CN113918912A

CN113918912A - Identity authentication method, system, equipment and medium based on brain print recognition

Info

Publication number: CN113918912A
Application number: CN202111183097.0A
Authority: CN
Inventors: 姚一鸣; 徐亮
Original assignee: OneConnect Financial Technology Co Ltd Shanghai
Current assignee: OneConnect Smart Technology Co Ltd; OneConnect Financial Technology Co Ltd Shanghai
Priority date: 2021-10-11
Filing date: 2021-10-11
Publication date: 2022-01-11

Abstract

The application relates to an artificial intelligence technology, and provides an identity authentication method, system, medium and device based on brain print recognition, wherein the method comprises the following steps: acquiring identity associated information for identifying the identity of a user, and performing first identity authentication on the user; if the first identity authentication is passed, displaying a plurality of candidate pictures; acquiring an electroencephalogram generated by a user according to an authentication picture, wherein the authentication picture is determined by the user from a plurality of candidate pictures; recognizing the electroencephalogram signals to obtain a first recognition result; recognizing the electroencephalogram signals to obtain a second recognition result; carrying out weighted summation on the first recognition result and the second recognition result to obtain an identity authentication comprehensive probability value; comparing the comprehensive probability value of the identity authentication with a preset threshold value to finish the second identity authentication of the user; and if the comprehensive probability value of the identity authentication is greater than the preset threshold value, the identity authentication passes the second time. The accuracy of single recognition is improved by integrating recognition results of different methods.

Description

Identity authentication method, system, equipment and medium based on brain print recognition

Technical Field

The invention relates to the field of artificial intelligence, in particular to an identity authentication method, system, medium and equipment based on brain print recognition.

Background

Identity authentication means that the identity of the other party is confirmed by a certain means, so that the physical identity of the other party is ensured to correspond to the virtual digital identity. User identity authentication plays a very important role in asset security, data protection, risk control and the like in various industries such as finance, communication, internet and the like.

Today, the user identity authentication can be generally divided into three categories: 1. certain secret information set or known by the user, such as a password, etc.; 2. a certain physical object held by the user, such as an identity card, a smart card, a storage medium and the like, in which private information is stored; 3. a certain biological characteristic specific to the user, such as a face, a voice, a fingerprint, an iris, etc. However, the above authentication methods all have risks of being attacked or leaked. Although there are many authentication methods for authenticating a user, generally, only one of the authentication methods is used for authentication, and the accuracy of a single authentication method needs to be improved.

Disclosure of Invention

The invention provides an identity authentication method, system, medium and equipment based on brain print recognition, and mainly aims to solve the problem that the traditional authentication mode is low in safety.

In order to achieve the above object, the present invention provides an identity authentication method based on brain print recognition, comprising:

acquiring identity associated information for identifying the identity of a user, and performing first identity authentication on the user based on the identity associated information;

if the first identity authentication is passed, displaying a plurality of candidate pictures through a display module; the candidate pictures comprise a plurality of interference pictures which are randomly generated and at least one pre-stored authentication picture provided by a user;

acquiring an electroencephalogram signal generated by the user according to an authentication picture through an electroencephalogram signal acquisition module, wherein the authentication picture is determined by the user from a plurality of candidate pictures;

recognizing the electroencephalogram signal by utilizing a pre-trained kernel function-based support vector machine model for recognizing the identity of the user to obtain a first recognition result; the first recognition result represents a probability value that the user belongs to a certain identity;

recognizing the electroencephalogram signal by utilizing a pre-trained brain print recognition model for recognizing the identity of the user to obtain a second recognition result; the second recognition result represents a probability value that the user belongs to a certain identity;

carrying out weighted summation on the first recognition result and the second recognition result to obtain an identity authentication comprehensive probability value;

comparing the comprehensive probability value of the identity authentication with a preset threshold value to finish the second identity authentication of the user; and if the comprehensive probability value of the identity authentication is larger than the preset threshold value, passing the identity authentication for the second time.

Optionally, the method further comprises: preprocessing the electroencephalogram signals, comprising:

denoising the electroencephalogram signal by using a recess filter to obtain a first electroencephalogram signal;

filtering the first electroencephalogram signal by using a low-pass filter to obtain a second electroencephalogram signal;

and denoising the second electroencephalogram signal through an independent component analysis algorithm to obtain a third electroencephalogram signal.

Optionally, the recognizing the electroencephalogram signal by using a pre-trained kernel function-based support vector machine model for recognizing the identity of the user to obtain a first recognition result includes:

acquiring a power spectral density feature vector of the electroencephalogram signal;

and inputting the power spectral density feature vector into the support vector machine model to obtain a first identification result.

Optionally, the acquiring the power spectral density feature vector of the electroencephalogram signal includes:

segmenting the electroencephalogram signals according to the frequency distribution of the human brain to obtain electroencephalogram signals of a plurality of wave bands;

acquiring the power spectral density of the electroencephalogram signals of the multiple wave bands;

and (4) forming a feature vector by the power spectral density corresponding to the electroencephalogram signals of each wave band.

