Method and device for recognizing entrance of personnel in confidential place based on multi-modal biological recognition

Abstract

The invention provides a method and a device for recognizing entrance of personnel in a confidential place based on multi-modal biological recognition, and the method comprises the steps: carrying out the early-stage preparation, collecting the personal information and at least two biological characteristics of participants, constructing a multi-modal biological characteristic data set, processing the data through a data fusion technology, and carrying out the recognition of entrance of personnel in a confidential place. Extracting features, calculating a fusion feature vector, encrypting, and storing into a database to guarantee data security; performing admission verification: when a participant enters, acquiring multi-modal biological features according to instructions, and generating a fusion feature vector to be identified by a system; and performing identity verification, and performing identity judgment by calculating the Pearson correlation coefficient of the to-be-recognized feature vector and each user feature vector in the database. The method is combined with a multi-mode biological recognition fusion technology, the limitation of a single biological recognition technology is overcome, the robustness and the defense capability of the system are improved, meanwhile, personal privacy and data security are ensured, and a solid technical guarantee is provided for smooth proceeding of activities in confidential places.

Description

Technical Field

[0001] The present invention relates to the field of biometric technology, and in particular to a method and device for identifying personnel entering a confidential place based on multimodal biometrics. Background Art

[0002] Identification of people entering confidential events is an important part of ensuring event security and preventing information leakage. In highly sensitive events or those involving important secrets, accurate and rapid identification of the identities of people entering is crucial. This is not only related to the smooth progress of the event, but also directly related to the maintenance of national security or corporate interests.

[0003] Traditional entry identification methods mainly rely on manual verification of identity documents, invitation letters, etc., but this method has problems such as low efficiency, easy errors, and difficulty in anti-counterfeiting. With the development of science and technology, modern confidentiality activities have begun to adopt more advanced and efficient identification technologies, such as biometric technology (fingerprint, facial recognition, etc.), electronic invitation letters and QR code scanning, and smart cards based on RFID (radio frequency identification) technology. As the carrier of these technologies, entry identification devices have also shown a diversified development trend. From simple card readers and barcode scanners to smart terminals that integrate multiple identification functions, these devices not only improve identification efficiency, but also enhance identification accuracy and security. For example, some high-end confidentiality activities will use smart terminals that integrate facial recognition and fingerprint recognition functions to double-verify entrants to ensure that nothing goes wrong.

[0004] However, in the existing technology, a single biometric recognition is usually used. This single biometric recognition system is easily affected by factors such as environmental noise, clarity of photographic capture, individual changes, etc., and has high false recognition and missed recognition rates, which brings certain safety risks to the event venue.

[0005] Therefore, in order to ensure the safety of the event venue, ensure that the identity of each entrant is authentic and correct, and prevent illegal intrusion, these have become key problems that event organizers urgently need to solve, and have put forward an urgent need for the innovation and upgrading of trusted identity recognition technology. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides a method and device for identifying personnel entering confidential places based on multimodal biometrics, which combines multimodal biometrics fusion technology to achieve accurate, efficient and secure identity identification of personnel entering the venue. This method overcomes the limitations of a single biometric technology, improves the robustness and defense capabilities of the system, while also ensuring personal privacy and data security, providing a solid technical guarantee for the smooth progress of large-scale events.

[0007] The present invention is achieved through the following technical solutions:

[0008] On the one hand, the present invention provides a method for identifying personnel entering a confidential place based on multimodal biometrics, the method specifically comprising:

[0009] Preliminary preparation: collect the personal user information and at least two types of biometric data of the participants, establish the personal multimodal biometric data of the participants, use data fusion technology to extract data features, calculate feature vectors and fuse feature vectors of the personal multimodal biometric data of the participants, obtain multimodal biometric fusion feature vectors, and encrypt and store them in the database to ensure data security;

[0010] Entrance verification: When entering the event, participants follow the prompts to collect multimodal biometric features in sequence to obtain the multimodal biometric fusion feature vector of the user to be identified;

[0011] In subsequent processing, the Pearson correlation coefficient between the multimodal biometric fusion feature vector of the user to be identified and the biometric fusion feature vector of each user in the database is calculated, and based on the Pearson correlation coefficient, it is determined whether it is the real identity. If so, entry is allowed, otherwise entry is denied.

