A method and apparatus for personnel verification based on distributed homomorphic encryption

Abstract

This invention relates to the field of data security technology and provides a method and apparatus for personnel verification based on distributed homomorphic encryption. In this embodiment, when verifying personnel using homomorphic encryption technology, the homomorphic key pair is sent to the homomorphic encryption collection node and the homomorphic operation verification node through the homomorphic key management and distribution node. Then, the homomorphic encryption collection node collects data, encrypts it into groups, and sends them to the paired homomorphic operation verification node. Finally, the homomorphic operation verification node performs encrypted calculation and verification, enabling homomorphic encryption and encrypted calculation to be performed on mobile terminal devices, reducing the computational pressure on the server side.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of data security, and particularly relates to a personnel verification method and device based on distributed homomorphic encryption. BACKGROUND

[0002] The development and maturity of machine learning technology provide an effective method for mining useful information from massive data. Traditional machine learning usually needs to aggregate data for model training to realize detection, classification and prediction of events. However, due to commercial competition and other factors, industrial production data has certain privacy, and not all manufacturers are willing to share their data.

[0003] In this case, homomorphic encryption technology, as a method that can perform homomorphic encryption on encrypted data without decryption, has been increasingly popular. However, due to the high cost of existing homomorphic encryption technology, it is basically used on the server side, which leads to excessive server computing pressure and greatly limits the practical application of homomorphic encryption technology in many cases. SUMMARY

[0004] Therefore, the application provides a personnel verification method and device based on distributed homomorphic encryption, which is used to solve the problem that the use of homomorphic encryption technology on the server side leads to excessive server computing pressure and limits the use scenarios of homomorphic encryption technology.

[0005] The application is implemented as follows:

[0006] The application provides a personnel verification method based on distributed homomorphic encryption, applied to a node cluster, the node cluster comprising a homomorphic key management and distribution node, a homomorphic encryption collection node and a homomorphic operation verification node, wherein the homomorphic encryption collection node is paired with the homomorphic operation verification node in advance, and the method comprises the following steps:

[0007] generating a homomorphic key pair by the homomorphic key management and distribution node, and sending the homomorphic key pair to the paired homomorphic encryption collection node and homomorphic operation verification node;

[0008] collecting face image information of a target person by the homomorphic encryption collection node, extracting feature codes from the face image information, and determining target code text data corresponding to the target person;

[0009] homomorphic encrypting the target code text data according to the homomorphic key pair to obtain corresponding ciphertext data, and grouping the ciphertext data;

[0010] sending the grouped ciphertext data to the homomorphic operation verification node paired with the homomorphic encryption collection node;

[0011] The homomorphic operation verification node performs comparison verification on the ciphertext data and a preset ciphertext comparison library, determines a comparison result of the ciphertext data, and decrypts the comparison result according to the homomorphic key pair, to determine a personnel verification result according to the decrypted comparison result.

[0012] Optionally, the homomorphic encryption collection node is paired with the homomorphic operation verification node in advance through a key exchange protocol, including:

[0013] The homomorphic encryption collection node and the homomorphic operation verification node negotiate and select some common key parameters, the homomorphic encryption collection node generates a collection key according to the key parameters, and sends the collection key to the homomorphic operation verification node;

[0014] The homomorphic operation verification node generates a verification key according to the key parameters, and sends the verification key to the homomorphic encryption collection node;

[0015] The homomorphic encryption collection node determines a shared key through the received verification key, the homomorphic operation verification node determines a shared key through the received collection key, and the homomorphic encryption collection node and the homomorphic operation verification node complete pairing through the shared key.

[0016] Optionally, the pairing mode of the homomorphic encryption collection node and the homomorphic operation verification node includes 1:1, 1:n, m:n, wherein the n is the number of homomorphic operation verification nodes, and the m is the number of homomorphic encryption collection nodes.

