Facial recognition methods, facial recognition systems, terminal devices and storage media

Abstract

This application discloses a face recognition method, a face recognition system, a terminal device, and a storage medium. The face recognition method includes: acquiring a face image to be recognized and extracting initial face features corresponding to the face image; secretly segmenting the initial face features to obtain at least a first face feature to be recognized and a second face feature to be recognized; determining a first difference between the first face feature to be recognized and a plurality of first reference face features in a first server, and determining a second difference between the second face feature to be recognized and a plurality of second reference face features in a second server; the first and second reference face features are obtained by secretly segmenting reference face features corresponding to reference face images; and obtaining a recognition result corresponding to the face image to be recognized based on the first and second differences. The face recognition method of this application stores the segmented face features on different servers for computation, reducing the risk of face feature leakage.

Description

Technical Field

[0001] This application relates to the field of identity recognition technology, and in particular to a face recognition method, face recognition system, terminal device and storage medium. Background Technology

[0002] With the development of technology, the application scenarios of facial recognition and comparison are becoming increasingly widespread, such as facial recognition gates and facial recognition payment systems. The principle of facial recognition is to extract the facial features of the person to be identified from an image containing a face and compare them with existing facial features in a database.

[0003] The increasing number of facial data breaches poses a security threat to users as malicious actors exploit this data. Therefore, ensuring the security of user facial data while implementing facial recognition is crucial. Summary of the Invention

[0004] This application provides a face recognition method, a face recognition system, a terminal device, and a storage medium.

[0005] One technical solution adopted in this application is to provide a face recognition method, which includes:

[0006] A face image to be identified is acquired, and the initial face features corresponding to the face image to be identified are extracted; the initial face features are secretly segmented to obtain at least a first face feature to be identified and a second face feature to be identified; a first difference is determined between the first face feature to be identified and several first reference face features in a first server, and a second difference is determined between the second face feature to be identified and several second reference face features in a second server; the first reference face features and the second reference face features are obtained by secretly segmenting the reference face features corresponding to the reference face image; the recognition result corresponding to the face image to be identified is obtained based on the first difference and the second difference.

[0007] The determination of the first difference between the first face feature to be identified and a plurality of first reference face features in the first server includes: determining the first difference between the first face feature to be identified and each of the first reference face features in the first server, and obtaining the first difference corresponding to each of the first reference face features; the determination of the second difference between the second face feature to be identified and a plurality of second reference face features in the second server includes: determining the second difference between the second face feature to be identified and each of the second reference face features in the second server, and obtaining the second difference corresponding to each of the second reference face features.

[0008] The process of obtaining the recognition result corresponding to the face image to be recognized based on the first difference and the second difference includes: determining the correspondence between the first reference face features and the second reference face; determining the target first difference and the target second difference based on the correspondence; determining the similarity based on the target first difference and the target second difference, thereby obtaining multiple similarities corresponding to the face image to be recognized; and obtaining the recognition result corresponding to the face image to be recognized based on the multiple similarities.

[0009] The recognition result of the face image to be recognized is obtained based on multiple similarities, including: if any similarity meets the preset conditions, the recognition result is successful; if all similarities do not meet the preset conditions, the recognition result is unsuccessful.

[0010] The similarity determination based on the first target difference and the second target difference includes: reconstructing the first face feature to be identified, the second face feature to be identified, the first reference face feature and the second reference face feature based on the first target difference and the second target difference; and determining the similarity using the first face feature to be identified, the second face feature to be identified, the first reference face feature and the second reference face feature.

[0011] The similarity determination process, which utilizes the first face feature to be identified, the second face feature to be identified, the first reference face feature, and the second reference face feature, includes: determining a first sub-face feature of the same dimension in the first face feature to be identified and the second face feature to be identified; determining a second sub-face feature of the same dimension in the first reference face feature and the second reference face feature; and determining the similarity using the first sub-face feature and the second sub-face feature of the same dimension.

[0012] The process of secretly segmenting the initial facial features to obtain at least a first facial feature to be identified and a second facial feature to be identified includes: performing additive secret segmentation on the initial facial features to obtain at least a first facial feature to be identified and a second facial feature to be identified.