[0012] Furthermore, in the preliminary preparation, the multimodal biometric data includes: fingerprint, face, iris, gait and voiceprint, etc.;

[0013] Furthermore, in the preliminary preparation, the data fusion technology is used to extract data features, calculate feature vectors, and fuse feature vectors of the participants' personal multimodal biometric data to obtain a multimodal biometric vector data set, which specifically includes:

[0014] The multimodal biometric data is cleaned, denoised, and normalized to improve the quality of the biometric data.

[0015] Extracting features from the multimodal biometric data of each participant, obtaining feature vectors, and aggregating the obtained feature vectors to obtain a multimodal biometric vector data set for each participant;

[0016] Furthermore, the method of extracting features from the multimodal biometric data of each participant, obtaining feature vectors, and aggregating the obtained feature vectors to obtain a multimodal biometric vector data set for each participant also includes: in the multimodal biometric data, using word embedding or TF-IDF method to extract text features for text data, using convolutional neural network to extract image features for image data, and using Mel-frequency cepstral coefficients to extract audio features for audio data; selecting, reducing the dimension of each type of feature data, and removing redundant information, and then obtaining feature vectors for each type of data to obtain a multimodal biometric vector data set.

[0017] Perform cross-modal fusion based on the multimodal biometric feature vector dataset of each participant to obtain a multimodal biometric fusion feature vector of each participant;

[0018] Furthermore, the cross-modal fusion specifically refers to concatenating biometric feature vectors of different modalities into a longer vector.

[0019] Furthermore, the calculation process of the Pearson correlation coefficient includes:

[0020] Calculate the mean of the fused feature vector X of the user features to be identified respectively The mean of the fused feature vector Y of the biometric features of a single user in the database The calculation formula is:

[0021] Among them, X i and Y i Respectively represent the values ​​of the two eigenvectors in the i-th dimension, and n represents the dimension of the eigenvector;

[0022] Based on the mean of the fused feature vector, the standard deviation S of the features of the user to be identified and the biometric features of each user in the database are calculated respectively. X , S Y , which is used to measure the discreteness of feature data. The calculation formula is:

[0023]

[0024] Calculate the overall error Cov(X,Y) between two variables, that is, covariance, and the calculation formula is:

[0025]

[0026] The Pearson correlation coefficient r between the fused feature vector of the biometric feature to be identified and the fused feature vector of the biometric feature of each user in the database is calculated using the following formula:

[0027]

[0028] Furthermore, the Pearson correlation coefficient is used to determine whether the identity is real, and the criteria are:

[0029] When the Pearson correlation coefficient between two fused feature vectors is r∈[0.75, 1], it indicates that there is a significant relationship between the two, and the user to be identified is a real identity user; otherwise, it indicates that the user to be identified is a non-real identity user.

[0030] On the one hand, the present invention provides a device for identifying personnel entering a confidential place based on multimodal biometrics, the device comprising a display module, a multimodal biometrics collection module, a data processing and fusion center module, a security and privacy protection module, an electronic ticketing and identity authentication module, and an emergency response and fault handling module, wherein:

[0031] The display module is an LCD display screen, which is used to display prompt information of the multimodal biometric collection module, the data processing and fusion center module, the security and privacy protection module, the electronic ticketing and identity authentication module, and the emergency response and fault handling module;

[0032] The multimodal biometrics collection module is used to obtain multimodal biometrics of the participant user;

[0033] The data processing and fusion center module is used to process the data initialization processing and feature fusion of the multimodal biometric features of the participant user, determine whether the identity authentication is successful, and make corresponding decisions;

[0034] The security and privacy protection module is used to protect user privacy and prevent user privacy leakage;

[0035] The electronic ticketing and identity verification module is used by the participant user to purchase electronic tickets before entering the venue and to request identity verification when entering the venue;

[0036] The emergency response and fault handling module is used to respond to emergency events and handle abnormal faults to ensure the normal development of activities.