[0017] Optionally, the generation of the homomorphic key pair by the homomorphic key management and distribution node and the sending of the homomorphic key pair to the paired homomorphic encryption collection node and homomorphic operation verification node include:

[0018] The homomorphic key management and distribution node generates a key pair according to a preset homomorphic encryption algorithm, sends a public key of the key pair to the homomorphic encryption collection node, and simultaneously sends a private key of the key pair to the homomorphic operation verification node paired with the homomorphic encryption collection node.

[0019] Optionally, the preset homomorphic encryption algorithm includes:

[0020] Paillier encryption algorithm, ElGamal encryption algorithm, RSA homomorphic encryption algorithm, BFV (Brakerski-Fan-Vercauteren) homomorphic encryption algorithm, CKKS (Cheon-Kim-Kim-Song) homomorphic encryption algorithm.

[0021] Optionally, after the target code text data corresponding to the target personnel is determined, the method further includes:

[0022] Random noise is added to the target code text data to increase the data security of the target code text data.

[0023] Optionally, the feature code extraction on the face image information to determine the target code text data corresponding to the target personnel includes:

[0024] The corresponding feature code is extracted from the face image information through a preset face feature extraction algorithm, and the face feature extraction algorithm includes a feature point calibration algorithm, a local binary pattern algorithm, a principal component analysis algorithm, or a deep learning algorithm.

[0025] The feature code is associated with identity information of the target personnel to generate target code text data, and the identity information includes an ID of the target personnel.

[0026] Optionally, the grouping of the ciphertext data includes:

[0027] The number of bytes of the target code text data is determined, and the ciphertext data is padded and grouped according to a preset grouping data size.

[0028] The application further provides a personnel verification device based on distributed homomorphic encryption, applied to a node cluster, the node cluster including a homomorphic key management distribution node, a homomorphic encryption collection node, and a homomorphic operation verification node, wherein the homomorphic encryption collection node is paired with the homomorphic operation verification node in advance, and the device includes:

[0029] A distribution unit is configured to generate a homomorphic key pair through the homomorphic key management distribution node, and send the homomorphic key pair to the paired homomorphic encryption collection node and homomorphic operation verification node.

[0030] A collection unit is configured to collect face image information of a target personnel through the homomorphic encryption collection node, extract a feature code from the face image information, and determine target code text data corresponding to the target personnel.

[0031] An encryption and grouping unit is configured to homomorphically encrypt the target code text data according to the homomorphic key pair to obtain corresponding ciphertext data, and group the ciphertext data.

[0032] A sending unit is configured to send the grouped ciphertext data to the homomorphic operation verification node paired with the homomorphic encryption collection node.

[0033] The verification unit is used by the homomorphic operation verification node to compare and verify the ciphertext data with a preset ciphertext comparison library, determine the comparison result of the ciphertext data, decrypt the comparison result according to the homomorphic key, and determine the personnel verification result based on the decrypted comparison result.

[0034] The following beneficial technical effects can be achieved by implementing the technical solution disclosed herein:

[0035] By optimizing distributed homomorphic encryption in a node cluster, that is, by sending homomorphic key pairs to homomorphic encryption acquisition nodes and homomorphic operation verification nodes through homomorphic key management and distribution nodes, the homomorphic encryption acquisition nodes collect data and encrypt and send it to the paired homomorphic operation verification nodes, and then the homomorphic operation verification nodes perform encrypted computation and verification. This enables homomorphic encryption and encrypted computation to be performed on mobile terminal devices during personnel verification, reducing the computational pressure on the server side. Attached Figure Description

[0036] The accompanying drawings illustrate exemplary embodiments of the present disclosure and, together with the description thereof, serve to explain the principles of the present disclosure. These drawings are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this specification.