Optionally, the acquiring power spectral densities of the electroencephalogram signals of the plurality of bands includes:

and calculating the power spectral density corresponding to the electroencephalogram signal of each wave band through a welch algorithm.

Optionally, the recognizing the electroencephalogram signal by using a pre-trained brain print recognition model for recognizing the identity of the user to obtain a second recognition result includes:

performing feature extraction on the electroencephalogram signals by using a pre-trained brain print recognition model to obtain brain print features;

and identifying the electroencephalogram signal according to the brain print characteristics to obtain a second identification result.

Optionally, performing feature extraction on the electroencephalogram signal, including:

extracting the frequency characteristics of the electroencephalogram signals by utilizing a time convolution module in a brain print recognition model;

extracting the spatial features of the electroencephalogram signals by utilizing a spatial convolution module in a brain print recognition model;

fusing the frequency characteristic and the space characteristic by utilizing a separable convolution module in a brain print recognition model to obtain a brain print characteristic;

the recognizing the electroencephalogram signal according to the brain print characteristics comprises the following steps:

and identifying the electroencephalogram signals by utilizing a classification module in a brain print identification model based on the brain print characteristics.

In order to achieve the above object, the present invention provides an identity authentication system based on brain print recognition, comprising:

the first authentication module is used for acquiring identity associated information for identifying the identity of a user and performing first identity authentication on the user based on the identity associated information;

the image display module is used for displaying a plurality of candidate images when the first identity authentication is passed; the candidate pictures comprise a plurality of interference pictures which are randomly generated and at least one pre-stored authentication picture provided by a user;

the electroencephalogram signal acquisition module is used for acquiring an electroencephalogram signal generated by the user according to an authentication picture, wherein the authentication picture is determined by the user from a plurality of candidate pictures;

the first identification module is used for identifying the electroencephalogram signal by utilizing a pre-trained kernel function-based support vector machine model for identifying the identity of the user to obtain a first identification result; the first recognition result represents a probability value that the user belongs to a certain identity;

the second identification module is used for identifying the electroencephalogram signal by utilizing a pre-trained brain print identification model for identifying the identity of the user to obtain a second identification result; the second recognition result represents a probability value that the user belongs to a certain identity;

the comprehensive authentication module is used for weighting and summing the first identification result and the second identification result to obtain an identity authentication comprehensive probability value;

the second authentication module is used for comparing the comprehensive probability value of the identity authentication with a preset threshold value so as to finish the second identity authentication of the user; and if the comprehensive probability value of the identity authentication is larger than the preset threshold value, passing the identity authentication for the second time.

In order to achieve the above object, the present invention provides a computer device, which includes a memory and a processor, wherein the memory stores computer readable instructions, and the computer readable instructions, when executed by the processor, cause the processor to execute the steps of the identity authentication method based on brain print recognition.

To achieve the above object, the present invention provides a storage medium storing computer readable instructions, which when executed by one or more processors, cause the one or more processors to perform the steps of the method for authenticating identity based on brain print recognition.

As described above, the identity authentication method, system, medium and device based on brain print recognition provided by the present invention have the following beneficial effects:

the invention relates to an identity authentication method based on brain print recognition, which obtains the identity authentication result by obtaining the associated information of a client to be authenticated and authenticating the identity of the client to be authenticated based on the associated information; the identity of the client is authenticated for the first time through the associated information of the client; outputting a plurality of candidate pictures when the first identity authentication is passed; determining an authentication picture corresponding to the client to be authenticated from a plurality of candidate pictures; the authentication picture is selected for the second authentication on the basis of the first authentication, so that the authentication accuracy can be improved; acquiring an electroencephalogram signal of the client to be authenticated when watching the authentication picture; segmenting the electroencephalogram signals according to the human brain frequency, and acquiring the power spectral density corresponding to each band of electroencephalogram signals; the electroencephalogram signals are segmented to obtain a plurality of sub-bands, and the electroencephalogram signals of each sub-band are integrated to reflect the difference between different individuals under the same or different states, so that the electroencephalogram signals can be better identified. Recognizing the electroencephalogram signals by utilizing a pre-trained kernel function-based support vector machine model for recognizing the identity of a user and the power spectral density corresponding to the electroencephalogram signals of each wave band to obtain a first recognition result; recognizing the electroencephalogram signal by utilizing a pre-trained brain print recognition model based on a neural network and used for recognizing the identity of the user and the electroencephalogram signal to obtain a second recognition result; and obtaining a comprehensive identification result through the first identification result and the second identification result, namely, the identity of the client can be considered to be authenticated for the third time. The invention carries out comprehensive authentication on the client by combining the support vector machine model and the brain print recognition model based on the neural network, avoids the phenomenon of misjudgment caused by a single authentication mode as much as possible, and improves the accuracy of identity recognition.

