WO2018112799A1

WO2018112799A1 - Visual evoked potential-based identity verification method and device

Info

Publication number: WO2018112799A1
Application number: PCT/CN2016/111324
Authority: WO
Inventors: 袁鹏; 薛希俊; 姚骏
Original assignee: 华为技术有限公司
Priority date: 2016-12-21
Filing date: 2016-12-21
Publication date: 2018-06-28
Also published as: CN110121711A

Abstract

A visual evoked potential-based identity verification method, comprising: generating an M-sequence stimulus signal corresponding to a user (201); acquiring an electroencephalogram signal generated by the user in response to the M-sequence stimulus signal (202); calculating a correlation coefficient between the acquired electroencephalogram signal and an authentication electroencephalogram signal of the user (203); and determining whether the correlation coefficient is greater than a threshold value of a matching coefficient (204). If the correlation coefficient is greater than the threshold value of the matching coefficient, then the identity authentication is successful. If not, then the identity authentication has failed. The M-sequence stimulus signal is used to cause the user to generate a specific electroencephalogram signal pattern, and the correlation coefficient between the acquired electroencephalogram signal and the authentication electroencephalogram signal of the user is used to determine an identity verification result. The characteristic of the M-sequence stimulus signal allows easy detection of an electroencephalogram signal. Low correlation between electroencephalogram signals from different subjects and stable characteristics and easy collection of an electroencephalogram signal resolve the issues of electroencephalogram pattern-based identity verification methods in the prior art.

Description

Method and device for identifying identity based on visual evoked potential

Technical field

Embodiments of the present invention relate to the field of information security, and in particular, to an image recognition method and apparatus based on visual evoked potentials.

Background technique

In the era of universal interconnection, the number of connected smart devices has grown exponentially, and information security issues have become particularly important. Research on various identification methods has also become a research hotspot. People began to use biometrics for identity recognition. Biometrics technology generally refers to the use of some physiological characteristics or behavioral characteristics inherent in the human body for identification. The physiological characteristics of the human body generally include: human face, fingerprint, palm shape, iris, etc.; human behavior characteristics may include: handwriting, gait, and the like. However, the existing biometrics technology also faces some problems. The fake finger made of gelatin can successfully fool the fingerprint identification system. The false iris feature etched on the contact lens can make the iris recognition system indistinguishable. true and false.

In recent years, people have begun to consider the use of Electroencephalogram (EEG) as a new type of biometrics in identity recognition. Studies have shown that even under the same external stimulus or when people are thinking about the same problem, the brain electrical signals induced by the brains of different subjects are different, that is, EEG has significant individual differences. At the same time, EEG has many advantages such as difficulty in copying and forgery, and can be modulated by the subject's own attention. At present, there are some identification methods based on various modes of EEG, such as identification methods based on resting EEG, identification methods based on EEG features of imaginary motion states, EEG identification methods based on P300 event-related potentials, etc. . However, there are some unsatisfactory places, such as: the EEG signal used is relatively low signal-to-noise ratio, difficult to detect; the signal characteristics are not stable enough, and are easily affected by the state of the user; it is often necessary to collect multiple lead EEG signals. Or a large number of training samples, the use is not convenient enough.

Summary of the invention

In view of this, an embodiment of the present invention provides a method for identifying an identity based on a visual evoked potential and The device uses the M sequence to induce the unique characteristics of the brain electricity, so that the identification signal is easier to detect, the acquisition is convenient, and the influence of the human mental state is less, and the signal characteristics are more stable.

In a first aspect, the present application provides a method for identifying an evoked potential based on a visual evoked potential, comprising: generating a user-supplied M-sequence stimulation signal to stimulate a user; and collecting an EEG signal generated by the user for the M-sequence stimulation signal; Correlation coefficient between the collected EEG signal and the authenticated EEG signal of the user, the authenticated EEG signal of the user is an EEG signal generated by the user for the corresponding M-sequence stimulus signal for identification, and the correlation coefficient is used for The degree of similarity between the two signals is measured; if the correlation coefficient is greater than the matching coefficient threshold, the identity authentication is passed; otherwise, the identity authentication fails.

It can be understood that the M-sequence stimulation signal herein generally refers to a visual signal that changes in the form of an M-sequence, and induces a user to generate an EEG signal by stimulating the user for a certain period of time. By adopting the above method, since the characteristics of the M-sequence stimulation signal are utilized, the EEG signal is easy to detect, the recognition accuracy is high, the EEG signal characteristics are stable and easy to collect, and the existing EEG-based identity recognition methods are solved. The problem.

Further, the M-sequence stimulation signal corresponding to the user is obtained by performing T-time shift on the original M-sequence stimulation signal, wherein the length of the original M-sequence stimulation signal is N, 0≤T≤N-1.

In one possible implementation, Ki is the number of shifts of the shift of the T-times shifting _I times T, 0≤K _i ≤N-1, the M-sequence corresponding to the user a stimulus signal to the user The stimulation includes: for the original M-sequence stimulation signal, sequentially shifting K _{i to} generate the T- _th M-sequence stimulation signal, and using the T- _th M-sequence stimulation signal to stimulate the user; calculating the collected EEG signal Before the correlation coefficient with the user's certified EEG signal, the method further comprises: intercepting the acquired EEG signal according to the stimulation sequence of the M-sequence stimulation signal of the T-time shift, and obtaining the M-sequence stimulation signal corresponding to each shift EEG signal. An M-sequence stimulation signal that produces a shift for the original M-sequence stimulation signal and an EEG signal corresponding to the M-sequence stimulation signal obtained each time are achieved.

In a possible implementation, the method further includes: generating a synchronization signal when starting to stimulate the user with the M- _th order shifted M-sequence stimulation signal; and then shifting the collected EEG signal according to T times The M-sequence stimulation signal stimulation sequence is intercepted, and the EEG signal corresponding to the M-sequence stimulation signal of each shift is obtained, including: the stimulation sequence of the acquired EEG signal according to the T-shifted M-sequence stimulation signal, according to The synchronization signal is intercepted to obtain an EEG signal corresponding to the M-sequence stimulation signal of each shift. By using the synchronization signal, the accuracy of the EEG signal corresponding to the M-sequence stimulation signal of each shift is intercepted from the acquired EEG signal.

In a possible implementation, calculating a correlation coefficient between the collected EEG signal and the user's certified EEG signal includes: separately calculating the EEG signal corresponding to each acquired M-sequence stimulation signal and the user's The correlation coefficient of the EEG signal is verified, and the correlation coefficient calculated each time is summed according to the corresponding weight, and the correlation coefficient between the collected EEG signal and the user's certified EEG signal is obtained.

Further, the method further comprises: adjusting the influence of the correlation coefficient between the EEG signal corresponding to the M-sequence stimulus signal of each shift and the user's certified EEG signal on the user's identification accuracy, and adjusting the M of each shift The weight of the correlation coefficient corresponding to the EEG signal corresponding to the sequence stimulation signal.

