the image is greater than or equal to a preset value, the person may be authenticated.

Referring back to FIG. 1 , when the ID image and the face image match, the electronic signature issuing unit 150 may generate and issue a face ID based on the face image. Face ID may include year/month/day/hour:minute:second at the time of signing, name of a signer, photo of signer, and name of signer (e-mail/contact information).

The electronic signature issuing unit 150 may generate a hash value of the electronic signature to be given to the electronic document and issue it together with the face ID. The hash value is issued for the corresponding digital signature and may include a part of the Face ID encryption value.

The electronic signature issuing unit 150 may issue document identification information for the electronic document. The document identification information is identification information obtained by encrypting Face ID, which can be used to check whether a document is forged or forged after authentication of the electronic signature is completed, and can be implemented as a QR code or the like.

The electronic document recorder 160 may combine the face ID with the electronic document by applying the face ID as an electronic signature of the electronic document. That is, when the electronic signature is issued, the electronic document recording unit 160 may be reduced or expanded to a predetermined size according to the form of the electronic document and displayed in combination with the electronic document.

The server controller 170 may control the overall operation of the server 100 .

The server control unit 170 may control to design and provide a form of an electronic document. That is, the server control unit 170 may provide an electronic document designing a form in which identity authentication settings and identification fields are set through facial recognition for performing Face ID authentication.

The server control unit 170 may determine whether the information regarding the electronic document has been forged or altered before or after transmission. In the present invention, as a method of checking forgery, forgery or not by comparing hash values using a hash function to check whether forgery or not, according to an embodiment, an appropriate method (MD-5, Kerberos etc.) to check whether the electronic document has been forged or forged.

In addition, the server control unit 170 may authenticate the ID (UDID) of the mobile terminal and input the user ID and password of the mobile terminal to authenticate the mobile terminal and a user using the same.

If the codec format of the photo or audio file, which is the proof material of the contract proof of the electronic document, does not match the codec supported by the operating system of the mobile terminal, the server control unit 170 supports these photos or audio files in the mobile terminal ( ). Transcoding that converts to a codec format can be performed.

The server control unit 170 may check whether the electronic document is forged or falsified through the timestamp part certificate verification module.

When the conclusion of the contract of the electronic document is completed, the server control unit 170 may notify both parties to the contract and interested parties of the conclusion of the contract. Here, as a method of notifying both parties to the contract and interested parties that the contract has been concluded, the conclusion of the contract may be notified by sending a text message to the mobile phone of the contracting party or the contracting party's e-mail.

In addition, the server control unit 170 may perform an operation according to technical regulations in order to provide a face recognition-based electronic signature.

Here, the technical regulations may include electronic Identification, Authentication and trust Services (eIDAS), PDF Advanced Electronic Signature (PAdES), Integration With Face Recognition (Digital Signature), Know Your Customer (KYC), Fast Identity Online (FIDO), etc. can

Electronic IDentification, Authentication and trust Services (EU, 2014) (eIDAS) is a standard specification for electronic identification and electronic signatures in the European Union (EU) market. regulations to support ways in which users can safely conduct online business, such as electronic money transfers or transactions with public services. Accordingly, in Europe, an electronic signature must satisfy the eIDAS standard, and the electronic signature based on the face recognition of the present invention is defined according to this standard.

That is, according to the electronic signature standard, the server control unit 170 provides unique identification information to connect with the signer, and identifies whether the data included in the electronic document is tampered with after it is signed. In addition, when the signed data is changed, the server control unit 170 may display the digital signature as invalid.

In addition, the signer has sole control over the data used to generate the digital signature.

Meanwhile, the server control unit 170 may provide a certificate for the electronic signature, an electronic certificate for verifying the identity of the signer, and linking the electronic signature validation data with the corresponding person.

Such an electronic signature can be technically implemented according to the XAdES, PAdES, and CAdES standards (standards) for digital signatures designated by ETSI.

PAdES (PDF Advanced Electronic Signatures) refers to the EU standard specification for digital signature on PDF and a standard specification for defining advanced electronic signatures.

The present invention complies with the PAdES standard because an electronic document that has been signed electronically between the parties is converted into PDF and provided.