[0037] Figure 1 A flowchart illustrating the method provided in this application embodiment;

[0038] Figure 2 This is a structural diagram of the device provided in the embodiments of this application; Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0040] It should be noted that, where there is no conflict, the embodiments and features described in this disclosure can be combined with each other. This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0041] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0042] like Figure 1As shown, a flowchart of a personnel verification method based on distributed homomorphic encryption provided by an embodiment of the present application is provided. The method is applied to a node cluster, which includes a homomorphic key management distribution node, a homomorphic encryption collection node, and a homomorphic operation verification node. The homomorphic encryption collection node is paired with the homomorphic operation verification node in advance. The flowchart can include the following steps:

[0043] In step S101, the homomorphic key management distribution node generates a homomorphic key pair and sends the homomorphic key pair to the paired homomorphic encryption collection node and homomorphic operation verification node.

[0044] In this embodiment, the homomorphic key management distribution node can generate a key pair according to a preset homomorphic encryption algorithm, send the public key of the key pair to the homomorphic encryption collection node to enable the homomorphic encryption collection node to encrypt the collected data, and send the private key of the key pair to the homomorphic operation verification node paired with the homomorphic encryption collection node to enable the homomorphic operation verification node to process the encrypted data and then decrypt the data.

[0045] For example, in the Paillier homomorphic encryption algorithm, the process of determining the key pair is as follows:

[0046] 1. The homomorphic key management distribution node determines two parameters a and b, and the product of a and b is denoted as n.

[0047] 2. Calculate λ = lcm(a-1, b-1), where lcm represents the least common multiple.

[0048] 3. Select a random integer g, which satisfies the order of g is n^2.

[0049] 4. Calculate μ = (L(g^λ mod n^2))^{-1} mod n, where L(x) = (x-1) / n.

[0050] 5. Determine the public key of the key pair as (n, g) and send the public key to the homomorphic encryption collection node; determine the private key of the key pair as (λ, μ) and send the private key to the homomorphic operation verification node paired with the homomorphic encryption collection node.

[0051] As another embodiment of the present application, the preset homomorphic encryption algorithm includes:

[0052] Paillier encryption algorithm, ElGamal encryption algorithm, RSA homomorphic encryption algorithm, BFV (Brakerski-Fan-Vercauteren) homomorphic encryption algorithm, CKKS (Cheon-Kim-Kim-Song) homomorphic encryption algorithm.

[0053] As another embodiment of the present application, the homomorphic encryption collection node is paired with the homomorphic operation verification node through a key exchange protocol in advance, comprising:

[0054] The homomorphic encryption collection node and the homomorphic operation verification node negotiate and select some common secret key parameters, the homomorphic encryption collection node generates a collection secret key according to the secret key parameters, and sends the collection secret key to the homomorphic operation verification node;

[0055] The homomorphic operation verification node generates a verification secret key according to the secret key parameters, and sends the verification secret key to the homomorphic encryption collection node;

[0056] The homomorphic encryption collection node determines a shared secret key through the received verification secret key, the homomorphic operation verification node determines a shared secret key through the received collection secret key, and the homomorphic encryption collection node and the homomorphic operation verification node complete pairing through the shared secret key.

[0057] In the above pairing process, the pairing mode of the homomorphic encryption collection node and the homomorphic operation verification node includes 1:1, 1:n, m:n, wherein the n is the number of homomorphic operation verification nodes, and the m is the number of homomorphic encryption collection nodes.

[0058] In this embodiment, the homomorphic encryption collection node can be a mobile device for face collection, and the homomorphic operation verification node can be a mobile device for verification. According to the number of each device, the collection device and the verification device can be paired in a 1:1, 1:many, and many:many mode.

[0059] Step S102, collecting face image information of a target person through the homomorphic encryption collection node, extracting feature codes from the face image information, and determining target code data corresponding to the target person.

[0060] In this embodiment, the homomorphic encryption collection node is provided with a collection device such as a camera, and the ambient light around the homomorphic encryption collection node can be ensured to be sufficient to avoid excessive shadows or reflections. By adjusting the angle and distance of the collection device, the collection device can collect the front fixed area, and when collecting the face, the appropriate zoom can be maintained. After collecting the face image information, the quality of the collected face image can also be checked to ensure that the image is clear, without blur or overexposure.