[0013] Another technical solution adopted in this application is to provide a face recognition system, which includes: a acquisition end for acquiring a face image to be recognized; a processing server connected to the acquisition end for receiving the face image to be recognized, extracting the initial face features corresponding to the face image to be recognized, and performing secret segmentation on the initial face features to obtain at least a first face feature to be recognized and a second face feature to be recognized; a first server connected to the processing server for receiving the first face feature to be recognized sent by the processing server and determining a first difference between the first face feature to be recognized and a plurality of first reference face features in the first server; and a second server connected to the processing server for receiving the second face feature to be recognized sent by the processing server and determining a second difference between the second face feature to be recognized and a plurality of second reference face features in the second server; wherein the first reference face feature and the second reference face feature are obtained by secretly segmenting the reference face features corresponding to the reference face image; the processing server is also used to receive the first difference and the second difference, and obtain the recognition result corresponding to the face image to be recognized based on the first difference and the second difference.

[0014] Another technical solution adopted in this application is to provide a terminal device, which includes a memory and a processor coupled to the memory; wherein the memory is used to store program data, and the processor is used to execute the program data to implement the face recognition method as described above.

[0015] Another technical solution adopted in this application is to provide a computer storage medium for storing program data, which, when executed by a computer, is used to implement the face recognition method described above.

[0016] The beneficial effects of this application are as follows: This application provides a face recognition method, a face recognition system, a terminal device, and a storage medium. By secretly segmenting the initial face features, at least a first face feature to be recognized and a second face feature to be recognized are obtained; this makes the first face feature to be recognized and the second face feature to be recognized more secure. Then, a first difference is determined by comparing them with several first reference face features in a first server, and a second difference is determined by comparing them with several second reference face features in a second server. Finally, the recognition result corresponding to the face image to be recognized is obtained based on the first difference and the second difference, thereby improving the security of face data in the entire face recognition process. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating an embodiment of the face recognition method provided in this application;

[0019] Figure 2 yes Figure 1 A flowchart illustrating a sub-step of S4 in the process;

[0020] Figure 3 yes Figure 2 Flowchart of sub-step S43 in the middle;

[0021] Figure 4 This is a flowchart illustrating an embodiment of the secret segmentation and storage of baseline facial features provided in this application;

[0022] Figure 5 This is a schematic diagram of another embodiment of the face recognition method provided in this application;

[0023] Figure 6 This is a schematic diagram of the structure of an embodiment of the face recognition system provided in this application;

[0024] Figure 7 This is a schematic diagram of the structure of an embodiment of the terminal device provided in this application;

[0025] Figure 8 This is a schematic diagram of the structure of an embodiment of the computer storage medium provided in this application. Detailed Implementation

[0026] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0027] The reference to "embodiment" in this application means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0028] The steps in the embodiments of this application are not necessarily processed in the order described. The steps can be rearranged, deleted, or added as needed. The step descriptions in the embodiments of this application are only optional combinations of sequences and do not represent all possible combinations of steps in the embodiments of this application. The order of steps in the embodiments should not be considered as a limitation of this application.

[0029] The terms "first," "second," etc., used in this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, or apparatuses.

[0030] Furthermore, although the terms "first," "second," etc., are used repeatedly in this application to describe various data (or devices, applications, instructions, or operations), these data (or devices, applications, instructions, or operations) should not be limited by these terms. These terms are only used to distinguish one type of data (or device, application, instruction, or operation) from another type of data (or device, application, instruction, or operation).

[0031] Please see Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the face recognition method provided in this application.

[0032] like Figure 1 As shown, the face recognition method provided in this application may include:

[0033] S1, acquire the face image to be identified, and extract the initial face features corresponding to the face image to be identified.

[0034] The face recognition method provided in this application is executed by a face recognition system. This face recognition system includes a data acquisition terminal, a processing server, a first server, and a second server. The first and second servers are non-colluding servers.

[0035] In some application scenarios, the process of acquiring the face image to be identified and extracting the initial facial features corresponding to the face image is executed by the acquisition end.

[0036] In some embodiments, the face recognition system acquires an image of the face to be recognized, i.e., captures a face image; the face recognition system extracts facial features from the image of the face to be recognized to obtain initial facial feature information X.

[0037] Optionally, before the facial feature extraction step, the facial recognition system can preprocess the facial image to be recognized, such as performing one or more image processing methods, such as image brightness adjustment, face detection, and face correction.