The server control unit 170 may be uniquely connected to the signer to identify the signer. Also, the server control unit 170 may store control of data used only by the signer to generate a signature.

When the data attached to the signature is changed after the digital signature, the server control unit 170 can easily identify it in PaDES.

Digital Signature (a kind of handwritten signatures) is a mathematical technique for verifying the authenticity of an electronic signature or contract. refers to an engineering technique based on a mathematical algorithm to maintain and verify the uniqueness and authenticity of the signature.

The member principle of a digital signature can output a private key and a corresponding public key. In this case, when a message and a private key are given, it is a signature algorithm that generates a signature, and when a message and a public key and a signature are given, it is a signature verification algorithm that accepts or rejects the claim of authenticity of the message. To implement a digital signature, a KPI, a block chain, a public certificate, etc. may be used.

Electronic know your customer or know your client (eKYC) can be used by companies of all sizes to ensure that a proposed customer, agent, consultant or distributor is compliant with anti-bribery regulations and is in fact the person they claim to be (the principal). there is. In particular, it is key to accurately identify who the person online is (who is who you are). It can be mainly used for anti-money laundering (AML) or customer identification in the financial sector. Accordingly, the present invention can be described as eKYC as a process for technically verifying whether a user who has accessed online for an electronic signature is the person who signed the signature.

FIDO (Fast IDentity Online) is a technology proposed to solve the problem that password-based user authentication is inexpensive and convenient, but vulnerable to supplementation. can

The storage unit 180 may store all information related to the electronic document. Here, all of the information may include images of electronic documents, photos such as metadata and ID images, verification materials such as voice and fingerprints, electronic signatures, and electronic documents for which a contract has been completed.

3 is a flowchart illustrating a method for providing a face recognition-based digital signature service according to an embodiment of the present invention, and FIG. 4 is an exemplary diagram for explaining a Face ID issuance process according to an embodiment of the present invention, FIG. 5 is an exemplary diagram illustrating an electronic document combined with Face ID according to an embodiment of the present invention.

1 and 3 , the signer's ID image may be attached from the signer's mobile terminal through an application, or a real-time facial image of the signer may be acquired through the camera of the mobile terminal.

Looking at the process of acquiring a real-time facial image with reference to FIG. 4, if you touch to take a picture of the application as shown in (a), you can take a facial image through the shooting function using the camera supported by the mobile terminal as shown in (b). there is.

Accordingly, the contract verification unit 120 may receive the signer's ID image, and the signature management unit 130 may receive a real-time facial image captured through the signer's mobile terminal (S310). In this case, the face image may include at least one image including a front image of the face. Furthermore, the face image may include three images including front/left/right images of the face.

Meanwhile, the contract verification unit 120 may extract verification data from the received ID image through Optical Character Recognition (OCR). Here, the verification data may include information such as letters, numbers, and photos extracted from the ID image. Furthermore, the contract verification unit 120 may verify the authenticity of the ID image by using the extracted verification data.

Thereafter, the authenticator 140 may determine whether the ID image and the facial image are matched by deep learning analysis (S320). Furthermore, when the ID image is verified, the authenticator 140 may determine whether the ID image and the face image match. The authenticator 140 may perform self-authentication on the face image using a deep learning model learned based on the data set. That is, when the ID image and the face image match each other, it means that the person is authenticated as the person in the digital signature.

The authenticator 140 extracts specific attribute spaces such as background, facial skin tone, color, and texture from the ID image, extracts specific attribute spaces such as background, facial skin tone, color, and texture from the facial image, and compares and analyzes the facial image If the accuracy of the information is 00% or higher, you can authenticate yourself.

The electronic signature issuing unit 150 may generate and issue a face ID based on the face image when the ID image and the face image match (S330:Y) (S340). As shown in (c) of FIG. 4, Face ID generates and provides a preview including year/month/day/hour:minute:second, signer's name, signer's photo, and signer's name (e-mail/contact information) at the time of signing. can

In addition, the electronic signature issuing unit 150 may generate a hash value of the electronic signature to be given to the electronic document and issue it together with the face ID. The hash value is issued for the corresponding digital signature and may include a part of the Face ID encryption value. Accordingly, a face ID to be applied as an electronic signature as shown in (d) of FIG. 4 may be issued.