[0061] As another embodiment of the present application, after determining the target code data corresponding to the target person, it further comprises:

[0062] Random noise is added to the target code text data to increase the data security of the target code text data.

[0063] In this embodiment, noise can be introduced by adding random values to the code text data, for example, using a random number generator to generate some random numbers and adding them to each element in the code text data. Noise can also be introduced by multiplying a random number in the code text data, and noise can also be introduced by blurring the code text data. There are many ways to add random noise to the target code text data, which are not limited by the present application.

[0064] As another embodiment of the present application, the feature code extraction of the face image information and the determination of the target code text data corresponding to the target personnel include:

[0065] The corresponding feature code is extracted from the face image information by a preset face feature extraction algorithm, and the face feature extraction algorithm includes a feature point calibration algorithm, a local binary pattern algorithm, a principal component analysis algorithm or a deep learning algorithm.

[0066] The feature code is associated with the identity information of the target personnel to generate target code text data, and the identity information includes the ID of the target personnel.

[0067] In this embodiment, the extraction of the feature code can be performed by the following steps:

[0068] 1. Taking the feature point calibration algorithm as an example, a suitable face key point detector can be selected, such as dlib, MTCNN, etc. These detectors can detect the key points in the face image.

[0069] 2. The selected face key point detector is used to detect the key point positions in the face image. Generally, these key points can include eye, eyebrow, nose, mouth and other regions.

[0070] 3. The feature code is determined based on the key point positions, for example, the image blocks of the regions around the key points are extracted, and then these image blocks are converted into a fixed length vector, and the vector is determined as the feature code.

[0071] In this embodiment, after the homomorphic encryption collection node collects the face image information, an ID is assigned to the target personnel corresponding to the face image information. After determining the feature code of the face image information of the target personnel, the feature code and the ID are combined to generate target code text data.

[0072] Step S103, homomorphic encryption of the target code text data is performed according to the homomorphic key to obtain corresponding ciphertext data, and the ciphertext data is grouped.

[0073] In the embodiment, the homomorphic encryption collection node encrypts the target ciphertext data according to the face image information after generating the target ciphertext data, and then encrypts the ciphertext data through the public key sent by the homomorphic key management distribution node. The ciphertext data is then grouped. There are many grouping methods, for example, fixed-length grouping, that is, grouping the ciphertext data according to its fixed bit length or byte number. This method can make the size of the group consistent, facilitating processing and calculation.

[0074] The ciphertext data can also be grouped according to the type of feature code, for example, grouping the feature codes of the same organ into the same type, and grouping the homomorphic encrypted data according to the relevant logic. For example, grouping the feature codes corresponding to the eyes, nose, ears and other organs to meet specific query or calculation requirements.

[0075] In the above embodiment, the grouping of the ciphertext data includes:

[0076] Determining the byte number of the target ciphertext data, and padding the ciphertext data according to the preset grouping data size.

[0077] In the embodiment, the byte number of the target ciphertext data does not change after homomorphic encryption. Therefore, the byte number of the target ciphertext data is determined, and the ciphertext data is padded according to the byte number of the target ciphertext data. For example, the byte number of the target ciphertext data is 500 bytes, and the preset grouping data size is 128 bytes. The ciphertext data can be filled with blank characters to 512 bytes, and then divided into 512 / 128=4 groups.

[0078] Step S104, sending the grouped ciphertext data to the homomorphic operation verification node paired with the homomorphic encryption collection node.

[0079] In the embodiment, the homomorphic encryption collection node and the homomorphic operation verification node have been paired in advance. The homomorphic encryption collection node and the homomorphic operation verification node perform data transmission and authentication through the shared secret key determined when pairing, to ensure the validity of the ciphertext data.

[0080] Step S105, the homomorphic operation verification node compares and verifies the ciphertext data with the preset ciphertext comparison library, determines the comparison result of the ciphertext data, decrypts the comparison result according to the homomorphic key pair, and determines the personnel verification result according to the decrypted comparison result.