Drawings

Fig. 1 is a schematic diagram of an application environment of an identity authentication method based on brain print recognition according to an embodiment of the present invention;

FIG. 2 is a flowchart of an identity authentication method based on brain print recognition according to an embodiment of the present invention;

FIG. 3 is a flow chart of preprocessing of electroencephalogram signals in one embodiment of the present invention;

FIG. 4 is a flow chart of obtaining a first recognition result according to an embodiment of the present invention;

FIG. 5 is a flowchart illustrating obtaining a power spectral density feature vector of the electroencephalogram signal according to an embodiment of the present invention;

fig. 6 is a block diagram of an identity authentication system based on voiceprint recognition according to an embodiment of the present invention.

The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.

Detailed Description

It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The invention provides an identity authentication method based on brain print recognition, which is applied to an electronic device 1. Fig. 1 is a schematic diagram of an application environment of the identity authentication method according to an embodiment of the present invention.

In the present embodiment, the electronic device 1 may be a terminal device having an arithmetic function, such as a server, a smart phone, a tablet computer, a portable computer, or a desktop computer.

The electronic device 1 includes: a processor 12, a memory 11, an imaging device 13, a network interface 14, and a communication bus 15.

The memory 11 includes at least one type of readable storage medium. The at least one type of readable storage medium may be a non-volatile storage medium such as a flash memory, a hard disk, a multimedia card, a card-type memory 11, and the like. In some embodiments, the readable storage medium may be an internal storage unit of the electronic apparatus 1, such as a hard disk of the electronic apparatus 1. In other embodiments, the readable storage medium may also be an external memory 11 of the electronic device 1, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the electronic device 1.

In the present embodiment, the readable storage medium of the memory 11 is generally used for storing the identity authentication program 10 and the like installed in the electronic device 1. The memory 11 may also be used to temporarily store data that has been output or is to be output.

The processor 12 may be a Central Processing Unit (CPU), microprocessor or other data Processing chip in some embodiments, and is used for executing program codes stored in the memory 11 or Processing data, such as executing the authentication method program 10.

The imaging device 13 may be a part of the electronic device 1 or may be independent of the electronic device 1. In some embodiments, the electronic device 1 is a terminal device having a camera, such as a smart phone, a tablet computer, a portable computer, and the like, and then the camera 13 is the camera of the electronic device 1. In other embodiments, the electronic device 1 may be a server, and the camera 13 is independent from the electronic device 1 and connected to the electronic device 1 through a network, for example, the camera 13 is installed in a specific location, such as an office or a monitoring area, and captures a real-time image of a target entering the specific location in real time, and transmits the captured real-time image to the processor 12 through the network.

The network interface 14 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface), and is typically used to establish a communication link between the electronic apparatus 1 and other electronic devices.

The communication bus 15 is used to realize connection communication between these components.

Fig. 1 only shows the electronic device 1 with components 11-15, but it is to be understood that not all of the shown components are required to be implemented, and that more or fewer components may alternatively be implemented.

Optionally, the electronic device 1 may further include a user interface, the user interface may include an input unit such as a Keyboard (Keyboard), a voice input device such as a microphone (microphone) or other equipment with a voice recognition function, a voice output device such as a sound box, a headset, etc., and optionally the user interface may further include a standard wired interface, a wireless interface.

Optionally, the electronic device 1 may further comprise a display, which may also be referred to as a display screen or a display unit. In some embodiments, the display device may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an Organic Light-Emitting Diode (OLED) touch device, or the like. The display is used for displaying information processed in the electronic apparatus 1 and for displaying a visualized user interface.

Optionally, the electronic device 1 further comprises a touch sensor. The area provided by the touch sensor for the user to perform touch operation is called a touch area. Further, the touch sensor described herein may be a resistive touch sensor, a capacitive touch sensor, or the like. The touch sensor may include not only a contact type touch sensor but also a proximity type touch sensor. Further, the touch sensor may be a single sensor, or may be a plurality of sensors arranged in an array, for example.

The area of the display of the electronic device 1 may be the same as or different from the area of the touch sensor. Optionally, a display is stacked with the touch sensor to form a touch display screen. The device detects touch operation triggered by a user based on the touch display screen.

Optionally, the electronic device 1 may further include a Radio Frequency (RF) circuit, a sensor, an audio circuit, and the like, which are not described herein again.

The embodiment of the application can acquire and process related data based on an artificial intelligence technology. Among them, Artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.

The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.