According to the identity authentication situation, the weight of the correlation coefficient corresponding to the EEG signal corresponding to each shifted M-sequence stimulus signal is continuously updated, so that the final acquired EEG signal and the user's obtained are obtained during subsequent identity authentication. The correlation coefficient of the certified EEG signal is closer to more accurate, and the identity authentication result is more credible.

It can be understood that the authenticated EEG signal of the user here is also an EEG signal generated by the user for its corresponding M-sequence stimulation signal, which is pre-acquired and stored in the system for user identification.

Further, before generating the user-supplied M-sequence stimulation signal to stimulate the user, the method further includes: receiving T digits sequentially input by the user, determining the input T digits and the input sequence and the shift number of the T-time shift Same as the shift order. By using the sequence of the number of shifts of the M-sequence stimulus signal corresponding to the user as the password to be input by the user using the identification system, when the sequence of the number of shifts input by the user is correct, the subsequent identification process is entered; otherwise, the system can be adopted. The way of making and alerting increases the security and accuracy of identity.

In a second aspect, an embodiment of the present invention provides a visual evoked potential-based identification device, including: a stimulus generating unit for generating a user-supplied M-sequence stimulus signal for stimulating a user; and a signal collecting unit for collecting a user for the The EEG signal generated by the M sequence stimulation signal; the signal analysis unit is configured to calculate a correlation coefficient between the collected EEG signal and the user's certified EEG signal, and determine whether the correlation coefficient is greater than a matching coefficient threshold, and if the correlation coefficient is greater than the matching coefficient threshold, The identity authentication is passed, otherwise, the identity authentication does not pass. The user's certified EEG signal is an EEG signal generated by the user for the corresponding M-sequence stimulus signal for identification, and the correlation coefficient is used to measure the degree of similarity between the two signals.

In one possible implementation, the stimulus generating unit is an electronic device that produces a stimulus signal that changes in the form of an M sequence and visualizes the display to the user.

In a possible implementation, the signal acquisition unit includes a wearable brain electrical collection device including one or more dry electrodes for the user to wear.

In a possible implementation manner, the signal analysis unit includes a calculation subunit for calculating a correlation coefficient between the collected EEG signal and the user's certified EEG signal; the authentication subunit is configured to determine whether the correlation coefficient is greater than a matching coefficient threshold. If the correlation coefficient is greater than the matching coefficient threshold, the identity authentication is passed; otherwise, the identity authentication fails.

In a possible implementation manner, K _i is a shift of the T _ith shift in the T shift, and 0≤K _i ≤N-1, the stimulus generating unit is specifically used for the original M sequence stimulation signal , sequentially shifting K _{i to} generate the T- _th M-sequence stimulation signal, and using the T- _th M-sequence stimulation signal to stimulate the user; the signal acquisition unit includes: a collection sub-unit for collecting the user's stimulation for the M-sequence The EEG signal generated by the signal; the intercepting subunit is configured to intercept the acquired EEG signal according to the stimulation sequence of the M-sequence stimulation signal of the T-shift, and obtain the EEG corresponding to the M-sequence stimulation signal of each shift signal.

In a possible implementation manner, the stimulation generating unit is further configured to generate a synchronization signal when starting to stimulate the user with the M-th order stimulation signal of the Ti-th shift; the intercepting sub-unit is specifically used for the collected brain The electrical signal is intercepted according to the synchronization sequence of the M-sequence stimulation signal of the T-shift, and the EEG signal corresponding to the M-sequence stimulation signal of each shift is obtained.

In a possible implementation manner, the calculating subunit is specifically configured to: separately calculate a correlation coefficient between the collected EEG signal corresponding to each shifted M sequence stimulation signal and the user's certified EEG signal; The calculated correlation coefficients are summed according to the corresponding weights, and the correlation coefficient between the collected EEG signals and the user's certified EEG signals is obtained.

Further, the signal analysis unit further includes a weight management sub-unit, configured to influence the accuracy of the user's identification based on the correlation coefficient between the EEG signal corresponding to the M-sequence stimulus signal and the authenticated EEG signal of the user. And adjusting the weight of the correlation coefficient corresponding to the EEG signal corresponding to the M-sequence stimulation signal of each shift.

Further, the identity identification device further includes a key authentication unit, configured to receive T digits sequentially input by the user, and determine the input T digits and the input sequence, before generating the user-supplied M-sequence stimulation signal to stimulate the user. The number of shifts of the T-shift is the same as the shift order.

In another possible implementation, the signal analysis unit includes a processor and a memory, the memory is used to store the computer to execute the instruction, and the processor executes the computer to execute the instruction for calculating the collected EEG signal and the user's certified EEG signal. The correlation coefficient determines whether the correlation coefficient is greater than a matching coefficient threshold. If the correlation coefficient is greater than the matching coefficient threshold, the identity authentication passes, otherwise, the identity authentication fails.

Further, the identification device further includes an input/output unit for receiving T digits sequentially input by the user before generating the M-sequence stimulation signal corresponding to the user, and the processor is further configured to determine the input T numbers. And the input order is the same as the shift number and shift order of the T-time shift. By using the sequence of the number of shifts of the M-sequence stimulus signal corresponding to the user as the password to be input by the user using the identification system, the input/output unit receives the password input by the user, and when the sequence of the shift number input by the user is correct, the user enters Subsequent identification process, otherwise, the system can be raised and alarmed to increase the security and accuracy of identity recognition.

Through the above solution, the visual evoked potential-based identification method and apparatus provided by the embodiments of the present invention use the M-sequence stimulation signal to induce a user to generate a specific mode of EEG signals, and calculate the collected EEG signals and the user's certified EEG signals. Correlation coefficient, and the identification result is determined by the correlation coefficient. Due to the use of the characteristics of the M-sequence stimulation signal, the signal recognition method based on the EEG signal is easy to detect, the EEG signal correlation between different objects is small, and the EEG signal characteristics are stable and easy to collect, thus solving various existing problems. The problem of the identification method based on the EEG mode.

DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the present invention, Those skilled in the art can also obtain other drawings based on these drawings without paying any creative work.

1 is a schematic structural diagram of an identity recognition system according to an embodiment of the present invention;

2 is a flow chart of an identification method based on visual evoked potential;

3 is a flow chart of another method based on visual evoked potential identification;

Figure 4 is a schematic diagram of displacement of the M series stimulation signal;

Figure 5 is a schematic diagram of interception of EEG signals induced by M-series stimulation signals;

6 is a schematic diagram of a calculation method for correlation coefficients between acquired EEG signals and certified EEG signals;

Figure 7 is a schematic diagram of a safe identification system;

8 is a schematic structural diagram of an identification device based on visual evoked potential;

9 is a schematic structural diagram of another visual evoked potential based identification device;

FIG. 10 is a schematic structural diagram of another visual evoked potential based identification device.

detailed description

The technical solutions in the embodiments of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention.