[0081] In the embodiment, the ciphertext comparison library can be established by historical ciphertext data. For example, the homomorphic encryption collection node obtains multiple face image information from the face information library of other systems, generates corresponding ciphertext data from the face image information, and sends the ciphertext data to the homomorphic operation verification node. The homomorphic operation verification node establishes the ciphertext comparison library from the ciphertext data generated from the face information library of other systems.

[0082] In the embodiment, the comparison of the ciphertext data can be implemented by using the calculation function provided in the homomorphic encryption technology. For example, the verification is implemented by homomorphic equality testing. The specific homomorphic equality testing method can be different due to the used homomorphic encryption algorithm. For example, when the Paillier encryption algorithm is used, the equality of two ciphertexts can be determined by comparing whether the homomorphic addition results of the two ciphertexts are equal. When the BFV homomorphic encryption scheme is used, the equality of two ciphertexts can be determined by checking whether the polynomial coefficients of the two ciphertexts are equal.

[0083] After the comparison result is determined by each homomorphic operation verification node, the comparison result is decrypted. The user can determine the personnel verification result according to the decrypted comparison result. For example, when the comparison result is that the matching degree between the personnel image data and the ID of a user is greater than 90%, the target personnel is determined to be the user corresponding to the ID.

[0084] At this point, the process shown in the flowchart is completed. Figure 1

[0085] In the embodiment of the application, a personnel verification method and device based on distributed homomorphic encryption are provided. In the embodiment, when the personnel verification is implemented by using the homomorphic encryption technology, the homomorphic key pair is sent to the homomorphic encryption collection node and the homomorphic operation verification node by the homomorphic key management distribution node. Then, the data is collected and encrypted by the homomorphic encryption collection node, and is sent to the paired homomorphic operation verification node. Then, the homomorphic operation verification node performs the secret calculation and verification, so that the homomorphic encryption and secret calculation can be implemented on the mobile terminal device, and the calculation pressure of the server side is reduced.

[0086] In another preferred embodiment, the personnel verification method based on distributed homomorphic encryption can be improved. For example, each region in the face image information can be split, and different tasks can be assigned to each homomorphic encryption collection node. For example, the homomorphic encryption collection node A obtains the feature code of the eye region in the face image information. Similarly, the homomorphic encryption collection node B obtains the feature code of the nose region, and the homomorphic encryption collection node C obtains the feature code of the mouth region. In this way, when the homomorphic operation verification node establishes the ciphertext comparison library in advance, only the ciphertext data of the corresponding region needs to be stored, so as to reduce the data amount of each homomorphic operation verification node.​

[0087] As Figure 2 shown, the embodiment of the present application also provides a personnel verification device based on distributed homomorphic encryption, applied to a node cluster, the node cluster comprising a homomorphic key management distribution node, a homomorphic encryption collection node and a homomorphic operation verification node, wherein the homomorphic encryption collection node is paired with the homomorphic operation verification node in advance, and the device comprises:

[0088] a distribution unit 201 configured to generate a homomorphic key pair by the homomorphic key management distribution node, and send the homomorphic key pair to the paired homomorphic encryption collection node and homomorphic operation verification node;

[0089] a collection unit 202 configured to collect face image information of a target person by the homomorphic encryption collection node, extract feature codes from the face image information, and determine target code data corresponding to the target person;

[0090] an encryption grouping unit 203 configured to homomorphically encrypt the target code data according to the homomorphic key pair to obtain corresponding ciphertext data, and group the ciphertext data;

[0091] a sending unit 204 configured to send the grouped ciphertext data to the homomorphic operation verification node paired with the homomorphic encryption collection node;

[0092] a verification unit 205 configured to compare and verify the ciphertext data with a preset ciphertext comparison library by the homomorphic operation verification node, determine a comparison result of the ciphertext data, decrypt the comparison result according to the homomorphic key pair, and determine a personnel verification result according to the decrypted comparison result.