[0038] S2, perform secret segmentation on the initial facial features to obtain at least the first and second facial features to be identified.

[0039] In this process, the initial facial features are secretly segmented by addition to obtain at least the first and second facial features to be identified.

[0040] In some embodiments, the face recognition system performs additive secret segmentation on the initial face feature information to obtain at least a first face feature to be identified and a second face feature to be identified. Specifically, the face recognition system performs additive secret segmentation on the initial face feature information to obtain at least a first initial face feature segment and a second initial face feature segment.

[0041] In some embodiments, the face recognition system performs additive secret segmentation on the initial face feature information X to obtain the first segment X of the initial face feature. 1 Compared with the initial facial features, the second segment X 2 .

[0042] Among them, the first segment of the initial facial features X 1 Compared with the initial facial features, the second segment X 2 satisfy:

[0043]

[0044]

[0045] X = X 1 +X 2

[0046] Where X represents the initial facial feature information, X 1 X is the first segment of the initial facial features. 2 d represents the second segment of the initial face feature, and d represents the total number of features in each face feature segment.

[0047] In some application scenarios, the process of secretly segmenting the initial facial features is executed by the acquisition end.

[0048] In some application scenarios, after the acquisition end performs secret segmentation on the initial face, it sends the first face feature to be identified to the first server and the second face feature to be identified to the second server. The acquisition end then deletes the face image to be identified, the initial face feature information X, and the first segment X of the initial face feature. 1 And the second segment X of the initial facial features2 .

[0049] S3, determine the first difference between the first face feature to be identified and a plurality of first reference face features in the first server, and determine the second difference between the second face feature to be identified and a plurality of second reference face features in the second server.

[0050] The first and second reference facial features are obtained by secretly segmenting the reference facial features corresponding to the reference facial image. Specifically, the reference facial image is the facial image uploaded by the user to the processing server. In some application scenarios, the process of secretly segmenting the reference facial features corresponding to the reference facial image is executed on the processing server.

[0051] In some embodiments, the face recognition system calculates a first difference between the first facial features to be identified and several baseline facial features stored in a first server. Specifically, the face monitoring system calculates the corresponding difference between the initial facial feature first segment and the secret facial feature segment stored in the database of the first server.

[0052] In some embodiments, the face recognition system calculates a second difference between the second face feature to be identified and several baseline face features stored in a second server. Specifically, the face monitoring system calculates the corresponding difference between the initial second slice of face features and the secret slice of face features stored in the database in the second server.

[0053] The process of determining the first difference between the first facial feature to be identified and several first reference facial features in the first server is executed in the first server; the process of determining the second difference between the second facial feature to be identified and several second reference facial features in the second server is executed in the second server.

[0054] S4. Obtain the recognition result corresponding to the face image to be recognized based on the first difference and the second difference.

[0055] In some embodiments, the face recognition system compares a first difference calculated by a first server with a second difference calculated by a second server to complete the recognition of the face image to be recognized.

[0056] In some application scenarios, the process of obtaining the recognition result corresponding to the face image to be recognized based on the first difference and the second difference is executed by the processing server.

[0057] In some application scenarios, the processing server that executes the process of obtaining the recognition result corresponding to the face image to be recognized based on the first difference and the second difference, and the processing server that executes the process of secret segmentation of the reference face features corresponding to the reference face image, may not be the same server.

[0058] The above scheme, by secretly segmenting the initial facial features, obtains at least a first and a second facial feature to be identified; thus enhancing the security of the first and second facial features to be identified. Then, it determines a first difference with several first reference facial features in a first server and a second difference with several second reference facial features in a second server; finally, it obtains the recognition result corresponding to the facial image to be identified based on the first and second differences, thereby improving the security of facial data throughout the entire facial recognition process.

[0059] In some embodiments, the face recognition system may also perform the following steps during the execution of S3:

[0060] S31, determine the first difference between the first face feature to be identified and each first reference face feature in the first server, and obtain the first difference corresponding to each first reference face feature.

[0061] In some embodiments, the first difference E corresponding to each first baseline face feature obtained by the first server i satisfy:

[0062]

[0063] in, The first facial feature to be identified The corresponding first baseline facial features.

[0064] Where i = 1, 2, 3, ..., m. m is the total number of the first baseline facial features obtained by the first server.