The principle of using EEG for identification is that when the human brain stimulates externally, it will produce weak bioelectricity, that is, EEG, and the brain signals generated by different subjects are different for the same external stimulus. EEG has significant individual differences. The forms that usually produce brain waves are: event-related potential (ERP) and visual evoked potential (VEP). Among them, ERP is caused by external events to induce human brain cognitive activity. The general signal-to-noise ratio is very low. A typical ERP signal such as P300 signal, P300 appears as a positive peak appearing at a delay of 300ms from the event occurrence time. Usually when a person observes a small probability of a novel event (for example, a continuous "beep" sound, a "drip" sound suddenly appears), which induces a P300 event-related potential in the human brain. VEP is generally induced by external visual stimuli.

FIG. 1 is a schematic structural diagram of a system to which a method according to an embodiment of the present invention is applied. The whole system mainly includes three parts: the stimulus generating unit, the signal collecting unit and the signal analyzing unit.

The stimulation generating unit is operative to generate a stimulation signal. Optionally, in the embodiment of the invention, a device (e.g., a variable brightness LED lamp) that produces a stimulation signal in the form of an M sequence is included. The M sequence is an abbreviation of the longest linear shift register sequence and is a pseudo random sequence, a pseudo noise code or a pseudo random code. It is easy to generate, has strong regularity, has good autocorrelation and good cross-correlation properties. Given an M-sequence, it is almost orthogonal to its new sequence generated by arbitrary displacement. Due to the M sequence Nature, using M-sequence external stimuli (generally visual stimuli) to induce EEG signals (which can be regarded as a type of VEP), has the following characteristics: the steep peak characteristics of the autocorrelation function can make EEG signals easier to detect; The amplitude frequency response and phase frequency response of the subject EEG can directly correspond to the system physiological characteristics of the human primary visual cortex, and when the two M sequences induce EEG signals without phase (or delay) difference, the correlation coefficient is very high. (such as two EEG signals induced by the same M sequence for the same subject), and when the two M sequences induce EEG signals with phase (or delay) differences from each other (such as those induced by the same M sequence for different subjects) The two EEG signals have a correlation coefficient close to zero, which makes the identification system based on this structure more distinguishable and more robust; the generated EEG signals are mainly concentrated in the human brain occipital region, only need Less electrodes (such as an electrode) can collect the corresponding signals, which is convenient to use; as a signal induced by the primary visual cortex, no need for advanced human It is known to participate in activities, so by the person's mental state less affected, more stable signal characteristics. Therefore, in the embodiment of the present invention, the M-sequence stimulation signal is used to induce the EEG for identification, and the stimulation generating unit generates the M-stimulus signal visible to the user.

The signal acquisition unit comprises an electrode placed in the scalp of the occipital region of the brain (corresponding to the primary visual cortex region), and is responsible for collecting the EEG signal generated by the user induced by the M-sequence signal. Alternatively, it may be one or more worn by a user. A wearable brain electrical collection device with dry electrodes. At the beginning of the stimulation, the user looks at the stimulation signal generated by the stimulation generating unit with the eye and maintains the attention, and generates electroencephalogram in the occipital region of the scalp of the brain. At this time, the EEG signal can be collected by the wearable EEG collecting device. The collected EEG signals can be transmitted (for example, preferentially by wireless transmission) to the signal analysis unit for subsequent analysis processing.

The signal analysis unit includes two parts: signal feature extraction and matching recognition. The signal feature extraction part transforms the collected EEG signals to obtain corresponding features that can be used for contrast recognition, for example, by means of synchronization, data interception, etc., to obtain an EEG signal for matching identification; and the matching recognition part will The processed features are matched with the stored user feature data for identity recognition. The stored user characteristic data is an EEG signal acquired by the user before being stimulated by the M-sequence signal. If the stimulus and the acquired signal match the stored signal, the identity authentication is successful.

Through the above technical solution, the visual evoked potential based identification method provided by the embodiment of the present invention uses the M-sequence stimulation signal to induce the user to generate a specific mode of the EEG signal, and the calculation is collected. The correlation coefficient between the EEG signal and the user's certified EEG signal, and the identity authentication result is determined by the correlation coefficient. The identification method utilizes the characteristics of the M-sequence stimulation signal, making the EEG signal easy to detect, the EEG signal correlation between different objects is small, the EEG signal characteristics are stable and easy to collect, and the various EEG-based modes are solved. The problem with the identification method.

In conjunction with the schematic diagram of the identity recognition system shown in FIG. 1, an embodiment of the present invention provides a method for identifying an identity based on a visual evoked potential. As shown in FIG. 2, the specific process includes:

Step 201: The identity recognition system generates a M-sequence stimulation signal corresponding to the user, and stimulates the user.

The stimulation generating unit in the activation system generates an M-sequence stimulation signal, where the M-sequence stimulation signal generally refers to a visual signal that changes in the form of an M sequence, such as an LED lamp whose brightness can be changed. It can be understood that it is necessary to stimulate the user for a certain time to induce the user to generate an EEG signal, but the time of the stimulation is not strictly limited. In theory, the longer the effect, the better, but for the M-sequence stimulation signal, there is generally 1 The EEG signal generated by the -2 second stimulus can meet the recognition requirements.

Step 202: Acquire an EEG signal generated by the user for the M-sequence stimulation signal.

The signal acquisition unit collects the EEG signals generated by the user. In signal acquisition, the user wears a wearable EEG acquisition device that contains one or more dry electrodes placed on the scalp of the user's cerebral occipital region (corresponding to the primary visual cortical region). When the stimulation starts, the user looks at the M-sequence stimulation signal generated by the stimulation generating unit with eyes and maintains attention. At this time, the EEG signal induced by the M-sequence of the user's brain can be collected by the wearable EEG acquisition device, and the collected EEG signals can be transmitted to the signal analysis unit for subsequent analysis and identification. Here, the transmission method can be wireless transmission (such as Bluetooth), cable connection transmission (such as direct connection of signal lines).

Step 203: Calculate a correlation coefficient between the collected EEG signal and the user's certified EEG signal.

Here, the authenticated EEG signal of the user refers to an EEG signal for identification generated by the user for its corresponding M-sequence stimulation signal, and is generally a pre-acquired EEG signal generated by the user for its corresponding M-sequence stimulation signal. , saved in the system for user identification. The correlation coefficient here is used to measure the degree of similarity between the two signals and can be used to measure the degree of similarity between the acquired EEG signals and the pre-existing user EEG signals. Alternatively, a Pearson linear correlation coefficient can be used.

Due to differences in the physiological structure of the brain (such as different visual pathway delays, the spectral response of the visual system is different), even between the stimulation of the same M-stimulus signal, different human-induced EEG signals There is also a big difference. When the two M-sequence stimulus-induced EEG signals have no phase (or delay) difference from each other, the correlation coefficient is very high (such as the two brains of the same user induced by the same M-sequence stimulus signal). Electrical signal), and when the two M-sequence stimulus-induced EEG signals have phase (or delay) differences from each other (such as two EEG signals of different users induced by the same M-sequence stimulus signal), the correlation coefficient will be Very low. It can be understood that the correlation coefficient between the EEG signal collected by the step 202 and the authenticated EEG signal of the user is calculated here for determining whether the collected EEG signal has high correlation with the user's authentication EEG signal.