[0093] As another embodiment of the present application, the homomorphic encryption collection node is paired with the homomorphic operation verification node in advance through a key exchange protocol, comprising:

[0094] the homomorphic encryption collection node and the homomorphic operation verification node negotiate and select some common key parameters, the homomorphic encryption collection node generates a collection key according to the key parameters, and sends the collection key to the homomorphic operation verification node;

[0095] the homomorphic operation verification node generates a verification key according to the key parameters, and sends the verification key to the homomorphic encryption collection node;

[0096] The homomorphic encryption collection node determines a shared key through the received verification key, the homomorphic operation verification node determines a shared key through the received collection key, and the homomorphic encryption collection node and the homomorphic operation verification node complete pairing through the shared key;

[0097] The pairing mode of the homomorphic encryption collection node and the homomorphic operation verification node includes 1:1, 1:n, m:n, wherein the n is the number of homomorphic operation verification nodes, and the m is the number of homomorphic encryption collection nodes;

[0098] The homomorphic key management distribution node generates a homomorphic key pair and sends the homomorphic key pair to the paired homomorphic encryption collection node and homomorphic operation verification node, including:

[0099] The homomorphic key management distribution node generates a key pair according to a preset homomorphic encryption algorithm, sends the public key of the key pair to the homomorphic encryption collection node, and simultaneously sends the private key of the key pair to the homomorphic operation verification node paired with the homomorphic encryption collection node;

[0100] The preset homomorphic encryption algorithm includes:

[0101] Paillier encryption algorithm, ElGamal encryption algorithm, RSA homomorphic encryption algorithm, BFV (Brakerski-Fan-Vercauteren) homomorphic encryption algorithm, CKKS (Cheon-Kim-Kim-Song) homomorphic encryption algorithm;

[0102] After determining the target code text data corresponding to the target personnel, the method further includes:

[0103] Random noise is added to the target code text data to increase the data security of the target code text data;

[0104] The feature code extraction on the face image information and the determination of the target code text data corresponding to the target personnel include:

[0105] A preset face feature extraction algorithm is used to extract the corresponding feature code from the face image information, and the face feature extraction algorithm includes a feature point calibration algorithm, a local binary pattern algorithm, a principal component analysis algorithm, or a deep learning algorithm;

[0106] The feature code is associated with the identity information of the target personnel to generate target code text data, and the identity information includes the ID of the target personnel;

[0107] The ciphertext data is grouped, including:

[0108] The number of bytes of the target ciphertext data is determined, and the ciphertext data is padded and grouped according to a preset grouping data size.

[0109] The personnel verification method based on distributed homomorphic encryption provided in the above embodiments and the personnel verification device based on distributed homomorphic encryption provided by the personnel verification method based on distributed homomorphic encryption can collect data by a homomorphic encryption collection node (collection device) and group the data, and can perform secret state calculation and verification by a homomorphic operation verification node (verification device), so that homomorphic encryption and secret state calculation can be performed on a mobile terminal device, and the calculation pressure of a server end is reduced.

[0110] The embodiments also disclose a computer device, which comprises a processor and a memory, and the memory stores at least one instruction loaded and executed by the processor to implement the personnel verification method based on distributed homomorphic encryption.

[0111] In addition, in the implementation of the personnel verification device based on distributed homomorphic encryption in the above examples, the logical division of each program module is only illustrative, and in actual application, the above functions can be completed by different program modules according to needs, for example, for the configuration requirements of corresponding hardware or the convenience of software implementation, that is, the internal structure of the device for optimizing the face picture quality evaluation model is divided into different program modules to complete all or part of the functions described above.

[0112] In the description of the present specification, the description of the terms "one embodiment / way", "some embodiments / ways", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment / way or example are included in at least one embodiment / way or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment / way or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments / ways or examples. In addition, the person skilled in the art can combine and combine the different embodiments / ways or examples described in the present specification and the features of the different embodiments / ways or examples without contradiction.