[0065] S32, determine the second difference between the second face feature to be identified and each second reference face feature in the second server, and obtain the second difference corresponding to each second reference face feature.

[0066] In some embodiments, the second server obtains a second difference F corresponding to each second baseline face feature. i satisfy:

[0067]

[0068] in, The second facial feature to be identified The corresponding second baseline facial features.

[0069] Where i = 1, 2, 3, ..., m. m is the total number of the first baseline facial features obtained by the second server.

[0070] For the steps of obtaining the recognition result corresponding to the face image to be identified based on the first difference and the second difference, please refer to [link / reference]. Figure 2 , Figure 2 yes Figure 1 A flowchart illustrating a sub-step of S4.

[0071] like Figure 2 As shown, the steps for obtaining the recognition result corresponding to the face image to be recognized based on the first difference and the second difference may include:

[0072] S41, determine the correspondence between the first reference facial features and the second reference facial features.

[0073] Since the first and second reference face features are obtained based on the reference face features and then stored on the first and second servers respectively, it is necessary to establish a correspondence between the first and second reference face features so that the associated first and second reference face features can be found based on the correspondence during subsequent recognition.

[0074] S42, based on the correspondence, determine the first difference and the second difference of the target.

[0075] Since the first and second differences are obtained by comparing them with the first and second baseline facial features respectively, once the correspondence between the first and second baseline facial features is determined, the first and second differences will also have this correspondence. Therefore, the target first difference and target second difference can be determined based on the correspondence.

[0076] For example, multiple first reference facial features and multiple second reference facial features correspond to multiple first differences and second differences. By determining the correspondence between the first reference facial features and the second reference facial features, the target first difference and the target second difference can be determined from the multiple first differences and second differences.

[0077] In one embodiment, the first reference facial features include A1, A2, and A3, and the second reference facial features include B1, B2, and B3. The first differences obtained by comparing the first reference facial features with the first facial features to be identified are a1, a2, and a3, respectively, and the second differences obtained by comparing the second reference facial features with the second facial features to be identified are b1, b2, and b3, respectively.

[0078] In this embodiment, the face recognition system determines the relationship between the first reference face feature A1 and the corresponding second reference face feature B1, the relationship between the first reference face feature A2 and the corresponding second reference face feature B2, and the relationship between the first reference face feature A3 and the corresponding second reference face feature B3, thereby determining the relationship between the first difference a1 and the corresponding second difference b1, the first difference a2 and the corresponding second difference b2, and the first difference a3 and the corresponding second difference b3. This improves the accuracy of recognition.

[0079] The above steps determine the first and second differences of the target based on the correspondence, thereby improving the accuracy of facial feature information matching.

[0080] S43, similarity is determined based on the first difference and the second difference of the target, and then multiple similarities are obtained for the face image to be identified.

[0081] Specifically, based on the first difference and the second difference of the target, the first face feature to be identified, the second face feature to be identified, the first reference face feature, and the second reference face feature are restored.

[0082] For detailed steps on determining similarity based on the first and second differences of the target, please refer to [link / reference]. Figure 3 , Figure 3 yes Figure 2 A flowchart illustrating a sub-step in S43.

[0083] like Figure 3 As shown, it includes:

[0084] S431, determine the first sub-face feature of the same dimension in the first face feature to be identified and the second face feature to be identified.

[0085] In some embodiments, the face detection system determines a first sub-face feature of the same dimension in the first face feature to be identified and the second face feature to be identified, and restores the first face feature to be identified and the second face feature to be identified to obtain the corresponding first sub-face feature.

[0086] S432, determine the second sub-face feature of the same dimension in the first reference face feature and the second reference face feature.

[0087] In some embodiments, the face detection system determines a first sub-face feature of the same dimension in the first reference face feature and the second reference face feature, and restores the first reference face feature and the second reference face feature to obtain the corresponding second sub-face feature.

[0088] For example, the method for determining the first sub-face feature and the second sub-face feature in S431 and S432 can be that the processing server in the face recognition system receives the first feature and the second feature sent by the first server and the second server respectively, and the processing server determines the first sub-face feature and the second sub-face feature.

[0089] Among them, the first difference E i Second difference F i The relationship between the first sub-face feature and the second sub-face feature of the same dimension satisfies:

[0090]

[0091]

[0092] ...