[0037] Furthermore, in the multimodal biometric collection module, the multimodal biometric collection module specifically adopts a unitized design and flexibly configures a variety of biometric feature collection devices, including: a face collection unit, a fingerprint collection unit, an iris collection unit, and a voiceprint collection unit, wherein:

[0038] The face acquisition unit uses a high-resolution camera and an infrared sensor to accurately capture and identify the facial features of the participants, and supports the acquisition of facial features under multiple angles and lighting conditions;

[0039] The fingerprint recognition unit uses a high-precision optical or capacitive fingerprint sensor to collect the fingerprint information of the participant to ensure the clarity and integrity of the fingerprint image;

[0040] The iris acquisition unit uses high-precision iris imaging technology to capture iris detail features, has the characteristics of high precision and high stability, and can effectively resist forgery attacks;

[0041] The voiceprint collection unit integrates a microphone array and uses speech recognition and processing technology to extract voiceprint features.

[0042] Furthermore, in the data processing and fusion center module, the data processing and fusion center module also includes a multimodal data processing unit, a data fusion algorithm unit, and a comparison and decision unit, wherein:

[0043] The multimodal data processing unit is responsible for receiving the biometric data from each biometric recognition module and performing preprocessing (such as denoising, normalization, etc.);

[0044] The data fusion algorithm unit adopts advanced data fusion technology to fuse multiple biometric data to form a more comprehensive and accurate identity information representation;

[0045] The comparison and decision unit compares the fused biometric data with the pre-stored database, determines whether the identity authentication is successful based on preset thresholds and rules, and makes corresponding decisions.

[0046] Further, in the security and privacy protection module, the security and privacy protection module also includes a data encryption and storage unit, an access control unit and a privacy policy and declaration unit, wherein:

[0047] The data encryption and storage unit uses encryption technology to encrypt, store and transmit biometric data to ensure data security;

[0048] The access control unit is set up with strict access rights management, so that only authorized personnel can access sensitive data and operate equipment;

[0049] The privacy policy and declaration unit clearly informs participants of the collection, use, storage and protection methods of personal information through the LCD display screen of the display module, respecting and protecting the privacy rights of participants.

[0050] Further, in the electronic ticketing and identity authentication module, the electronic ticketing and identity authentication module further includes an electronic ticketing unit, an identity authentication interface unit and an admission control unit, wherein:

[0051] The electronic ticketing unit is used to support online ticket purchase, electronic ticket verification and other functions; after purchasing electronic tickets through official channels, participants need to present the electronic ticket QR code for verification when entering the venue;

[0052] The identity authentication interface unit is connected to the multimodal biometric collection module and is used to receive identity authentication requests and transmit biometric data to the data processing and fusion center module for comparison;

[0053] The entry control unit controls the opening and closing of the entry gate or the access control system according to the identity verification result, so as to realize the orderly entry of personnel.

[0054] Further, in the emergency response and fault handling module, the emergency response and fault handling module also includes a fault monitoring and alarm unit, an emergency plan unit and a backup plan unit, wherein:

[0055] The aforementioned emergency plan, formulate a detailed emergency plan, including response measures for emergencies such as system failure, network interruption, large-scale personnel congestion, etc.;

[0056] The fault detection and alarm system is equipped with a fault detection mechanism. Once a device fault or abnormality is found, an alarm signal is immediately issued and relevant personnel are notified to handle it;

[0057] The backup plan sets up backup equipment or plans at key links to ensure that services can be quickly restored in emergency situations to avoid affecting the normal progress of activities.

[0058] Compared with the prior art, the present invention has the following advantages:

[0059] The present invention perfectly overcomes the limitations of a single biometric technology in the prior art by integrating a variety of biometric recognition technologies, thereby improving the robustness and defense capabilities of the system; at the same time, by optimizing the entry process, strengthening privacy protection and security mechanisms, and formulating emergency response mechanisms, the security of entry to event organizations is improved, providing a strong guarantee for the smooth conduct of confidential activities.

[0060] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0062] Figure 1 This is a flow chart of a method for identifying personnel entering a confidential place based on multimodal biometrics.

[0063] Figure 2 This is a structural diagram of a personnel entry identification device for confidential places based on multimodal biometrics.

[0064] Figure 3 This is a structural diagram of the multimodal biological collection module.