The electronic signature issuing unit 150 may issue document identification information for the electronic document. As shown in (e) of FIG. 4 , the document identification information is identification information obtained by encrypting the Face ID, and can be used to check whether the document is forged or falsified after authentication of the electronic signature is completed, and can be implemented as a QR code or the like.

The electronic document recorder 160 may apply the Face ID as an electronic signature of the electronic document and combine it with the electronic document (S350). That is, when the electronic signature is issued, the electronic document recording unit 160 may be reduced or expanded to a predetermined size according to the form of the electronic document and displayed in combination with the electronic document.

The electronic document recording unit 160 may record an electronic document in which face ID and document identification information are combined as shown in FIG. 5 . Accordingly, the electronic document for which the contract is completed may be stored in the storage unit 180 .

6 is a flowchart for explaining a method for providing a face recognition-based electronic signature service according to an embodiment of the present invention, and FIG. 7 is an exemplary diagram illustrating an electronic signature field of the signature management unit according to an embodiment of the present invention.

1, 6 and 7 , in the server 100 , the server control unit 170 may design a form of an electronic document and provide it to the mobile terminals 200_1 and 200_2 . That is, the server control unit 170 may set the signature field 10 (S410). The server control unit 170 may provide an electronic document designing a form in which an identity authentication setting 15 and an identification card field are set through facial recognition for performing Face ID authentication.

Referring to FIG. 7 , in the case of the signature field 10 , when the recipient designation 13 is performed, the designation of a person to sign 13A may be set through a pop-up window. In the signer information, the name written in the 'signature field' is automatically entered, and designation can be completed by designating the signer's e-mail and mobile phone number.

Accordingly, the server control unit 170 may transmit the signature information to the mobile terminals 200_1 and 200_2 requiring the signer's signature information (S420).

The mobile terminals 200_1 and 200_2 that have received the signature information are installed with an application (APP) that performs a function to receive a face recognition-based electronic signature service, so that the signer's ID image is input or facial recognition is performed using the application. can be input through (S430). That is, the signer's ID image may be attached through the application or a real-time facial image of the signer may be acquired through the camera of the mobile terminals 200_1 and 200_2.

Accordingly, in the server 100, the contract verification unit 120 receives the signer's ID image (S440), and can extract verification data from the received signer's ID image through Optical Character Recognition (OCR). . Here, the verification data may include information such as letters, numbers, and photos extracted from the ID image. Furthermore, the contract verification unit 120 may verify the authenticity of the ID image by using the extracted verification data.

The signature management unit 130 may receive a real-time facial image captured through the signer's mobile terminals 200_1 and 200_2 (S440). In this case, the face image may include at least one image including a front image of the face. Furthermore, the face image may include three images including front/left/right images of the face.

The authenticator 140 may perform signer authentication by determining whether the ID image and the facial image are matched by deep learning analysis (S450). That is, the authentication unit 140 is authenticated as the person used as the electronic signature when the ID image and the face image match each other. User authentication for facial images can be performed using a deep learning model generated by learning a pre-stored data set based on a neural network algorithm. That is, the authenticator 140 extracts a specific attribute space such as background, facial skin tone, color, and texture from the ID image using the deep learning model, and specific attribute space such as background, facial skin tone, color, and texture from the facial image. After extracting and analyzing the face image, if the accuracy of the facial image is 00% or more, the person can be authenticated.

Thereafter, when the ID image and the face image match, the electronic signature issuing unit 150 may generate and issue a face ID based on the face image (S460). Face ID may include year/month/day/hour:minute:second at the time of signing, name of a signer, photo of signer, and name of signer (e-mail/contact information). In addition, the electronic signature issuing unit 150 may generate a hash value of the electronic signature to be given to the electronic document and issue it together with the face ID. The hash value is issued for the corresponding digital signature and may include a part of the Face ID encryption value.

The electronic signature issuing unit 150 may issue document identification information for the electronic document (S470). The document identification information is identification information obtained by encrypting Face ID, which can be used to check whether a document is forged or forged after authentication of the electronic signature is completed, and can be implemented as a QR code or the like.