The application is operational with numerous general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like. The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

In the apparatus embodiment shown in fig. 1, an operating system, and an identity authentication program 10 may be included in a memory 11, which is a machine-readable medium; when the processor 12 executes the identity authentication program 10 stored in the memory 11, the steps of implementing the identity authentication method based on brain print recognition as shown in fig. 2 are as follows:

s20, obtaining identity associated information for identifying the identity of a user, and performing first identity authentication on the user based on the identity associated information;

s21, if the first identity authentication is passed, displaying a plurality of candidate pictures through the display module; the candidate pictures comprise a plurality of interference pictures which are randomly generated and at least one pre-stored authentication picture provided by a user;

s22, acquiring an electroencephalogram signal generated by the user according to an authentication picture through an electroencephalogram signal acquisition module, wherein the authentication picture is determined by the user from a plurality of candidate pictures;

s23, recognizing the electroencephalogram signal by using a pre-trained kernel function-based support vector machine model for recognizing the identity of the user to obtain a first recognition result; the first recognition result represents a probability value that the user belongs to a certain identity;

s24, recognizing the electroencephalogram signal by using a pre-trained brain print recognition model for recognizing the identity of the user to obtain a second recognition result; the second recognition result represents a probability value that the user belongs to a certain identity;

s25, carrying out weighted summation on the first recognition result and the second recognition result to obtain an identity authentication comprehensive probability value;

s26, comparing the comprehensive probability value of the identity authentication with a preset threshold value to finish the second identity authentication of the user; and if the comprehensive probability value of the identity authentication is larger than the preset threshold value, passing the identity authentication for the second time.

Through the mode, the method can realize multiple times of authentication of the user, and comprehensively authenticates the user by combining the support vector machine and the brain print recognition model based on the neural network, so that the misjudgment phenomenon of a single recognition mode is avoided as much as possible. Compared with the traditional biological authentication method, the method improves the problems that the traditional virtual password is easy to leak and attack, the entity password medium is easy to lose, the privacy is easy to leak and the like, and improves the security of authentication.

The steps are explained in detail below.

In step S20, identity related information for identifying the identity of the user is acquired, and the user is authenticated for the first time based on the identity related information.

In this embodiment, a large amount of user identity associated information is stored in the database in advance, the identity associated information of each user corresponds to the user identity information, and the user identity information can be determined through the user identity associated information. When user authentication is required, acquiring identity associated information of a user, comparing the identity associated information of the user with identity associated information of the user stored in a database in advance, and calculating the similarity between the identity associated information of the user and the identity associated information of the user in the database; and under the condition that the similarity between the identity associated information of the user and the identity associated information of the user in the database is greater than or equal to a preset similarity threshold value, the identity associated information of the user is considered to exist in the database.

The preset similarity threshold may be 98%, that is, when the similarity between the identity related information of the user and the identity related information of the user in the database exceeds 98%, it is determined that the identity related information of the user exists in the database. In this embodiment, a plurality of preset similarity thresholds may be set, for example, 85%, 90%, and 95%, and different similarity thresholds may be selected according to needs.

In this embodiment, the identity association information includes at least one of the following: face feature information, human body feature information and feature information with identity identification marks. The face feature information includes face features such as eyes, nose, mouth, and the like. And the face characteristic information is captured by a face capturing machine. The human body characteristic information comprises fingerprint characteristic information, voiceprint characteristic information, iris characteristic information, palm print characteristic information, vein characteristic information and the like. The characteristic information with the identity identification mark, namely the pre-stored characteristic information which can determine the user identity information through the identity identification information, can be a string of character strings, such as an identity card number or other numbers, and the number has uniqueness, so that the corresponding identity information can be obtained through the number; the character string may be pre-stored in an identification card, such as an identification card or RFID card. Identity associated information is perceived through intelligent perception equipment deployed at an entrance and an exit of a management area, and the intelligent perception equipment comprises: face identification terminal, AI camera, intelligent snapshot machine, card reading equipment etc..

In step S21, if the first authentication is passed, displaying a plurality of candidate pictures through the display module; the candidate pictures comprise a plurality of interference pictures which are randomly generated and at least one pre-stored authentication picture which is provided by a user.

After the identity information of the user is identified, a plurality of pictures can be displayed through the display module to serve as candidate pictures. The candidate pictures comprise at least one authentication picture reserved by the user and a plurality of randomly generated interference pictures, and the interference pictures are other pictures except the authentication pictures. The authentication picture needs to be reserved in a database when the user performs identity authentication for the first time. When the user identity authentication is performed, the user needs to determine a reserved authentication picture from a plurality of pictures. For example, a plurality of pictures are displayed on a display screen of the display device, and the user determines the authentication picture by clicking the picture displayed on the display screen, or each picture has a serial number, and the user determines the authentication picture by inputting the serial number.

If the user is authenticated according to the identity associated information of the user, if the identity of the user is not identified, the user can be considered as a new user, and at the moment, the user needs to upload a picture as an authentication picture and store the picture in the database. Meanwhile, the associated characteristic information of the new user and the identity information corresponding to the associated characteristic information are correspondingly stored in the database. The authentication picture may be a picture uploaded from a local storage device or a picture uploaded by taking a picture in real time.

In step S22, the electroencephalogram signal generated by the user according to the authentication picture determined by the user from the plurality of candidate pictures is acquired by the electroencephalogram signal acquisition module.

The electroencephalogram of the user can be obtained by pushing Visual stimuli (authentication pictures) to the user, the electroencephalogram of the user refers to VEP (Visual Evoked Potential) signal data, the VEP signal is an electroencephalogram induced during a specific Visual stimulus period (for example, picture stimulus), and the electroencephalograms induced when each person sees a specific picture are different. Thus, this characteristic of the VEP signal can be used for biometric identification to verify the identity of the person being stimulated.