[0065] Figure 4This is the module structure diagram of the data processing and fusion center.

[0066] Figure 5 It is the structure diagram of the security and privacy protection module.

[0067] Figure 6 It is the structure diagram of the electronic ticketing and identity authentication module.

[0068] Figure 7 It is the structural diagram of the emergency response and fault handling module. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] In one embodiment, see Figure 1 , provides a method for identifying personnel entering a confidential place based on multimodal biometrics, the method specifically comprising the following steps:

[0071] Step S1: Preliminary preparation, collecting the personal user information and at least two types of biometric data of the participants, establishing the personal multimodal biometric data of the participants, using data fusion technology to extract data features, calculate feature vectors and fuse feature vectors of the personal multimodal biometric data of the participants, obtain multimodal biometric fusion feature vectors, and encrypt and store them in the database to ensure data security;

[0072] Furthermore, the multimodal biometric data includes fingerprints, faces, irises, gait, and voiceprints;

[0073] Furthermore, data fusion technology is used to extract data features, calculate feature vectors, and fuse feature vectors from the participants' personal multimodal biometric data to obtain a multimodal biometric vector data set, which specifically includes the following steps:

[0074] Step S101: performing operations such as cleaning, denoising and normalization on the multimodal biometric data to improve the quality of the biometric data.

[0075] Step S102: extracting features from the multimodal biometric data of each participant, obtaining feature vectors, and aggregating the obtained feature vectors to obtain a multimodal biometric vector data set for each participant;

[0076] Furthermore, in step S102, it also includes:

[0077] In multimodal biometric data, word embedding or TF-IDF method is used to extract text features for text data, convolutional neural network is used to extract image features for image data, and Mel frequency cepstral coefficient is used to extract audio features for audio data;

[0078] Each type of feature data is selected, dimensionally reduced, and redundant information is removed, and then the feature vectors of each type of data are obtained to obtain a multimodal biological feature vector dataset.

[0079] Step S103: performing cross-modal fusion based on the multimodal biometric feature vector dataset of each participant to obtain a multimodal biometric fusion feature vector of each participant;

[0080] Furthermore, in step S103, the cross-modal fusion specifically refers to concatenating biometric feature vectors of different modalities into a longer vector.

[0081] Step S2: Entry verification: When entering the event, participants follow the prompts to collect multimodal biometric features in sequence to obtain a multimodal biometric fusion feature vector of the user to be identified;

[0082] Step S3: Subsequent processing, calculating the Pearson correlation coefficient between the multimodal biometric fusion feature vector of the user to be identified and the biometric fusion feature vector of each user in the database, and judging whether it is the real identity based on the Pearson correlation coefficient. If so, entry is allowed, otherwise entry is denied.

[0083] Furthermore, in step S3, the calculation process of the Pearson correlation coefficient includes the following steps:

[0084] Step S301: Calculate the mean of the fused feature vector X of the user features to be identified respectively The mean of the fused feature vector Y of the biometric features of a single user in the database , the calculation formula is:

[0085] Among them, X i and Y i Respectively represent the values ​​of the two eigenvectors in the i-th dimension, and n represents the dimension of the eigenvector;

[0086] Step S302: Based on the mean of the fused feature vector, the standard deviation S of the feature of the user to be identified and the biometric feature of each user in the database are calculated respectively. X , S Y , which is used to measure the discreteness of feature data. The calculation formula is:

[0087]

[0088] Step S303: Calculate the overall error Cov(X, Y) between the two variables, that is, the covariance, and the calculation formula is:

[0089]

[0090] Step S304: Calculate the Pearson correlation coefficient r between the fused feature vector of the biometric feature to be identified and the fused feature vector of the biometric feature of each user in the database, and the calculation formula is:

[0091]

[0092] Furthermore, in step S3, the Pearson correlation coefficient is used to determine whether the identity is real, and the determination criteria are:

[0093] When the Pearson correlation coefficient between two fused feature vectors is r∈[0.75, 1], it indicates that there is a significant relationship between the two, and the user to be identified is a real identity user; otherwise, it indicates that the user to be identified is a non-real identity user.