[0093]

[0094] in, and and and These are the first sub-face features of the same dimension.

[0095] in, and and and These are the second sub-face features of the same dimension.

[0096] Where i = 1, 2, 3, ..., m.

[0097] S433, similarity is determined using the first and second sub-face features of the same dimension.

[0098] The processing server uses the first facial feature to be identified, the second facial feature to be identified, the first reference facial feature, and the second reference facial feature to determine the similarity.

[0099] In some embodiments, the processing server uses first and second sub-face features of the same dimension to determine face similarity.

[0100] Alternatively, the Euclidean distance between the first and second sub-face features of the same dimension can be used as the face similarity.

[0101] Alternatively, the squared value of the Euclidean distance between the first and second sub-face features of the same dimension can be used as the face similarity.

[0102] For example, the first sub-face feature X and the second sub-face feature Y to be identified i Similarity s i satisfy:

[0103] s i =(XY) i ) 2 =(x1-y i1 ) 2 +(x2-y i2 ) 2 +…+(x d -y id ) 2

[0104] Where x1 and y i1The first and second sub-face features are defined in the same dimension.

[0105] S44, based on multiple similarities, obtain the recognition result corresponding to the face image to be identified.

[0106] If any similarity meets the preset conditions, the recognition result is considered successful.

[0107] If all similarity scores do not meet the preset conditions, the recognition result will be "recognition failed".

[0108] In some embodiments, the preset condition is a similarity s i The similarity s is less than or equal to the comparison threshold t. i If the similarity s is less than or equal to the comparison threshold t, the face recognition system outputs a successful recognition result; if the similarity s is less than or equal to the comparison threshold t, the system outputs a successful recognition result. i If the value exceeds the comparison threshold t, the face recognition system outputs a recognition failure result.

[0109] For example, a face recognition system calculates similarity s i The difference between the comparison threshold t and the value of s, if any s i -t≤0 means the face recognition system has found a baseline face that matches the face image to be recognized, and the recognition is successful; if for all s i In general, all satisfy s i -t>0 means that the face recognition system did not find any baseline face information that matches the face image to be recognized, and the recognition failed.

[0110] Where i = 1, 2, ..., m.

[0111] Please see Figure 4 , Figure 4 This is a flowchart illustrating an embodiment of the secret segmentation and storage of baseline facial features provided in this application.

[0112] like Figure 4 As shown, the process of secretly segmenting and storing baseline facial features may include:

[0113] Send base database information.

[0114] In some embodiments, the user uploads a database containing reference face images to the client, and the client sends the database information to the feature extraction server.

[0115] Feature extraction is performed on the base database information to obtain base database features.

[0116] In some embodiments, the feature extraction server extracts features from the baseline face image in the base database information to obtain the base database feature information Y.

[0117] The base database feature information is denoted as Y = [Y1, Y2, ..., Y]. m ].

[0118] Where m represents the total number of faces in the base database feature information.

[0119] Among them, Y i =(y i1 ,y i2 ,…,y id ).

[0120] Where i = 1, 2, 3, ..., m; and d is the number of features in a single face image.

[0121] Additive secret partitioning is performed on the features of the base database to obtain the first secret partition and the second secret partition.

[0122] In some embodiments, the feature extraction server performs additive secret sharding on the base database feature information to obtain the first secret shard Y of the base database feature information. i 1 And the second secret fragment Y of the base database feature information i 2 , where i = 1, 2, ..., m.

[0123] Specifically, the content of the addition secret piecewise computation includes:

[0124] Y i =Y i 1 +Y i 2

[0125]

[0126]

[0127] Send the first secret fragment to the first computing server.

[0128] In some embodiments, the feature extraction server divides the base database feature information into a first secret fragment Y. i 1 Send to the first computing server.

[0129] Send the second secret fragment to the second computing server.

[0130] In some embodiments, the feature extraction server divides the base database feature information into a second secret fragment Y. i 2 Send to the second computing server.

[0131] After sending, delete the base database information, base database characteristics, and base database characteristic secret fragments.

[0132] In some embodiments, the client deletes the base database information, and the feature extraction server deletes the base database information and the base database feature information Y = [Y1, Y2, ..., Y]. m ], and the first secret fragment Y of the base database feature information i 1 And the second secret fragment Y of the base database feature information i 2 .