Then, the electronic document recording unit 160 may apply the Face ID as an electronic signature of the electronic document and combine it with the electronic document (S480). That is, when the electronic signature is issued, the electronic document recording unit 160 may be reduced or expanded to a predetermined size according to the form of the electronic document and displayed in combination with the electronic document.

The server control unit 170 may distribute the contracted electronic document (S490). The contracted electronic document may be implemented as a file in PDF format, and may be transmitted to the signer's mobile terminals 200_1 and 200_2 through an electronic e-mail or application.

Accordingly, the method of providing a face recognition-based electronic signature service can provide a digital signature service with improved reliability and improved safety.

So far, the present invention has been described in detail with reference to the preferred embodiments shown in the drawings. These embodiments are not intended to limit the present invention, but are merely illustrative, and should be considered in an illustrative rather than a restrictive sense. The true technical protection scope of the present invention should be determined by the technical spirit of the appended claims rather than the above description. Although specific terms are used in this specification, they are only used for the purpose of describing the concept of the present invention and are not used to limit the meaning or scope of the present invention described in the claims. Each step of the present invention does not necessarily have to be performed in the order described, and may be performed in parallel, selectively, or individually. Those of ordinary skill in the art to which the present invention pertains will understand that various modifications and equivalent other embodiments are possible without departing from the essential technical spirit of the present invention as claimed in the claims. It is to be understood that equivalents include both currently known equivalents as well as equivalents developed in the future, ie, all elements invented to perform the same function, regardless of structure.

Claims

receiving a signer's ID image and real-time facial image;

determining whether the identification image and the facial image are matched by deep learning analysis;

generating and issuing a face ID based on the face image when the ID image and the face image match; and,

The face recognition-based electronic signature service providing method comprising a; applying the face ID as an electronic signature of the electronic document and combining the electronic document with the electronic document.
According to claim 1,

The step of receiving the ID image input is,

extracting verification data from the ID image through optical character recognition; and,

Including; verifying the authenticity of the ID image using the extracted verification data;

When the identification image is verified, the facial recognition-based electronic signature service providing method, characterized in that it is determined whether the identification image and the facial image match
3. The method of claim 2,

When the Face ID is issued, deleting the extracted verification data; Facial recognition-based electronic signature service providing method comprising the.
According to claim 1,

The step of determining whether the identification image and the facial image are matched by deep learning analysis,

storing a training data set for the facial image; and,

Performing authentication of the face image using a deep learning model through training based on the data set; Facial recognition-based electronic signature service providing method comprising a.
According to claim 1,

The step of generating and issuing the Face ID comprises:

A method for providing a facial recognition-based electronic signature service, characterized in that the hash value of the electronic signature to be given to the electronic document is generated and issued together with the face ID.
6. The method of claim 5,

The step of generating and issuing the Face ID comprises:

A face recognition-based electronic signature service providing method, characterized in that issuing document identification information for the electronic document.
Contract proof that receives the signer's ID image input;

a signature management unit receiving a real-time facial image of the signer;

an authentication unit that determines whether the identification image and the facial image are matched by deep learning analysis;

an electronic signature issuing unit that generates and issues a face ID based on the face image when the ID image and the face image match; and,

A face recognition-based electronic signature service providing server comprising a; an electronic contract record unit that applies the Face ID as an electronic signature of an electronic document and combines it with the electronic document.
8. The method of claim 7,

The contract certificate is

Extracting verification data through optical character recognition from the ID image, and verifying the authenticity of the ID image using the extracted verification data,

The authentication unit,

When the identification image is verified, the facial recognition-based digital signature service providing server, characterized in that it is determined whether the identification image and the facial image match.
9. The method of claim 8,

The contract certificate is

When the Face ID is issued, the facial recognition-based digital signature service providing server, characterized in that the deletion of the extracted verification data.
8. The method of claim 7,

The authentication unit,

a data set storage unit for storing a training data set for a facial image; and,

A face recognition-based digital signature service providing server comprising a; a deep learning unit that authenticates the face image using a deep learning model through training based on the data set.
According to claim 1,

The electronic signature issuing unit generates a hash value of the electronic signature to be given to the electronic document and issues it together with the face ID.
12. The method of claim 11,

The electronic signature issuing unit, facial recognition-based electronic signature service providing server, characterized in that for issuing document identification information for the electronic document.