[0094] In another embodiment, see Figure 2 , Figure 2 The structure diagram of a personnel entry identification device for confidential places based on multimodal biometrics, the device includes: a display module, a multimodal biometrics collection module, a data processing and fusion center module, a security and privacy protection module, an electronic ticketing and identity authentication module, and an emergency response and fault handling module;

[0095] The display module is an LCD display screen, which is used to display prompt information of the multimodal biometric collection module, the data processing and fusion center module, the security and privacy protection module, the electronic ticketing and identity authentication module, and the emergency response and fault handling module;

[0096] The multimodal biometrics collection module is used to obtain multimodal biometrics of the participant user;

[0097] For further information, see Figure 3 , Figure 3 : is a structural diagram of a multimodal biometric collection module. The multimodal biometric collection module specifically adopts a modular design and flexibly configures a variety of biometric feature collection devices, including: a face collection unit, a fingerprint collection unit, an iris collection unit, and a voiceprint collection unit, wherein:

[0098] The face acquisition unit uses a high-resolution camera and an infrared sensor to accurately capture and identify the facial features of the participants, and supports the acquisition of facial features under multiple angles and lighting conditions;

[0099] The fingerprint recognition unit uses a high-precision optical or capacitive fingerprint sensor to collect the fingerprint information of the participant to ensure the clarity and integrity of the fingerprint image;

[0100] The iris acquisition unit uses high-precision iris imaging technology to capture iris detail features, has the characteristics of high precision and high stability, and can effectively resist forgery attacks;

[0101] The voiceprint collection unit integrates a microphone array and uses speech recognition and processing technology to extract voiceprint features.

[0102] The data processing and fusion center module is used to process the data initialization processing and feature fusion of the multimodal biometric features of the participant user, determine whether the identity authentication is successful, and make corresponding decisions;

[0103] For further information, see Figure 4 , Figure 4 : is a structural diagram of a data processing and fusion center module, wherein the data processing and fusion center module also includes a multimodal data processing unit, a data fusion algorithm unit, and a comparison and decision unit, wherein:

[0104] The multimodal data processing unit is responsible for receiving the biometric data from each biometric recognition module and performing preprocessing (such as denoising, normalization, etc.);

[0105] The data fusion algorithm uses advanced data fusion technology to fuse multiple biometric data to form a more comprehensive and accurate identity information representation;

[0106] The comparison and decision module compares the fused biometric data with the pre-stored database, determines whether the identity authentication is successful based on preset thresholds and rules, and makes corresponding decisions.

[0107] The security and privacy protection module is used to protect user privacy and prevent user privacy leakage;

[0108] For further information, see Figure 5 , Figure 5 : is a structural diagram of a security and privacy protection module, wherein the security and privacy protection module further includes a data encryption and storage unit, an access control unit, and a privacy policy and statement sheet, wherein:

[0109] The data encryption and storage unit uses encryption technology to encrypt, store and transmit biometric data to ensure data security;

[0110] The access control unit is set up with strict access rights management, so that only authorized personnel can access sensitive data and operate equipment;

[0111] The privacy policy and declaration unit clearly informs participants of the collection, use, storage and protection methods of personal information through the LCD display screen of the display module, respecting and protecting the privacy rights of participants.

[0112] The electronic ticketing and identity verification module is used by the participant user to purchase electronic tickets before entering the venue and to request identity verification when entering the venue;

[0113] For further information, see Figure 6 , Figure 6 : is a structural diagram of an electronic ticketing and identity authentication module, wherein the electronic ticketing and identity authentication module further comprises an electronic ticketing module, an identity authentication interface unit and an admission control unit, wherein:

[0114] The electronic ticketing module is used to support online ticket purchase, electronic ticket verification and other functions; after the participants purchase the electronic ticket through the official channel, they need to present the electronic ticket QR code for verification when entering;

[0115] The identity authentication interface unit is connected to the multimodal biometric collection module and is used to receive identity authentication requests and transmit biometric data to the data processing and fusion center module for comparison;

[0116] The entry control unit controls the opening and closing of the entry gate or the access control system according to the identity verification result, so as to realize the orderly entry of personnel.