[0133] The above scheme involves the feature extraction server extracting the base database feature information, performing secret fragmentation, and sending it to two computing servers for storage. This reduces the risk of facial feature leakage, protects the security of the base database feature information storage, and also reduces the storage overhead of the base database feature information.

[0134] Please see Figure 5 , Figure 5 This is a schematic diagram of another embodiment of the face recognition method provided in this application.

[0135] like Figure 5 As shown, the facial recognition process may include:

[0136] Collect and capture images.

[0137] In some embodiments, the front-end acquisition device captures images, that is, it captures images of the face to be identified.

[0138] Extract facial features of the person to be identified.

[0139] In some embodiments, the front-end acquisition device extracts the facial feature X to be identified. For example, the facial feature to be identified corresponds to the initial facial feature described above.

[0140] Optionally, before extracting the facial features to be identified, the front-end acquisition device may perform an image preprocessing step on the facial image to be identified.

[0141] Optionally, the image preprocessing steps may include one or more image processing methods such as image contrast adjustment, image brightness adjustment, face detection, and face correction.

[0142] Addition secret segmentation is performed on the facial features to be identified to obtain the first secret segment and the second segment.

[0143] In some embodiments, the front-end acquisition device performs additive secret segmentation on the facial features to be identified, obtaining the first secret segment X of the facial features to be identified. 1 And the second secret segment X of the facial features to be identified 2 For example, the first secret segment of the facial feature to be identified and the second secret segment of the facial feature to be identified correspond to the first initial facial feature and the second initial facial feature, respectively.

[0144] The calculation of specific secret fragments includes:

[0145]

[0146]

[0147] X = X 1 +X 2

[0148] Send the first secret fragment to the first computing server.

[0149] In some embodiments, the front-end acquisition device divides the first secret slice X of the facial features to be identified. 1 The data is sent to the first computing server. The first computing server is the first server mentioned above.

[0150] Send the second secret fragment to the second computing server.

[0151] In some embodiments, the front-end acquisition device will divide the second secret slice X of the facial features to be identified. 2 The data is sent to the second computing server. The second computing server is the second server described above.

[0152] The front-end acquisition device completes the first secret segmentation of the facial features to be identified. 1 Send to the first computing server and the second secret fragment X of the facial features to be identified 2 After sending the data to the second computing server, the front-end acquisition device deletes the image of the face to be identified, the facial feature X to be identified, and the secret fragment X of the facial feature to be identified. 1 and X 2 This operation reduces the leakage of facial images and facial feature information while also reducing the storage overhead of the front-end acquisition device.

[0153] For each base database feature first secret slice and the face feature to be identified first secret slice, calculate the first difference.

[0154] In some embodiments, the first computing server stores the base database feature information in a first secret fragment Y. i 1 First Secret Slice of Facial Features to be Identified X 1 Perform the calculation of the first difference.

[0155] Specifically, the first computing server performs the following calculations:

[0156]

[0157] Among them, E i This is the first difference.

[0158] Where i = 1, 2, ..., m.

[0159] For each second secret slice of the base database features and the second secret slice of the face features to be identified, calculate the second difference.

[0160] In some embodiments, the second computing server stores the second secret fragment Y of the underlying database feature information. i 2 And the second secret segment X of the facial features to be identified 2 Perform the second difference calculation.

[0161] Specifically, the second computing server performs the following calculations:

[0162]

[0163] Among them, F i This is the second difference.

[0164] Where i = 1, 2, ..., m.

[0165] Send the first difference to the comparison server.

[0166] In some embodiments, the first computing server will calculate the first difference E. i Send to the comparison server.

[0167] The comparison server is the processing server mentioned above.

[0168] Send the second difference to the comparison server.

[0169] In some embodiments, the second computing server will use the second difference F i Send to the comparison server.

[0170] Calculate the facial feature similarity based on the first and second differences.

[0171] In some embodiments, the comparison server performs facial feature similarity calculations based on the results sent by the first computing server and the second computing server.

[0172] Optionally, the calculation of facial feature similarity can be performed by calculating the Euclidean distance between the facial features to be identified and the feature information in the database.

[0173] Alternatively, the facial feature similarity can be calculated by taking the square of the Euclidean distance between the facial features to be identified and the feature information in the database.