Optionally, when the signal acquisition unit uses multiple dry electrodes for acquisition, the collected EEG signal generated by the user is a signal of multiple electrodes, and the mathematical expression is a matrix. At this time, classical data reduction methods such as Principal Component Analysis (PCA) or Canonical Correlation Analysis (CCA) can be used to reduce it to a one-dimensional time domain signal. For ease of expression and description, in the embodiments of the present invention, the generated EEG signals are regarded as one-dimensional time domain signal processing.

Step 204: Determine whether the obtained correlation coefficient is greater than a matching coefficient threshold. If the matching coefficient threshold is greater than the matching coefficient threshold, the identity authentication is passed. Otherwise, the identity authentication fails.

It can be understood that the matching coefficient threshold can be set according to actual system requirements, and optionally, according to the accuracy rate and/or the recall rate requirement, the accuracy rate is how many of the EEG signals passed by the identity authentication are the real users. The recall rate is how many times the identity authentication performed by the real user is correctly identified. These two values intuitively represent the specificity and sensitivity of system identification.

Optionally, steps 203 and 204 are implemented by the signal analysis unit shown in FIG. 1.

In the above method embodiment, the M-sequence stimulation signal is used to induce the user to generate a specific mode of the EEG signal, and the correlation coefficient between the collected EEG signal and the user's certified EEG signal is calculated, and the identity authentication result is determined by the correlation coefficient. The identification method utilizes the characteristics of the M-sequence stimulation signal, so that the EEG signal is easy to detect, the recognition accuracy is high, the EEG signal characteristics are stable and easy to collect, and the existing EEG-based identification methods are solved. problem.

As shown in FIG. 3, an embodiment of the present invention provides another method for identifying an identity based on a visual evoked potential. For the terms and implementation details not defined in this embodiment, reference may be made to the method embodiment of FIG. 2 above. The method comprises the following steps:

Step 301: Generate a M-sequence stimulation signal corresponding to the user, and stimulate the user. Used here The M-sequence stimulus signal corresponding to the user is an M-series stimulation signal obtained by shifting the original M-sequence stimulus signal corresponding to the user.

Determining a raw M-sequence stimulus signal for the user, such as an M-sequence stimulus signal of length N, and shifting it k as the M-sequence stimulus signal for the user, where the shift number k ranges from 0 ≤ k ≤ N-1. A plurality of shift numbers k can be selected for the original M-sequence stimulus signal, that is, a plurality of M-series stimulation signals are obtained for stimulating the user. Alternatively, stimulation for the original M-sequence signal is shifted T times, 0≤T≤N-1 Here, the first time T _I T times the number of shifts for shifting the shift K _{_i,} 0≤K _i ≤ N-1. That is, the number of shifts is a vector K, K = [k ₁ , k ₂ , ..., k _T ].

For the original M-sequence stimulation signal, K _{i is} sequentially shifted to generate the T- _th M-sequence stimulation signal, and the T- _th M-sequence stimulation signal is used to stimulate the user. For example, using the M-series stimulation signal shifted by T times, the number of shifts K _i in the T-shift is one of the sequences [k ₁ , k ₂ , . . . , k _T ], and the identification system is as follows. The timing shown in FIG. 4 is stimulated, and the stimulation generating unit first generates the M-sequence stimulation signal M ₁ shifted by k ₁ , stimulates the M-sequence stimulation signal M ₁ for a period of time, and then generates the M-sequence stimulation signal M ₂ shifted by k ₂ . Stimulation with M ₂ for a period of time, in sequence, until the T-M sequence stimulation signal is completely stimulated.

Step 302: Acquire an EEG signal generated by the user for the shifted M-sequence stimulation signal. The EEG signal generated by the user is collected by the signal acquisition unit, and the EEG signal here is induced by the shifted M series stimulation signal generated in Step 301. The method for collecting the EEG signal is basically the same as the step 202 in the previous embodiment, and details are not described herein again.

Step 303: Acquire an EEG signal corresponding to each shifted M-sequence stimulation signal. The collected EEG signals are intercepted according to the stimulation order of the M-sequence stimulation signals of the T-shifts, and the EEG signals corresponding to the M-sequence stimulation signals of each shift are obtained. As shown in FIG. 5, the acquired EEG signals are intercepted according to the order of stimulation (here, the sequence of sequences [k ₁ , k ₂ , . . . , k _T ]), and each shifted M-sequence stimulation signal k _{i is obtained.} Corresponding EEG signal X(k _i ).

Optionally, in step 301, when the user is stimulated by the shifted M-sequence stimulation signal, a synchronization signal is generated. In this step, after receiving the collected EEG signal, the signal acquisition unit performs the T-time shift. The stimulation sequence of the M-sequence stimulation signal of the bit is subjected to signal interception according to the synchronization signal to obtain an electroencephalogram signal X(k _i ) corresponding to each shifted M-sequence stimulation signal. Such as: X (k ₁ ), X (k ₂ ) and the like.

Step 304: Calculate a correlation coefficient between the collected EEG signal and the user's certified EEG signal. Calculate the correlation coefficient between the EEG signal corresponding to each shifted M-sequence stimulus signal and the user's certified EEG signal, and then calculate the correlation coefficient obtained by each calculation according to the corresponding weight. The correlation coefficient between the EEG signal and the user's certified EEG signal. Here, the user's certified EEG signal is also a pre-acquired EEG signal generated by the user for its corresponding M-sequence stimulus signal, stored in the system for user identification, and also for the brain generated by the shifted M-sequence stimulus signal. electric signal.

As shown in FIG. 6, the correlation coefficient r(k _i ) of the EEG signal X(k _i ) corresponding to each shifted M-sequence stimulus signal and the previously stored user Y(k _i ) is calculated to obtain r(k). ₁ ), r(k ₂ ), ..., r(k _T ), and the like. The calculation method and detailed description of the correlation coefficient are basically the same as the step 203 in the previous embodiment, and details are not described herein again. The correlation coefficients r(k _i ) calculated above are summed according to the corresponding weights W(k _i ), where r(k _i ) is the corresponding weight W(k _i ), indicating that the weight is for the shift k correlation coefficient r (k _{_i) i} M obtained stimulation signal sequence M (k _i) evoked EEG X (k _i) calculated in the final weights obtained in the weight coefficient. Similarly, r(k ₁ ) corresponds to W(k ₁ ), and r(k ₂ ) corresponds to W(k ₂ ). Multiplying each r(k _i ) by a corresponding weight W(k _i ), and summing all the obtained products to obtain a correlation coefficient between the final collected EEG signal and the user's certified EEG signal. Used for identity authentication.

Alternatively, a series of correlation coefficients r (k _i) calculated may be understood as a vector R, the respective correlation parameters r (k _i) corresponding to the weight W (k _i) is understood as a vector w, w each The element is the weight of the corresponding element in R. The inner product is calculated by using the vector R and the vector w to obtain the final total correlation coefficient. It can be understood that w is a column vector whose transposition is multiplied by R to represent the inner product.