[0113] In addition, the terms "first", "second", etc. are used only for the purpose of description, and should not be understood as indicating or implying relative importance or an indicated number of technical features. Thus, features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "a plurality of" is at least two, such as two, three, etc., unless otherwise explicitly and specifically limited.

[0114] Those skilled in the art should understand that the above embodiments are only for the purpose of clearly illustrating the present disclosure, and are not intended to limit the analysis scope of the present disclosure. Based on the above disclosure, other changes or modifications can also be made by those skilled in the art, and these changes or modifications are still within the analysis scope of the present disclosure.

Claims

1. A personnel verification method based on distributed homomorphic encryption, characterized in that, The application is applied to a node cluster containing a homomorphic key management distribution node, a homomorphic encryption collection node and a homomorphic operation verification node, wherein the homomorphic encryption collection node is paired with the homomorphic operation verification node in advance, and the method comprises the following steps: A homomorphic key pair is generated by the homomorphic key management distribution node, and the homomorphic key pair is sent to the paired homomorphic encryption collection node and homomorphic operation verification node; The homomorphic encryption collection node collects the face image information of a target person, extracts the feature code of the face image information, and determines the target code text data corresponding to the target person; The target code text data is homomorphic encrypted according to the homomorphic key pair to obtain corresponding ciphertext data, and the ciphertext data is grouped; The grouped ciphertext data is sent to the homomorphic operation verification node paired with the homomorphic encryption collection node; The homomorphic operation verification node compares and verifies the ciphertext data with a preset ciphertext comparison library, determines the comparison result of the ciphertext data, decrypts the comparison result according to the homomorphic key pair, and determines the personnel verification result according to the decrypted comparison result.

2. The method of claim 1, wherein, The homomorphic encryption collection node and the homomorphic operation verification node are paired through a key exchange protocol, comprising the following steps: The homomorphic encryption collection node and the homomorphic operation verification node negotiate and select some common key parameters, the homomorphic encryption collection node generates a collection key according to the key parameters, and sends the collection key to the homomorphic operation verification node; The homomorphic operation verification node generates a verification key according to the key parameters, and sends the verification key to the homomorphic encryption collection node; The homomorphic encryption collection node determines a shared key through the received verification key, and the homomorphic operation verification node determines a shared key through the received collection key, and the homomorphic encryption collection node and the homomorphic operation verification node complete pairing through the shared key.

3. The method of claim 1, wherein, The pairing mode of the homomorphic encryption collection node and the homomorphic operation verification node includes 1:1, 1:n, m:n, wherein n is the number of homomorphic operation verification nodes, and m is the number of homomorphic encryption collection nodes.

4. The method of claim 1, wherein, The homomorphic key pair is generated by the homomorphic key management distribution node, and the homomorphic key pair is sent to the paired homomorphic encryption collection node and homomorphic operation verification node, comprising the following steps: The homomorphic key management distribution node generates a key pair according to a preset homomorphic encryption algorithm, sends the public key of the key pair to the homomorphic encryption collection node, and sends the private key of the key pair to the homomorphic operation verification node paired with the homomorphic encryption collection node.

5. The method of claim 4, wherein, The preset homomorphic encryption algorithm comprises: Paillier encryption algorithm, ElGamal encryption algorithm, RSA homomorphic encryption algorithm, BFV (Brakerski-Fan-Vercauteren) homomorphic encryption algorithm, CKKS (Cheon-Kim-Kim-Song) homomorphic encryption algorithm.

6. The method of claim 1, wherein, The determining of the target code text data corresponding to the target personnel further comprises: Adding random noise in the target code text data to increase the data security of the target code text data.

7. The method of claim 1, wherein, The feature code extraction from the face image information and the determining of the target code text data corresponding to the target personnel comprise: Extracting the corresponding feature code from the face image information through a preset face feature extraction algorithm, the face feature extraction algorithm comprising a feature point calibration algorithm, a local binary pattern algorithm, a principal component analysis algorithm or a deep learning algorithm; Associating the feature code with the identity information of the target personnel to generate target code text data, the identity information comprising an ID of the target personnel.