When the electroencephalogram signals of the user are collected, the user needs to wear the electroencephalogram signal collecting module, and the electroencephalogram signals are collected through the electroencephalogram signal collecting module while watching the authentication picture. The EEG signal acquisition module comprises a lead EEG acquisition cap, a wireless EEG/ERP amplifier and a notebook computer. Wherein the lead brain electricity collecting cap is made of waterproof fabric, the appearance of the lead brain electricity collecting cap is similar to a hair band, two ends of the lead brain electricity collecting cap are respectively provided with a magic tape, and the lead brain electricity collecting cap is in a circular ring structure after being pasted. The middle part of the inner side of the lead electroencephalogram acquisition cap is provided with 8 electrode plates, the electrode types are dry electrodes, the position distribution of the dry electrodes conforms to the international 10-20 system standard (the positions of OZ, O1, O2, POZ, PO3, PO4, PO5 and PO6 respectively), and the dry electrodes are responsible for acquiring electroencephalogram signals of the occipital lobe areas of the brain. The wireless EEG/ERP amplifier is connected to the middle part of the outer side of the lead EEG acquisition cap in a magnetic attraction mode. The wireless EEG/ERP amplifier is responsible for collecting the EEG signals collected by the lead EEG collection cap, and the EEG signals are wirelessly transmitted to the notebook computer for EEG signal storage after being processed by signal amplification, analog-to-digital conversion and the like.

The electroencephalogram signal acquisition module can be selected from a device which is good in portability, low in cost, mature and capable of achieving the purpose of acquiring the electroencephalogram of the position of the occipital lobe of the brain, and the device is not described in detail here.

In an embodiment, in order to improve the calculation accuracy, the electroencephalogram signal needs to be preprocessed before calculating the power spectral density of the electroencephalogram signal or extracting the electroencephalogram feature corresponding to the electroencephalogram signal. As shown in fig. 3, the preprocessing of the electroencephalogram signal includes:

s31, denoising the electroencephalogram signal by using a recess filter to obtain a first electroencephalogram signal;

because the acquired electroencephalogram signals are mixed with power frequency noise generated when electronic and electric equipment works, the electroencephalogram signals need to be subjected to noise reduction treatment, and specifically, a recess filter can be adopted to perform noise reduction treatment on the electroencephalogram signals.

S32, filtering the first electroencephalogram signal by using a low-pass filter to obtain a second electroencephalogram signal;

because the effective frequency range of brain scalp electroencephalogram is 0-50Hz, a low-pass filter algorithm is adopted to extract the effective wave range and filter the ineffective wave range.

And S33, denoising the second electroencephalogram signal through an independent component analysis algorithm to obtain a third electroencephalogram signal.

Because physiological artifacts such as blink, eye movement, head movement, heart rhythm and the like are inevitably mixed in the acquired electroencephalogram signals, interference signals cannot be completely eliminated after the acquired electroencephalogram signals are processed by the filter, so that different source signals can be divided and removed by using an independent component analysis algorithm to obtain cleaner electroencephalogram signals.

An Independent Component Analysis (ICA) method is one of blind source separation technologies, when the method is used for denoising multi-electrode electroencephalogram signals, each electroencephalogram acquisition electrode is used as a signal source, each independent source is separated from acquired composite signals, then original electroencephalogram signals are screened from the independent sources, noise components are removed, and the purpose of denoising is achieved.

And when the electroencephalogram signals are identified by utilizing the support vector machine model and the brain print identification model in the follow-up process, the identification is finished based on the third electroencephalogram signal obtained through preprocessing.

In step S23, recognizing the electroencephalogram signal by using a pre-trained kernel function-based support vector machine model for recognizing the user identity, to obtain a first recognition result; the first recognition result represents a probability value that the user belongs to a certain identity.

As shown in fig. 4, the recognizing the electroencephalogram signal by using a pre-trained kernel function-based support vector machine model for recognizing the identity of the user to obtain a first recognition result includes:

s41, acquiring a power spectral density feature vector of the electroencephalogram signal;

s42, inputting the power spectral density feature vector to the support vector machine model to obtain a first identification result.

A Support Vector Machine (SVM) is a supervised learning method for classifying data. By using the method, the low-dimensional indivisible data is mapped to a proper high-dimensional feature space through a certain nonlinear function, and more accurate classification and discrimination are realized. In the training stage of the support vector machine model, data can be acquired through experiments, a source data set is combined for training and testing, and the model is stored after the acceptable accuracy is achieved.