Step 305: Determine whether the obtained correlation coefficient is greater than a matching coefficient threshold. If the matching coefficient threshold is greater than the matching coefficient threshold, the identity authentication is passed. Otherwise, the identity authentication fails. The specific implementation and detailed description are basically the same as the step 204 in the previous embodiment, and details are not described herein again.

Further, the M-sequence stimulation signal of each shift may be adjusted according to the influence of the correlation coefficient between the EEG signal corresponding to the M-sequence stimulus signal and the user's certified EEG signal on the user's identification accuracy. The weight of the correlation coefficient corresponding to the corresponding EEG signal. Each of the elements W(k _i ) in the above is the weight of the corresponding element r(k _i ) in R. Alternatively, all W(k _i ) can be initialized to the same size, both being 1/T. . The value of the element in w (ie, the weight) can be continuously updated according to the identity authentication situation. For example, when the correlation coefficient r(k _i ) corresponding to the M-sequence stimulation signal of a certain shift k _i is more easily distinguished from the real user. When compared with other users, the weight value corresponding to r(k _i ) can be increased, so that the correlation coefficient calculated by the E-sequence signal induced by the M-sequence obtained by each shift is summed according to the weight when the subsequent identity authentication is performed. The obtained acquired EEG signal is closer to the correlation coefficient of the user's certified EEG signal, and the identity authentication result is more credible.

Further, before generating the user-supplied M-sequence stimulation signal to stimulate the user, the identity authentication method in the embodiment of the present invention may further include the following steps: receiving T numbers sequentially input by the user, and determining the input T numbers and inputs. The order is the same as the shift number and shift order of the T-time shift.

Optionally, the sequence of the number of shifts of the M-sequence stimulus signal corresponding to the user is used as the personal identification number (PIN) to be input by the user using the identification system. If the error system is input, the system may choose to directly exit. , you can also start an alarm. When the PIN (ie, the sequence of shift numbers) input by the user coincides with the shift number and the shift order corresponding to the user stored in advance, the stimulus generating unit and the signal collecting unit are activated, and the system identity authentication proceeds to the next stage.

In the above method embodiment, the M-sequence stimulation signal is used to induce the user to generate an EEG signal, and the correlation coefficient between the collected EEG signal and the user's certified EEG signal is calculated, and the identity authentication result is determined by the correlation coefficient. The identification method utilizes the characteristics of the M-sequence stimulation signal, so that the EEG signal is easy to detect, the recognition accuracy is high, the EEG signal characteristics are stable and easy to collect, and the existing EEG-based identification methods are solved. problem. Based on the method embodiment corresponding to FIG. 2, the method of shifting the original M-sequence stimulus signal to generate the user-supplied M-sequence stimulus signal is adopted, and the correlation characteristics of the M-sequence stimulus signal are fully utilized, thereby further strengthening the identity. Accuracy and security of certification.

As shown in FIG. 7 , an embodiment of the present invention shows a safe identification system, wherein the stimulus generating unit and the signal analyzing unit can be integrated with the safe, and the signal collecting unit can be independent of the safe in order to facilitate the collection of the brain electrical signal. . Of course, this is only an example. In the specific implementation, these units may have other installation manners, and do not limit the scope of application and the scope of protection of the embodiments of the present invention. For the safe identification system shown in FIG. 7, an embodiment of the present invention provides a method for identifying an identity based on a visual evoked potential. Terms and conditions not defined in this embodiment For details, refer to the previous embodiment. The method comprises the following steps:

First, initialize the safe identification system. Determining a raw M-sequence stimulus signal for the user, length N=64, and selecting shift [12,8] as the shift sequence of the user, and realizing the real brain of the user induced by the two shifted M-sequence stimulus signals The electrical signals Y(12) and Y(8) are stored in the safe identification system. Set the initial matching coefficient threshold to 0.9.

In the identification, the new user who needs to be identified is provided with a wearable dry electrode cap (including a dry electrode placed at the scalp of the user's pillow area) to request safe identification. The system will ask the user to input the shift sequence if the user inputs [6,12,15] and other non-set sequences, the safe feedback user authentication failed, and the alarm is activated. If the user inputs [12, 8] correctly, the stimulation generating unit of the system is activated, and the M-sequence stimulation signal corresponding to the user is generated to authenticate the user. At the same time, the synchronization signal is sent to the electrode cap by wireless signal transmission to complete the synchronization of the stimulation and the acquisition signal.

The stimulation generating unit first blinks the signal with the M sequence of the shifted 12 for a period of time, and then blinks the signal with the M sequence of the shifted 8 for a period of time. The real-time EEG signal collected by the electrode cap is segmented and taken to X(12) and X(8) according to the received synchronization signal, and transmitted to the signal analysis unit in the safe identification system by wireless transmission.

The signal analysis unit calculates the linear correlation coefficient r(12) of X(12) and Y(12), and the correlation coefficient r(8) of X(8) and Y(12), respectively.

Calculate the final correlation coefficient used for identity authentication, that is, the correlation coefficient between the acquired EEG signal and the user's certified EEG signal = 0.5 * r (12) + 0.5 * r (8), here assume that the initial shift is set. The weights of the correlation coefficients of the M-sequence stimulus signals are the same, that is, 1/2=0.5. If the calculation result is >0.9, the identity authentication is successful; otherwise, the authentication fails and the alarm is started.

If the subsequent discovery, r(8) can reflect the difference between the real user and the non-real user for the result of the identity authentication, the corresponding weight can be adjusted, and the collected EEG signal and the user's authentication EEG signal can be updated. The correlation coefficient = 0.2 * r (12) + 0.8 * r (8), that is, the weight of r (8) is raised to 0.8, and the weight of r (8) is correspondingly reduced to 0.2, thereby improving the accuracy of authentication.

The method provided by the embodiment of the present invention is described in detail above with reference to FIG. 1 to FIG. 7 . FIG. 8 shows a possible structural diagram of a visual evoked potential based identification device according to the present application. The identity recognition device can implement the functions of the visual evoked potential based identification device in the method embodiment of FIG. 2 and FIG. 3 above, and the terms and implementation details not defined in this embodiment Reference may be made to the method embodiments of Figures 2 and 3 above. As shown in FIG. 8, the visual evoked potential based identification device 40 includes a stimulation generating unit 41, a signal acquisition unit 42, and a signal analysis unit 43. The stimulus generating unit 41 is configured to generate a user-supplied M-sequence stimulation signal to stimulate the user; the signal acquisition unit 42 is configured to collect an EEG signal generated by the user for the M-sequence stimulation signal; and the signal analysis unit 43 is configured to calculate the signal acquisition. The correlation coefficient between the EEG signal collected by the unit 42 and the authenticated EEG signal of the user determines whether the calculated correlation coefficient is greater than the matching coefficient threshold. If the correlation coefficient is greater than the matching coefficient threshold, the identity authentication passes; otherwise, the identity authentication fails. . Here, the user's authenticated EEG signal is an EEG signal generated by the user for the corresponding M-sequence stimulus signal for identification, and the correlation coefficient is used to measure the degree of similarity between the two signals.