[0174] In some embodiments, the comparison server performs the following calculations to compute the facial features of the person to be identified and the feature information in the database:

[0175]

[0176]

[0177] ...

[0178]

[0179] The facial similarity between the facial feature X to be identified and the feature Y in the database satisfies:

[0180] s i =(x1-y i1 ) 2 +(x2-y i2 ) 2 +…+(x d -y id ) 2

[0181] Compare the facial feature similarity and the comparison threshold.

[0182] In some embodiments, the comparison server stores a facial feature similarity comparison threshold t, and the comparison between the facial feature similarity and the comparison threshold is used to calculate s. i -t.

[0183] If s i -t≤0, recognition successful; otherwise s i -t>0, recognition failed.

[0184] In some embodiments, if s is satisfied i -t≤0, the comparison server outputs the successful recognition result to the front-end acquisition device; if s is satisfied... i -t>0, the comparison server outputs the recognition failure result to the front-end acquisition device.

[0185] Please see Figure 6 , Figure 6 This is a schematic diagram of an embodiment of the face recognition system provided in this application.

[0186] The face recognition system 100 of this application embodiment includes a data acquisition terminal 11, a processing server 12, a first server 13, and a second server 14. The first server 13 and the second server 14 are non-colluding servers. The first server 13 and the second server 14 cannot collude to recover the original image feature data, thus achieving privacy protection for face feature information.

[0187] In some embodiments, the acquisition terminal 11 is used to acquire an image of a face to be identified. The processing server 12, connected to the acquisition terminal 11, is used to receive the image of the face to be identified, extract the initial facial features corresponding to the image, and perform secret segmentation on the initial facial features to obtain at least a first facial feature and a second facial feature to be identified. A first server 13, connected to the processing server 12, is used to receive the first facial feature to be identified sent by the processing server 12 and determine a first difference between the first facial feature to be identified and several first reference facial features in the first server 13. A second server 14, connected to the processing server 12, is used to receive the second facial feature to be identified sent by the processing server 12 and determine a second difference between the second facial feature to be identified and several second reference facial features in the second server. The processing server 12 is also used to receive the first difference and the second difference, and obtain the recognition result corresponding to the image of the face to be identified based on the first difference and the second difference.

[0188] In some embodiments, the processing server 12 may include a comparison server for receiving a first difference and a second difference sent from the first server 13 and the second server 14 respectively, and comparing the first difference and the second difference to obtain the recognition result corresponding to the face image to be recognized.

[0189] In some embodiments, the acquisition end 11 can be an image front-end acquisition device. After acquiring the image of the face to be recognized, the acquisition end 11 performs preprocessing on the image. Specifically, the acquisition end 11 performs operations such as image brightness and contrast adjustment, face detection, and face correction.

[0190] Please continue reading Figure 7 , Figure 7 This is a schematic diagram of the structure of an embodiment of the terminal device provided in this application. The terminal device 500 of this application embodiment includes a processor 51 and a memory 52.

[0191] The processor 51 and the memory 52 are connected to the bus. The memory 52 stores program data, and the processor 51 is used to execute the program data to implement the face recognition method described in the above embodiments.

[0192] In this embodiment, processor 51 can also be referred to as a CPU (Central Processing Unit). Processor 51 may be an integrated circuit chip with signal processing capabilities. Processor 51 can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor, or processor 51 can be any conventional processor.

[0193] This application also provides a computer storage medium; please refer to [link / reference]. Figure 8 , Figure 8 This is a schematic diagram of a computer storage medium according to an embodiment of the present application. The computer storage medium 600 stores program data 61, which is used to implement the face recognition method of the above embodiment when executed by the processor.