As shown in fig. 5, in an embodiment, the acquiring the power spectral density feature vector of the electroencephalogram signal includes:

s51, segmenting the electroencephalogram signals according to the human brain frequency distribution to obtain electroencephalogram signals of multiple wave bands;

s52, acquiring the power spectral density of the electroencephalogram signals of the multiple wave bands;

and performing time-frequency conversion on the preprocessed electroencephalogram signals by using short-time Fourier transform or wavelet transform, and converting time domain signals into frequency domains. The human brain frequency can be roughly divided into 5 bands: the combination of delta band (0-3Hz), theta band (4-7Hz), alpha band (8-13Hz), beta band (14-30Hz), gamma band (31-50Hz), and 5 band power spectral densities reflects the differences between different individuals in the same or different states.

Specifically, the power spectral densities of the electroencephalogram signals of the multiple bands can be calculated by a welch algorithm.

Firstly, the received EEG signal x_N(N) is divided into L segments, each segment being M in length, i.e. N equals LM for x_N(n) when segmenting, allowing data of each segment to have data overlap. For example, if each piece of data overlaps by half, the number of pieces at this time is:

where M is the length of each segment of the signal.

For each segmentWindowing the data, denoted d₂(n) of (a). Separately calculating the power spectrum of each segment

P (W) is the power spectral density of the EEG signal of a certain power, W represents the frequency, xⁱ(n) is the electroencephalogram signal of the ith wave band, j is an imaginary number unit, and e is a natural logarithm.

S53, the power spectral densities corresponding to the electroencephalogram signals of each wave band are combined into a feature vector.

And (3) forming corresponding feature vectors by the power spectral densities of the 5 wave bands, and then inputting the corresponding feature vectors into a pre-trained support vector machine model to complete the identification of the electroencephalogram signals to obtain a first identification result. And outputting a first identification result after the classification and the discrimination of the support vector machine model, wherein the first identification result is a probability value, namely the probability value of the user belonging to a certain identity.

In step S24, recognizing the electroencephalogram signal by using a pre-trained brain print recognition model for recognizing the identity of the user to obtain a second recognition result; the second recognition result represents a probability value that the user belongs to a certain identity;

specifically, a pre-trained brain print recognition model is used for carrying out feature extraction on the electroencephalogram signals to obtain brain print features; and identifying the electroencephalogram signal according to the brain print characteristics to obtain a second identification result. The electroencephalogram signals are distinguished by the brain print recognition model and then output recognition results, and the recognition results are probability values.

The brain print recognition model may be an end-to-end convolutional neural network, and the output result is a probability value, i.e. a probability value of the user belonging to a certain identity. Data can be acquired through experiments at the present stage of the brain print recognition model training, the data sets are combined and separated for training and testing, and the model is stored after the acceptable accuracy is achieved.

In this embodiment, the brain print features in the electroencephalogram signal can be extracted according to the pre-trained brain print recognition model, and the brain print features in the electroencephalogram signal include frequency features and spatial features of each frequency feature. The spatial features include a plurality of different spatial features. When extracting the brain print features of the user, a plurality of different spatial features need to be mixed, and the brain print features of the user are determined according to the mixed features, wherein the brain print features are the mixed spatial features, and the mixing mode can be convolution operation.

When the identity of the user needs to be recognized, the user only needs to receive the same visual stimulation again, receive the electroencephalogram signal of the user, extract the brain print features in the electroencephalogram signal, and recognize the brain print features of the user through a pre-trained brain print recognition model, so that the identity of the user is recognized.

In one embodiment, the brain print recognition model comprises a temporal convolution module, a spatial convolution module, a separable convolution module, and a classification module;

the time convolution module is used for extracting the frequency characteristics of the electroencephalogram signals;

the spatial convolution module is used for extracting spatial features of the electroencephalogram signals;

the separable convolution module is used for fusing the frequency characteristic and the space characteristic to obtain a fused characteristic;

and the classification module is used for identifying the electroencephalogram signals based on the fusion characteristics.

When the brain print recognition model is used for identity recognition, the electroencephalogram signal is subjected to feature extraction through the brain print recognition model, and the method specifically comprises the following steps:

after the extraction of the brain print features is completed, the identification of the electroencephalogram signals according to the brain print features comprises the following steps:

Table 1 shows the specific composition of each module, as shown in table 1:

TABLE 1

The time convolution module comprises an input layer, a convolution layer and a batch normalization layer; the time convolution module is used for fusing information on time, extracting various time dimension information and obtaining frequency characteristics.

The spatial convolution module comprises a depth convolution layer, a batch normalization layer, an activation layer, an average pooling layer and an output layer; the space convolution module is used for fusing information of each space (different channels of electroencephalogram) to obtain space characteristics.

The separable convolution module comprises a separated convolution layer, a batch normalization layer, an activation layer, an average pooling layer and an output layer.

In the time convolution module, a feature map is output using 2D convolution (Conv2D) and a filter with parameters F1, followed by batch normalization.

In the spatial convolution module, deep convolution (depthwiseConv2D) is used to learn the spatial filter, then batch normalization is performed, and finally, an average pooling layer is used to reduce the number of features.