8. The method of claim 1, wherein, The grouping of the ciphertext data comprises: Determining the number of bytes of the target code text data, and filling and grouping the ciphertext data according to a preset grouping data size. 9.A personnel verification apparatus based on distributed homomorphic encryption, characterized by, The device is applied to a node cluster, the node cluster comprising a homomorphic key management distribution node, a homomorphic encryption collection node and a homomorphic operation verification node, wherein the homomorphic encryption collection node is paired with the homomorphic operation verification node in advance, and the device comprises: A distribution unit configured to generate a homomorphic key pair through the homomorphic key management distribution node, and send the homomorphic key pair to the paired homomorphic encryption collection node and homomorphic operation verification node; A collection unit configured to collect face image information of a target personnel through the homomorphic encryption collection node, extract a feature code from the face image information, and determine target code text data corresponding to the target personnel; An encryption and grouping unit configured to homomorphic encrypt the target code text data according to the homomorphic key pair to obtain corresponding ciphertext data, and group the ciphertext data; A sending unit configured to send the grouped ciphertext data to the homomorphic operation verification node paired with the homomorphic encryption collection node; A verification unit configured to compare and verify the ciphertext data with a preset ciphertext comparison library by the homomorphic operation verification node, determine a comparison result of the ciphertext data, decrypt the comparison result according to the homomorphic key pair, and determine a personnel verification result according to the decrypted comparison result.

10. The apparatus of claim 9, wherein, The homomorphic encryption collection node and the homomorphic operation verification node are paired through a key exchange protocol, comprising: The homomorphic encryption collection node and the homomorphic operation verification node negotiate and select some common key parameters, the homomorphic encryption collection node generates a collection key according to the key parameters, and sends the collection key to the homomorphic operation verification node; The homomorphic operation verification node generates a verification key according to the key parameters, and sends the verification key to the homomorphic encryption collection node; The homomorphic encryption collection node determines a shared key through the received verification key, the homomorphic operation verification node determines a shared key through the received collection key, and the homomorphic encryption collection node and the homomorphic operation verification node complete pairing through the shared key. The pairing mode of the homomorphic encryption collection node and the homomorphic operation verification node includes 1:1, 1:n, m:n, wherein the n is the number of homomorphic operation verification nodes, and the m is the number of homomorphic encryption collection nodes; The homomorphic key management and distribution node generates a homomorphic key pair and sends the homomorphic key pair to the paired homomorphic encryption collection node and homomorphic operation verification node, including: The homomorphic key management and distribution node generates a key pair according to a preset homomorphic encryption algorithm, sends the public key of the key pair to the homomorphic encryption collection node, and simultaneously sends the private key of the key pair to the homomorphic operation verification node paired with the homomorphic encryption collection node; The preset homomorphic encryption algorithm includes: Paillier encryption algorithm, ElGamal encryption algorithm, RSA homomorphic encryption algorithm, BFV (Brakerski-Fan-Vercauteren) homomorphic encryption algorithm, CKKS (Cheon-Kim-Kim-Song) homomorphic encryption algorithm; After determining the target code text data corresponding to the target personnel, the method further includes: Adding random noise to the target code text data to increase the data security of the target code text data; The feature code extraction on the face image information and the determination of the target code text data corresponding to the target personnel include: Extracting the corresponding feature code from the face image information through a preset face feature extraction algorithm, and the face feature extraction algorithm includes a feature point calibration algorithm, a local binary pattern algorithm, a principal component analysis algorithm, or a deep learning algorithm; Associating the feature code with the identity information of the target personnel to generate target code text data, and the identity information includes the ID of the target personnel; The grouping of the ciphertext data includes: Determining the byte number of the target code text data, and performing padding grouping on the ciphertext data according to a preset grouping data size.

Images