Alternatively, the stimulation generating unit 41 may be an electronic device that can generate a stimulation signal that changes in the form of an M sequence and visualize the display to the user.

Alternatively, the signal acquisition unit 42 can include a wearable brain electrical collection device that includes one or more dry electrodes for the user to wear.

The visual evoked potential based identification device provided by the embodiment uses the M-sequence stimulation signal to induce the user to generate a specific mode of the EEG signal, and calculates the correlation coefficient between the collected EEG signal and the user's certified EEG signal, and correlates The coefficient determines the identity authentication result. Because the identity recognition device utilizes the characteristics of the M-sequence stimulation signal, the EEG signal is easy to detect, the EEG signal correlation between different objects is small, and the EEG signal characteristics are stable and easy to collect, thus solving various existing EEG-based modes. The problem with the identification method.

Optionally, as shown in FIG. 9, on the basis of the identity identification device signal 40 shown in FIG. 8, further, the analysis unit 43 includes a calculation subunit 431 and an authentication subunit 432, and the calculation subunit 431 is used for calculation and acquisition. Correlation coefficient between the EEG signal and the user's certified EEG signal; the authentication subunit is configured to determine whether the correlation coefficient calculated by the calculation subunit is greater than a matching coefficient threshold, and if the correlation coefficient is greater than the matching coefficient threshold, the identity authentication is passed Otherwise, the identity authentication does not pass.

Optionally, the M-sequence stimulus signal corresponding to the user is obtained by performing T-time shift on the original M-sequence stimulus signal, wherein the length of the original M-sequence stimulus signal is N, 0≤T≤N-1. A plurality of shift numbers k can be selected for the original M-sequence stimulus signal, that is, a plurality of M-series stimulation signals are obtained for stimulating the user. Optionally, K _i is a shift of the T _ith shift in the T shift of the original M sequence stimulation signal, 0≤K _i ≤N-1, and the stimulus generating unit 41 is specifically configured to target the original M sequence The stimulation signal is sequentially shifted by K _{i to} generate the T- _th order M-sequence stimulation signal, and the T- _th M-sequence stimulation signal is used to stimulate the user. Then, as shown in FIG. 9, further, the signal acquisition unit 42 includes a collection subunit 421 and a clipping subunit 422, wherein the acquisition subunit 421 is configured to collect an EEG signal generated by the user for the M sequence stimulation signal, and intercept the subunit. 422 is configured to intercept the EEG signal collected by the acquisition subunit 421 according to the stimulation sequence of the M-sequence stimulation signal of the T-shift, to obtain an EEG signal corresponding to the M-sequence stimulation signal of each shift. The method and detailed description of generating the M series stimulation signal corresponding to the user and collecting the EEG signal by shifting are basically the same as those of the previous embodiment, and details are not described herein again.

Optionally, the stimulation generating unit 41 is further configured to generate a synchronization signal when the M-sequence stimulation signal is started to be stimulated by the M-th order stimulation signal, and the intercepting sub-unit 422 is specifically configured to: according to the collected EEG signal The stimulation sequence of the M-sequence stimulation signal of the T-shift is intercepted according to the synchronization signal generated by the received stimulation-generating unit 41, and the EEG signal corresponding to the M-sequence stimulation signal of each shift is obtained.

Correspondingly, the calculating sub-unit 431 is specifically configured to separately calculate correlation coefficients of the EEG signals corresponding to the M-sequence stimulation signals of each of the acquired offsets and the authentication EEG signals of the user, and correspondingly calculate the correlation coefficients according to each time. The weights are summed to obtain the correlation coefficient between the collected EEG signals and the user's certified EEG signals. Optionally, the signal analysis unit 43 further includes a weight management sub-unit 433, configured to accurately identify the user according to the correlation coefficient between the EEG signal corresponding to the M-sequence stimulus signal and the user's certified EEG signal. The effect of adjusting the weight of the correlation coefficient corresponding to the EEG signal corresponding to each shifted M-sequence stimulus signal. The calculation method and detailed description of the correlation coefficient are basically the same as those of the previous embodiment, and are not described herein again.

Further, the identification device 40 further includes a key authentication unit 44, configured to receive T numbers sequentially input by the user, and determine the input T numbers and inputs before generating the user-supplied M-sequence stimulation signal to stimulate the user. The order is the same as the shift number and shift order of the preset T-time shift.

The identification device 40 shown in FIG. 9 adopts a method of shifting the original M-sequence stimulation signal to generate a user-supplied M-sequence stimulation signal, and fully utilizing the correlation characteristics of the M-sequence stimulation signal, The accuracy and security of identity authentication are further enhanced.

FIG. 10 schematically illustrates another identity recognition device 40 in accordance with an embodiment of the present invention. As shown in FIG. 10, the signal analysis unit 43 includes a processor 531 and a memory 532 for storing a computer. Executing the instruction, the processor 531 executes a computer execution instruction to execute the function of the signal analysis unit described in the previous embodiment, and is used for calculating a correlation coefficient between the collected EEG signal and the authenticated EEG signal of the user, and determining whether the correlation coefficient is If the correlation coefficient is greater than the matching coefficient threshold, the identity authentication passes; otherwise, the identity authentication fails.

The processor 531 can be a general-purpose central processing unit (CPU), a microprocessor, an application specific integrated circuit (ASIC), or one or more integrated circuits for executing related programs. The technical solution provided by the embodiment of the present invention is implemented.

The memory 532 may be a read only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). When the technical solution provided by the embodiment of the present invention is implemented by software or firmware, the program code for implementing the technical solution provided by the embodiment of the present invention is stored in the memory 532 and executed by the processor 531.

Specifically, the memory 532 may be used to store computer execution instructions, and may also be used to store various information, for example, a preset matching coefficient threshold, an EEG signal generated by the user for the corresponding M-sequence stimulus signal, and the like for identification. . The processor 531 can read the information stored by the memory 532 or store the collected information to the memory 532. Further, when the identification device 40 is in operation, the processor 531 can read the computer execution instructions stored in the memory 532 to perform the functions of the signal analysis unit 43 described in the previous embodiment.

Similar to the technical solution in the previous embodiment, in the technical solution applied by the identity recognition device 40 shown in FIG. 10, the M-sequence stimulation signal corresponding to the user may also be obtained by performing T-transmission on the original M-sequence stimulation signal, wherein The length of the original M-sequence stimulus signal is N, 0 ≤ T ≤ N-1. K _i is the shift of the T _ith shift in the T-time shift, 0≤K _i ≤N-1, then the stimulus generating unit 41 is specifically used to sequentially shift K for the original M-sequence stimulus signal. _i generates a T- _th order M-sequence stimulus signal, and stimulates the user with the T- _th M-sequence stimulus signal. The signal acquisition unit 42 is specifically configured to collect an EEG signal generated by the user for the M-sequence stimulation signal, and intercept the acquired EEG signal according to the stimulation sequence of the M-sequence stimulation signal of the T-time shift, to obtain the M for each shift. The EEG signal corresponding to the sequence stimulation signal. Optionally, the stimulation generating unit 41 is configured to generate a synchronization signal when the M-sequence stimulation signal is started to be stimulated by the M-th order stimulation signal, and the signal acquisition unit 42 is specifically configured to collect the user generated for the M-sequence stimulation signal. The EEG signal is obtained by intercepting the acquired EEG signal according to the stimulation sequence of the M-sequence stimulation signal of the T-shift, according to the synchronization signal, to obtain an EEG signal corresponding to the M-sequence stimulation signal of each shift.