[0117] The emergency response and fault handling module is used to respond to emergency events and handle fault anomalies to ensure the normal development of activities;

[0118] For further information, see Figure 7 , Figure 7 : is a structural diagram of an emergency response and fault handling module, wherein the emergency response and fault handling module further includes a fault monitoring and alarm unit, an emergency plan unit and a backup plan unit, wherein:

[0119] The aforementioned emergency plan, formulate a detailed emergency plan, including response measures for emergencies such as system failure, network interruption, large-scale personnel congestion, etc.;

[0120] The fault detection and alarm system is equipped with a fault detection mechanism. Once a device fault or abnormality is found, an alarm signal is immediately issued and relevant personnel are notified to handle it;

[0121] The backup plan sets up backup equipment or plans at key links to ensure that services can be quickly restored in emergency situations to avoid affecting the normal progress of activities.

[0122] Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent substitutions for some of the technical features therein; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying personnel entering a confidential place based on multimodal biometrics, characterized in that: The method comprises: Preliminary preparation: collect the personal user information and at least two types of biometric data of the participants, establish the personal multimodal biometric data of the participants, use data fusion technology to extract data features, calculate feature vectors and fuse feature vectors of the personal multimodal biometric data of the participants, obtain multimodal biometric fusion feature vectors, and encrypt and store them in the database; Entrance verification: When entering the event, participants follow the prompts to collect multimodal biometric features in sequence. Based on the collected multimodal biometric features, a multimodal biometric fusion feature vector of the user to be identified is obtained; In subsequent processing, the Pearson correlation coefficient between the multimodal biometric fusion feature vector of the user to be identified and the biometric fusion feature vector of each user in the database is calculated, and based on the Pearson correlation coefficient, it is determined whether it is the real identity. If so, entry is allowed, otherwise entry is denied.

2. The method for identifying personnel entering a confidential place based on multimodal biometrics as claimed in claim 1, characterized in that: The multimodal biometric data includes fingerprints, faces, irises, gait, and voiceprints; The method uses data fusion technology to extract data features, calculate feature vectors, and fuse feature vectors on the personal multimodal biometric data of the participants to obtain a multimodal biometric vector data set, and also includes: Performing operations such as cleaning, denoising, and normalization on the multimodal biometric data to improve the quality of the biometric data; Extracting features from the multimodal biometric data of each participant, obtaining feature vectors, and aggregating the obtained feature vectors to obtain a multimodal biometric vector data set for each participant; Based on the multimodal biometric feature vector dataset of each participant, cross-modal fusion is performed to obtain a multimodal biometric fusion feature vector of each participant.

3. A method for identifying personnel entering a confidential place based on multimodal biometrics as claimed in claim 2, characterized in that: The method of extracting features from the multimodal biometric data of each participant, obtaining feature vectors, and aggregating the obtained feature vectors to obtain a multimodal biometric vector dataset of each participant also includes: In multimodal biometric data, word embedding and TF-IDF methods are used to extract text features for text data, convolutional neural networks are used to extract image features for image data, and Mel-frequency cepstral coefficients are used to extract audio features for audio data; Each type of feature data is selected, dimensionally reduced, and redundant information is removed, and then the feature vectors of each type of data are obtained to obtain a multimodal biological feature vector dataset.

4. The method for identifying personnel entering a confidential place based on multimodal biometrics as claimed in claim 1, characterized in that: The calculation process of the Pearson correlation coefficient includes: Calculate the mean of the fused feature vector X of the user features to be identified respectively The mean of the fused feature vector Y of the biometric features of a single user in the database The calculation formula is: Among them, X i and Y i Respectively represent the values ​​of the two eigenvectors in the i-th dimension, and n represents the dimension of the eigenvector; Based on the mean of the fused feature vector, the standard deviation S of the features of the user to be identified and the biometric features of each user in the database are calculated respectively. X , S Y , used to measure the discreteness of feature data, the calculation formula is: Calculate the overall error Cov(X,Y) between two variables, that is, covariance, and the calculation formula is: The Pearson correlation coefficient r between the fused feature vector of the biometric feature to be identified and the fused feature vector of the biometric feature of each user in the database is calculated using the following formula:

5. A method for identifying personnel entering a confidential place based on multimodal biometrics as claimed in claim 4, characterized in that: The determination of whether it is a real identity based on the Pearson correlation coefficient is based on the following criteria: when the Pearson correlation coefficient r∈[0.75, 1], the user to be identified is a real identity user, otherwise it is a non-real identity user.