In this embodiment, a spatial filter of a specific frequency can be extracted efficiently by combining the time convolution (Conv2D) and the spatial convolution (depthwiseConv 2D). And D is responsible for controlling the number of spatial filters to be learned by each feature map. Dropout techniques can be used in the spatial convolution module for overfitting and modeling.

In the Separable Convolution module, Separable Convolution (Separable Convolution) is used, and Separable Convolution is separated into a depth Convolution including a depth Convolution layer and a point Convolution with a parameter of F2. There are two advantages to using a depth separable convolution: 1) the number of parameters to be fitted is reduced. 2) And learning the feature kernel to separate the relation with the feature maps and summarize each feature map through the optimal combined output. The separable convolution module can distinguish how learning summarizes in time the individual feature mappings (transposed convolution) from optimally combining feature mappings (point convolution). Finally, the average pooling layer is also used to reduce the size.

In the classification module, the features are passed directly to the softmax output to reduce the number of free parameters in the model.

In step S25, the first recognition result and the second recognition result are weighted and summed to obtain an authentication comprehensive probability value.

After the identity authentication probability values output by the support vector machine model and the brain print recognition model are obtained, the identity authentication comprehensive probability value is obtained according to the two probability values, and then the identity of the user is authenticated according to the identity authentication comprehensive probability value.

In one embodiment, the authentication comprehensive probability value is calculated by weighted summation of the first recognition result and the second recognition result,

ComPr＝SVMPr×Coef_svm+EegnetPr×Coef_eegnet

wherein, ComPr is the comprehensive probability value of identity authentication; SVMPr is a probability value of the first recognition result; coef_svmThe weight value of the first recognition result; EegnetPr is a probability value of the second recognition result; coef_eegnetThe weight value of the second recognition result;

in an embodiment, the second identity authentication result is obtained through the identity authentication comprehensive probability value and a preset threshold.

Specifically, the comprehensive probability value of the identity authentication is compared with a preset threshold value to complete the second identity authentication of the user; and if the comprehensive probability value of the identity authentication is larger than the preset threshold value, passing the identity authentication for the second time.

And judging whether the comprehensive probability value of the identity authentication can meet a preset threshold value according to the following formula.

Wherein Threshold is a preset Threshold; IdResult is the final output result.

If the IdResult is true, outputting an identification result, and passing the user identity authentication; otherwise, abandoning the judgment result, and collecting the electroencephalogram signal again to analyze and judge the flow. After the identification fails for a plurality of times (for example, 3 times), the user identity authentication is not passed.

By the mode, multiple times of authentication of the user can be realized, and the user is comprehensively authenticated by combining the support vector machine and the brain print recognition model based on the neural network, so that the phenomenon of misjudgment is avoided as much as possible. Compared with the traditional biological authentication method, the method improves the problems that the traditional virtual password is easy to leak and attack, the entity password medium is easy to lose, the privacy is easy to leak and the like, and improves the security of authentication.

As shown in fig. 6, an embodiment of the present application further provides an identity authentication system based on brain print recognition, including:

a first authentication module 60, configured to obtain identity association information for identifying an identity of a user, and perform first identity authentication on the user based on the identity association information;

the picture display module 61 is used for displaying a plurality of candidate pictures when the first identity authentication is passed; the candidate pictures comprise a plurality of interference pictures which are randomly generated and at least one pre-stored authentication picture provided by a user;

an electroencephalogram signal acquisition module 62, configured to acquire an electroencephalogram signal generated by the user according to an authentication picture, where the authentication picture is determined by the user from multiple candidate pictures;

the first identification module 63 is configured to identify the electroencephalogram signal by using a pre-trained kernel function-based support vector machine model for identifying a user identity, so as to obtain a first identification result; the first recognition result represents a probability value that the user belongs to a certain identity;

the second recognition module 64 is configured to recognize the electroencephalogram signal by using a pre-trained brain print recognition model for recognizing the identity of the user, so as to obtain a second recognition result; the second recognition result represents a probability value that the user belongs to a certain identity;

the comprehensive authentication module 65 is configured to perform weighted summation on the first identification result and the second identification result to obtain an identity authentication comprehensive probability value;

the second authentication module 66 compares the comprehensive probability value of the identity authentication with a preset threshold value to complete the second identity authentication of the user; and if the comprehensive probability value of the identity authentication is larger than the preset threshold value, passing the identity authentication for the second time.

Through the mode, the method and the device can realize multiple times of authentication of the user, and comprehensively authenticate the user by combining the support vector machine and the brain print recognition model based on the neural network, so that the phenomenon of misjudgment is avoided as much as possible. Compared with the traditional biological authentication method, the method improves the problems that the traditional virtual password is easy to leak and attack, the entity password medium is easy to lose, the privacy is easy to leak and the like, and improves the security of authentication.