[0194] When the embodiments of this application are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0195] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A face recognition method, characterized by, The method comprises: obtaining a to-be-identified face image, and extracting initial face features corresponding to the to-be-identified face image; secretly fragmenting the initial face features to obtain at least first to-be-identified face features and second to-be-identified face features; determining first differences between the first to-be-identified face features and a plurality of first reference face features in a first server, and determining second differences between the second to-be-identified face features and a plurality of second reference face features in a second server; the first reference face features and the second reference face features are obtained by secretly fragmenting reference face features corresponding to reference face images; obtaining an identification result corresponding to the to-be-identified face image according to the first differences and the second differences; wherein the determination of the first differences between the first to-be-identified face features and the plurality of first reference face features in the first server comprises: determining the first differences between the first to-be-identified face features and each of the first reference face features in the first server to obtain the first difference corresponding to each of the first reference face features; the determination of the second differences between the second to-be-identified face features and the plurality of second reference face features in the second server comprises: determining the second differences between the second to-be-identified face features and each of the second reference face features in the second server to obtain the second difference corresponding to each of the second reference face features; the obtaining of the identification result corresponding to the to-be-identified face image according to the first differences and the second differences comprises: determining a correspondence between the first reference face features and the second reference face features; determining target first differences and target second differences based on the correspondence; determining a similarity based on the target first differences and the target second differences, and then obtaining a plurality of similarities corresponding to the to-be-identified face image; obtaining the identification result corresponding to the to-be-identified face image based on the plurality of similarities.

2. The method of claim 1, wherein, the obtaining of the identification result corresponding to the to-be-identified face image based on the plurality of similarities comprises: if any of the similarities meets a preset condition, the identification result is successful; if all of the similarities do not meet the preset condition, the identification result is unsuccessful.

3. The method of claim 1, wherein, the determination of the similarity based on the target first differences and the target second differences comprises: restoring the first to-be-identified face features, the second to-be-identified face features, the first reference face features, and the second reference face features based on the target first differences and the target second differences; determining the similarity by using the first to-be-identified face features, the second to-be-identified face features, the first reference face features, and the second reference face features.

4. The method of claim 3, wherein, the determination of the similarity by using the first to-be-identified face features, the second to-be-identified face features, the first reference face features, and the second reference face features comprises: determining first sub-face features of the same dimension in the first to-be-identified face features and the second to-be-identified face features; determining a second sub-personal feature of the same dimension in the first reference personal feature and the second reference personal feature; determining the similarity by using the first sub-personal feature and the second sub-personal feature of the same dimension.

5. The method of claim 1, wherein, The secret fragmentation of the initial personal feature includes: The secret fragmentation of the initial personal feature includes:

6. A face recognition system, characterized by, The face recognition system includes: a collection terminal for collecting a face image to be recognized; a processing server connected to the collection terminal, configured to receive the face image to be recognized, extract an initial personal feature corresponding to the face image to be recognized, and perform secret fragmentation on the initial personal feature to obtain at least a first personal feature to be recognized and a second personal feature to be recognized; a first server connected to the processing server, configured to receive the first personal feature to be recognized sent by the processing server, and determine a first difference between the first personal feature to be recognized and a plurality of first reference personal features in the first server; the determination of the first difference between the first personal feature to be recognized and the plurality of first reference personal features in the first server includes: determining a first difference between the first personal feature to be recognized and each of the first reference personal features in the first server to obtain the first difference corresponding to each of the first reference personal features; a second server connected to the processing server, configured to receive the second personal feature to be recognized sent by the processing server, and determine a second difference between the second personal feature to be recognized and a plurality of second reference personal features in the second server; the determination of the second difference between the second personal feature to be recognized and the plurality of second reference personal features in the second server includes: determining a second difference between the second personal feature to be recognized and each of the second reference personal features in the second server to obtain the second difference corresponding to each of the second reference personal features; The processing server is further configured to receive the first difference and the second difference, and obtain a recognition result corresponding to the face image to be recognized according to the first difference and the second difference; the obtaining of the recognition result corresponding to the face image to be recognized according to the first difference and the second difference includes: determining a corresponding relationship between the first reference personal feature and the second reference personal feature; determining a target first difference and a target second difference based on the corresponding relationship; determining a similarity based on the target first difference and the target second difference, and further obtaining a plurality of similarities corresponding to the face image to be recognized; obtaining the recognition result corresponding to the face image to be recognized based on the plurality of similarities.

7. A terminal device, characterized by comprising: The terminal device includes a memory and a processor coupled to the memory; The memory is configured to store program data, and the processor is configured to execute the program data to implement the face recognition method according to any one of claims 1 to 5.

8. A computer storage medium, characterized in that, The computer storage medium is configured to store program data, and the program data, when executed by a computer, is configured to implement the face recognition method according to any one of claims 1 to 5.

Images