6. A device for identifying personnel entering a confidential place based on multimodal biometrics, characterized in that: The device includes a display module, a multimodal biometric collection module, a data processing and fusion center module, a security and privacy protection module, an electronic ticketing and identity authentication module, and an emergency response and fault handling module, wherein: The display module is an LCD display screen, which is used to display the prompt information of the multimodal biometric collection module, the data processing and fusion center module, the security and privacy protection module, the electronic ticketing and identity authentication module, and the emergency response and fault handling module; The multimodal biometrics collection module is used to obtain multimodal biometrics of the participant user; The data processing and fusion center module is used to process the data initialization processing and feature fusion of the multimodal biometric features of the participant user, determine whether the identity authentication is successful, and make corresponding decisions; The security and privacy protection module is used to protect user privacy and prevent user privacy leakage; The electronic ticketing and identity verification module is used by the participant user to purchase electronic tickets before entering the venue and to request identity verification when entering the venue; The emergency response and fault handling module includes a fault monitoring and alarm unit, an emergency plan unit and a backup plan unit, which are used to respond to emergency events and handle fault anomalies to ensure the normal development of activities.

7. The device for identifying personnel entering a confidential place based on multimodal biometrics as claimed in claim 6, characterized in that: The multimodal biometric collection module specifically adopts a modular design and flexibly configures a variety of biometric feature collection devices, including: a face collection unit, a fingerprint collection unit, an iris collection unit, and a voiceprint collection unit, wherein: The face acquisition unit uses a high-resolution camera and an infrared sensor to accurately capture and identify the facial features of the participants, and supports the acquisition of facial features under multiple angles and lighting conditions; The fingerprint recognition unit uses a high-precision optical or capacitive fingerprint sensor to collect the fingerprint information of the participant to ensure the clarity and integrity of the fingerprint image; The iris acquisition unit uses high-precision iris imaging technology to capture iris detail features, has the characteristics of high precision and high stability, and can effectively resist forgery attacks; The voiceprint collection unit integrates a microphone array and uses speech recognition and processing technology to extract voiceprint features.

8. The device for identifying personnel entering a confidential place based on multimodal biometrics as claimed in claim 6, characterized in that: The data processing and fusion center module also includes a multimodal data processing unit, a data fusion algorithm unit, and a comparison and decision unit, wherein: The multimodal data processing unit is responsible for receiving the biometric data from each biometric recognition module and performing pre-processing; The data fusion algorithm uses advanced data fusion technology to fuse multiple biometric data to form a more comprehensive and accurate identity information representation; The comparison and decision module compares the fused biometric data with the pre-stored database, determines whether the identity authentication is successful based on preset thresholds and rules, and makes corresponding decisions.

9. The device for identifying personnel entering a confidential place based on multimodal biometrics as claimed in claim 6, characterized in that: The security and privacy protection module also includes a data encryption and storage unit, an access control unit, and a privacy policy and statement sheet, wherein: The data encryption and storage unit uses encryption technology to encrypt, store and transmit biometric data to ensure data security; The access control unit is set up with strict access rights management, so that only authorized personnel can access sensitive data and operate equipment; The privacy policy and declaration unit clearly informs participants of the collection, use, storage and protection methods of personal information through the LCD display screen of the display module, respecting and protecting the privacy rights of participants.

10. The device for identifying personnel entering a confidential place based on multimodal biometrics as claimed in claim 6, characterized in that: The electronic ticketing and identity verification module further comprises an electronic ticketing module, an identity verification interface unit and an admission control unit, wherein: The electronic ticketing module is used to support online ticket purchase, electronic ticket verification and other functions; The identity authentication interface unit is connected to the multimodal biometric collection module and is used to receive identity authentication requests and transmit biometric data to the data processing and fusion center module for comparison; The entry control unit controls the opening and closing of the entry gate or the access control system according to the identity verification result, so as to realize the orderly entry of personnel.

Images