After the identity information of the user is identified, a plurality of pictures are randomly output as candidate pictures through the picture display module. The multiple candidate pictures comprise authentication pictures and interference pictures reserved by the user, and the interference pictures are other pictures except the authentication pictures. The authentication picture needs to be reserved in a database when the user is authenticated for the first time. When user authentication is performed, a user needs to determine a reserved authentication picture from multiple pictures. For example, a display device displays a plurality of pictures on a display screen of the display device, and a user determines an authentication picture by clicking the pictures displayed on the display screen, or each picture has a serial number, and the user determines the authentication picture by inputting the serial number.

In an embodiment, the identity authentication system further includes a preprocessing module, configured to preprocess the electroencephalogram signal, where the preprocessing module includes: the device comprises a first filtering submodule, a second filtering submodule and a denoising submodule;

the first filtering submodule is used for carrying out denoising processing on the electroencephalogram signal by using a recess filter to obtain a first electroencephalogram signal;

The second filtering submodule is used for filtering the first electroencephalogram signal by using a low-pass filter to obtain a second electroencephalogram signal;

And the denoising submodule is used for denoising the second electroencephalogram signal through an independent component analysis algorithm to obtain a third electroencephalogram signal.

In one embodiment, the first identification module comprises: the feature vector acquisition submodule and the identification submodule;

the characteristic vector acquisition submodule is used for acquiring a power spectral density characteristic vector of the electroencephalogram signal;

and the identification submodule is used for inputting the power spectral density feature vector into the support vector machine model to obtain a first identification result.

In one embodiment, the feature vector obtaining sub-module includes: the device comprises a segmentation unit, a power spectral density acquisition unit and a combination unit;

the segmentation unit is used for segmenting the electroencephalogram signals according to the frequency distribution of the human brain to obtain electroencephalogram signals of a plurality of wave bands;

a power spectral density acquisition unit, configured to acquire power spectral densities of the electroencephalogram signals of the multiple bands;

where M is the length of each segment of the signal.

Windowing each segment of data, denoted d₂(n) of (a). Separately calculating the power spectrum of each segment

And the combination unit is used for combining the power spectral densities corresponding to the electroencephalogram signals of each wave band into a feature vector.

In an embodiment, the second identification module performs feature extraction on the electroencephalogram signal through a brain print identification model to obtain brain print features, and obtains a second identification result based on the brain print features.

Specifically, the preprocessed electroencephalogram signal is directly input into a brain-print recognition model to realize the recognition of the electroencephalogram signal, the electroencephalogram signal is distinguished by the brain-print recognition model and then a recognition result is output, and the recognition result is a probability value.

Table 2 shows the specific composition of each module, as shown in table 2:

TABLE 2

In one embodiment, the comprehensive authentication probability value is calculated by the comprehensive authentication module through the weighted summation of the first recognition result and the second recognition result,

ComPr＝SVMPr×Coef_svm+EegnetPr×Coef_eegnet

and for the second authentication module, obtaining a second identity authentication result through the identity authentication comprehensive probability value and a preset threshold value.

Specifically, the second authentication module compares the comprehensive probability value of the identity authentication with a preset threshold value to complete the second identity authentication of the user; and if the comprehensive probability value of the identity authentication is larger than the preset threshold value, passing the identity authentication for the second time.

Wherein Threshold is a preset Threshold; IdResult is the final output result.

In an embodiment of the present application, there is also provided a computer readable storage medium having stored thereon a computer program, which, when executed by a processor, causes the processor to perform the steps of the identity authentication method as shown in fig. 2.

An embodiment of the present application further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to perform the steps of the identity authentication method shown in fig. 2.

The above steps are substantially the same as the specific implementation of the identity authentication method and system based on brain print recognition, and are not described herein again.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.

It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

In the embodiments provided in the present invention, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other ways. For example, the above-described embodiments of the apparatus/terminal device are merely illustrative, and for example, the division of the modules or units is only one logical division, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments. Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) as described above and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.

It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a machine-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.

The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims

1. An identity authentication method based on brain print recognition is characterized by comprising the following steps:

2. The identity authentication method based on the brain print recognition according to claim 1, further comprising: preprocessing the electroencephalogram signals, comprising:

3. The identity authentication method based on the brain print recognition of claim 1, wherein the recognizing the electroencephalogram signal by using a pre-trained kernel function-based support vector machine model for recognizing the identity of the user to obtain a first recognition result comprises:

4. The identity authentication method based on the brain print recognition of claim 3, wherein the obtaining the power spectral density feature vector of the electroencephalogram signal comprises:

5. The identity authentication method based on the brain print recognition of claim 4, wherein the obtaining the power spectral density of the electroencephalogram signals of the plurality of bands comprises:

6. The identity authentication method based on the brain print recognition of claim 1, wherein the recognizing the electroencephalogram signal by using the pre-trained brain print recognition model for recognizing the identity of the user to obtain the second recognition result comprises:

7. The identity authentication method based on the brain print recognition of claim 6, wherein the feature extraction of the electroencephalogram signal comprises:

8. An identity authentication system based on brain print recognition, comprising:

9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the identity authentication method according to any one of claims 1 to 7 are implemented by the processor when executing the computer program.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the identity authentication method of any one of claims 1 to 7.