Further, the processor 531 in the signal analyzing unit 40 shown in FIG. 10 is specifically configured to separately calculate a correlation coefficient between the EEG signal corresponding to each acquired M-sequence stimulation signal and the authenticated EEG signal of the user, The correlation coefficient obtained each time is summed according to the corresponding weight, and the correlation coefficient between the collected EEG signal and the user's certified EEG signal is obtained.

Further, the processor 531 is further configured to adjust the influence of the correlation coefficient between the EEG signal corresponding to the M-sequence stimulus signal and the authenticated EEG signal of the user on the identification accuracy of the user, and adjust each shift. The weight of the correlation coefficient corresponding to the EEG signal corresponding to the M-sequence stimulus signal.

Further, the identification device 40 further includes an input/output (I/O) unit 54 for receiving T numbers sequentially input by the user before generating the user-supplied M-sequence stimulation signal to stimulate the user. The processor 531 is further configured to determine that the input T numbers and the input order are the same as the preset shift number and shift order of the T times shift. Alternatively, the input/output unit 54 herein may be an input keyboard.

Although the signal analysis unit 43 shown in FIG. 10 only shows the processor 531 and the memory 532, it will be understood by those skilled in the art in the specific implementation process that other devices necessary for normal operation are also included. For example, a communication interface and a bus, wherein the communication interface can employ a transceiver such as, but not limited to, a transceiver for implementing communication between the signal analysis unit 43 and the signal acquisition unit 42 and the stimulation generation unit 41. The bus can include a path for transferring information between the processor 531 and the memory 532. The bus may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, and at the same time, those skilled in the art will appreciate that the identification device 40 illustrated in FIG. 10 may also include hardware devices that implement other additional functions, as desired.

Those skilled in the art will appreciate that the various method steps and elements described in connection with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both, in order to clearly illustrate hardware and software. Interchangeability, in accordance with the functional generality in the above description The steps and composition of the various embodiments are described. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the solution. Different methods may be used to implement the described functionality for each particular application, but such implementation should not be considered to be beyond the scope of the present invention.

A person skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the device and the unit described above can refer to the corresponding process in the foregoing method embodiment, and details are not described herein again.

In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the device embodiments described above are merely illustrative. For example, the division of the unit/module is only a logical function division. In actual implementation, there may be another division manner, for example, multiple units or components may be used. Combinations can be integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, or an electrical, mechanical or other form of connection.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the embodiments of the present invention.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a standalone product, may be stored in one computer computer readable storage medium or as one or more instructions or code embodied on a computer readable medium. transmission. Computer readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one location to another. A storage medium may be any available media that can be accessed by a computer. By way of example and not limitation, computer readable media may comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, disk storage media or other magnetic storage device, or can be used for carrying or storing The desired program code in the form of a data structure or any other medium that can be accessed by a computer. Also. Any connection may suitably be a computer readable medium. For example, if the software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable , fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, wireless, and microwave are included in the definition of the medium to which they belong. As used in the present invention, a disk and a disc include a compact disc (CD), a laser disc, a compact disc, a digital versatile disc (DVD), a floppy disk, and a Blu-ray disc, wherein the disc is usually magnetically copied, and the disc is The laser is used to optically replicate the data. Combinations of the above should also be included within the scope of the computer readable media. Based on such understanding, the technical solution of the present invention is essential or part of the prior art, or all or part of the technical solution may be stored in a storage medium, including a plurality of instructions for causing a computer device (may be a personal computer, server, or network device, etc.) performing all or part of the steps of the methods described in various embodiments of the present invention.

It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not limited thereto; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that The technical solutions described in the foregoing embodiments are modified, or some of the technical features are equivalently replaced; and the modifications or substitutions do not deviate from the technical scope of the embodiments of the present invention.

Claims

An identification method based on visual evoked potentials, comprising:

Generating a user-supplied M-sequence stimulus signal to stimulate the user;

Acquiring an EEG signal generated by the user for the M-sequence stimulation signal;

Calculating a correlation coefficient between the collected EEG signal and the authenticated EEG signal of the user, the authenticated EEG signal of the user is an EEG generated by the user for the corresponding M-sequence stimulus signal for identification a signal that is used to measure the degree of similarity between two signals;

If the correlation coefficient is greater than the matching coefficient threshold, the identity authentication is passed; otherwise, the identity authentication fails.
The method according to claim 1, wherein the M-sequence stimulation signal corresponding to the user is obtained by performing T-transmission on the original M-sequence stimulation signal, wherein the length of the original M-sequence stimulation signal is N. , 0 ≤ T ≤ N-1.
The method according to claim 2, wherein K i is a shift number of the T ith shift in the T-time shift, and 0 ≤ K i ≤ N-1,

Generating a user-supplied M-sequence stimulation signal to stimulate the user, including:

And for the original M-sequence stimulation signal, sequentially shifting K i to generate the T- th M-sequence stimulation signal, and using the T- th M-sequence stimulation signal to stimulate the user;

Before the calculating the correlation coefficient between the collected EEG signal and the authenticated EEG signal of the user, the method further includes:

The collected EEG signals are intercepted according to the stimulation order of the M-sequence M-stimulus signals of the T-shifts, and the EEG signals corresponding to the M-sequence stimulation signals of each shift are obtained.
The method of claim 3, wherein the method further comprises:

At the beginning of the T i times with the M-sequence shifted stimulation signal to stimulate the user, generating a synchronization signal; the

The collected EEG signals are intercepted according to the T-shifted M-sequence stimulation signal stimulation sequence, and the EEG signals corresponding to the M-sequence stimulation signals of each shift are obtained, including: The EEG signal is intercepted according to the stimulation sequence of the M-sequence stimulation signal of the T-shift, according to the synchronization signal, to obtain an EEG signal corresponding to the M-sequence stimulation signal of each shift.
The method according to claim 3 or 4, wherein the calculating a correlation coefficient between the collected EEG signal and the authenticated EEG signal of the user comprises:

Calculating, respectively, a correlation coefficient between the collected EEG signal corresponding to the M-sequence stimulation signal of each shift and the authenticated EEG signal of the user;

The correlation coefficient obtained by each calculation is summed according to the corresponding weight, and the correlation coefficient between the collected EEG signal and the authenticated EEG signal of the user is obtained.
The method of claim 5, wherein the method further comprises:

Adjusting the M of each shift according to the influence of the correlation coefficient between the EEG signal corresponding to the M-sequence stimulus signal and the authenticated EEG signal of the user on the identification accuracy of the user The weight of the correlation coefficient corresponding to the EEG signal corresponding to the sequence stimulation signal.
The method according to any one of claims 2 to 6, wherein before the generating the user-supplied M-sequence stimulation signal to stimulate the user, the method further comprises:

Receiving T numbers sequentially input by the user, determining that the input T numbers and input order are the same as the shift number and shift order of the T times shift.
A visual evoked potential based identification device comprising:

a stimulus generating unit, configured to generate a user-supplied M-sequence stimulation signal to stimulate the user;

a signal acquisition unit, configured to collect an EEG signal generated by the user for the M-sequence stimulation signal;

a signal analysis unit, configured to calculate a correlation coefficient between the collected EEG signal and the authenticated EEG signal of the user, and determine whether the correlation coefficient is greater than a matching coefficient threshold, if the correlation coefficient is greater than the matching coefficient threshold , identity authentication passed, otherwise, identity authentication does not pass. Wherein, the authenticated EEG signal of the user is an EEG signal generated by the user for the identification of the corresponding M-sequence stimulus signal, and the correlation coefficient is used to measure the degree of similarity between the two signals.
The identity recognition device according to claim 8, wherein the signal analysis unit comprises:

a calculating subunit, configured to calculate a correlation coefficient between the collected EEG signal and the authenticated EEG signal of the user;

The authentication subunit is configured to determine whether the correlation coefficient is greater than a matching coefficient threshold. If the correlation coefficient is greater than the matching coefficient threshold, the identity authentication passes, otherwise, the identity authentication fails.
The identity recognition device according to claim 9, wherein the M-sequence stimulation signal corresponding to the user is obtained by performing T-transmission on the original M-sequence stimulation signal, wherein the length of the original M-sequence stimulation signal is It is N, 0 ≤ T ≤ N-1.
The identity recognition device according to claim 10, wherein K i is a shift of the T ith shift in the T-time shift, 0 ≤ K i ≤ N-1,

The stimulation generation unit is configured for stimulation of the original M-sequence signal, generated sequentially shifts the K i of the first sequence of M times T i stimulation signal with said second sequence of M times T i for the stimulation signal Said that the user is stimulating;

The signal acquisition unit includes:

a collecting subunit, configured to collect an EEG signal generated by the user for the M sequence stimulation signal;

The intercepting subunit is configured to intercept the collected EEG signal according to the stimulation sequence of the M-sequence M-stimulus signal of the T-time shift, to obtain an EEG signal corresponding to the M-sequence stimulation signal of each shift.
The identity recognition device of claim 11 wherein:

The stimulation generating unit is further configured to generate a synchronization signal when the M-sequence stimulation signal that is shifted by the Tith is started to stimulate the user;

The intercepting subunit is specifically configured to intercept the collected EEG signal according to the stimulation sequence of the M-sequence M-stimulus signal of the T-time shift according to the synchronization signal, to obtain an M sequence for each shift The EEG signal corresponding to the stimulation signal.
The identity recognition device according to claim 11 or 12, wherein the calculation subunit is specifically configured to:

Calculating, respectively, a correlation coefficient between the collected EEG signal corresponding to the M-sequence stimulation signal of each shift and the authenticated EEG signal of the user;

The correlation coefficient obtained by each calculation is summed according to the corresponding weight, and the correlation coefficient between the collected EEG signal and the authenticated EEG signal of the user is obtained.
The identity recognition device according to claim 13, wherein the signal analysis unit further comprises:

a weight management subunit, configured to determine, according to the correlation coefficient between the EEG signal corresponding to the M-sequence stimulation signal and the authentication EEG signal of the user, the identification accuracy of the user The effect of adjusting the weight of the correlation coefficient corresponding to the EEG signal corresponding to the M-sequence stimulus signal of each shift.
The identity recognition device according to any one of claims 10 to 14, further comprising:

a key authentication unit, configured to receive T digits sequentially input by the user before determining the M-sequence stimulation signal corresponding to the user, and determine T input digits and input order of the input The number of shifts of the T-time shift is the same as the shift order.
The identity recognition device of claim 8 wherein said signal analysis unit comprises a processor for storing computer execution instructions, said processor executing said computer execution instructions for computing And determining, by the correlation coefficient between the collected EEG signal and the authenticated EEG signal of the user, whether the correlation coefficient is greater than a matching coefficient threshold, and if the correlation coefficient is greater than the matching coefficient threshold, the identity authentication is passed; otherwise, Identity authentication does not pass.
The identity recognition device according to claim 16, wherein the M-sequence stimulation signal corresponding to the user is obtained by performing T-time shifting on the original M-sequence stimulation signal, wherein the length of the original M-sequence stimulation signal is It is N, 0 ≤ T ≤ N-1.
The identity recognition device according to claim 17, wherein K i is a shift of the T ith shift in the T shift, 0 ≤ K i ≤ N-1,

The stimulation generation unit is configured for stimulation of the original M-sequence signal, generated sequentially shifts the K i of the first sequence of M times T i stimulation signal with said second sequence of M times T i for the stimulation signal Said that the user is stimulating;

The signal acquisition unit is specifically configured to collect an EEG signal generated by the user for the M-sequence stimulation signal, and generate the EEG signal according to the stimulation sequence of the M-sequence stimulation signal of the T-time shift The interception is performed to obtain an EEG signal corresponding to the M-sequence stimulation signal of each shift.
The identity recognition device of claim 18, wherein:

The stimulation generating unit is further configured to generate a synchronization signal when the M-sequence stimulation signal that is shifted by the Tith is started to stimulate the user;

The signal acquisition unit is specifically configured to collect an EEG signal generated by the user for the M-sequence stimulation signal, and generate the EEG signal according to the stimulation sequence of the M-sequence stimulation signal of the T-time shift And intercepting according to the synchronization signal, and obtaining an EEG signal corresponding to the M-sequence stimulation signal of each shift.
The identity recognition device according to claim 18 or 19, wherein the processor is specifically configured to:

Calculating, respectively, a correlation coefficient between the collected EEG signal corresponding to the M-sequence stimulation signal of each shift and the authenticated EEG signal of the user;

The correlation coefficient obtained by each calculation is summed according to the corresponding weight, and the correlation coefficient between the collected EEG signal and the authenticated EEG signal of the user is obtained.
The identity recognition device according to claim 20, wherein the processor is further configured to:

Adjusting the M of each shift according to the influence of the correlation coefficient between the EEG signal corresponding to the M-sequence stimulus signal and the authenticated EEG signal of the user on the identification accuracy of the user The weight of the correlation coefficient corresponding to the EEG signal corresponding to the sequence stimulation signal.
The identity recognition device according to any one of claims 17 to 21, further comprising:

An input/output unit, configured to receive T numbers sequentially input by the user before the generating the user-supplied M-sequence stimulation signal to stimulate the user;

The processor is further configured to: determine that the T numbers and input orders of the input are the same as the shift number and shift order of the T times of shifts.