High-frequency cold-start visual evoked potential brain-computer interface system, device, equipment and medium

Abstract

The invention relates to the technical field of brain-computer interfaces, and discloses a high-frequency cold-start visual evoked potential brain-computer interface system, device, equipment and medium, the system comprises an instruction sequence generation module, a local dynamic flicker module, a user display interface and a decoding module, determining an integer instruction sequence length enabling the flicker frequency to be within a first preset range; determining a target instruction sequence according to the integer instruction sequence length; the local dynamic flicker module performs local dynamic flicker icon sorting on the target instruction sequence to determine a local dynamic flicker icon display sequence; the user display interface displays the local dynamic flicker icon display sequence to the target user; the decoding module receives a target electroencephalogram signal generated after a target user watches the local dynamic flicker icon display sequence; and the decoding module decodes the target electroencephalogram signal to obtain the target intention of the target user, so that the practicability of the system is improved.

Description

Technical Field

[0001] The present invention relates to the field of brain-computer interface technology, and in particular to a high-frequency cold-start visual evoked potential brain-computer interface system, device, equipment and medium. Background Art

[0002] Brain-computer interface technology provides users with a way to bypass the nerve-muscle channel and directly use the neural channel to interact with the outside world. With the development of society and brain-computer interface technology, high performance is no longer the only dimension for evaluating brain-computer interfaces. Requirements such as the diversity of application scenarios, user comfort and convenience are gradually highlighted. The existing stimulation paradigm has low-frequency stimulation that can easily cause user fatigue and even induce epilepsy; the stimulation mode cannot effectively avoid surrounding interference, resulting in a decrease in decoding accuracy; the phase-encoded stimulation paradigm decoding relies on precise triggering information and cannot achieve a cold start of the decoding algorithm. These problems limit the visual evoked potential brain-computer interface system, especially in practical applications in the fields of cognitive training and neurorehabilitation. Therefore, how to improve the performance of the visual evoked potential brain-computer interface system in practical applications has become an urgent problem to be solved. Summary of the invention

[0003] The present invention provides a high-frequency cold-start visual evoked potential brain-computer interface system, device, equipment and medium, which can solve the technical problem of how to improve the practicality of the visual evoked potential brain-computer interface system.

[0004] In the first aspect, a high-frequency cold-start visual evoked potential brain-computer interface system is provided, wherein the high-frequency cold-start visual evoked potential brain-computer interface system comprises an instruction sequence generation module, a local dynamic flickering module, a user display interface and a decoding module, wherein:

[0005] The instruction sequence generation module determines the integer instruction sequence length according to the instruction target number and the device refresh rate, so that the flickering frequency under the integer instruction sequence length is within a first preset range; wherein the flickering frequency is the quotient of the device refresh rate and the integer instruction sequence length;

[0006] The instruction sequence generation module determines the target instruction sequence according to the integer instruction sequence length;

[0007] The local dynamic flashing module sorts the local dynamic flashing icons of the target instruction sequence and determines the local dynamic flashing icon display sequence;

[0008] The user display interface displays a local dynamic flashing icon display sequence to the target user;

[0009] The decoding module receives the target EEG signal generated by the target user after watching the local dynamic flashing icon display sequence;

[0010] The decoding module decodes the target EEG signal to obtain the target intention of the target user.

[0011] In a second aspect, a high-frequency cold-start visual evoked potential brain-computer interface device is provided, comprising:

[0012] A first determination module is used to determine the integer instruction sequence length according to the instruction target number and the device refresh rate, so that the flickering frequency under the integer instruction sequence length is within a first preset range; wherein the flickering frequency is the quotient of the device refresh rate and the integer instruction sequence length;

[0013] A second determination module is used to determine a target instruction sequence according to the length of the integer instruction sequence;

[0014] A third determination module is used to sort the local dynamic flashing icons of the target instruction sequence and determine the local dynamic flashing icon display sequence;

[0015] A display module, used to display a local dynamic flashing icon display sequence to a target user;

[0016] A receiving module, used for receiving a target electroencephalogram signal generated by a target user after watching a local dynamic flashing icon display sequence;

[0017] The decoding module is used to decode the target EEG signal to obtain the target intention of the target user.

[0018] In a third aspect, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, processing steps of a brain-computer interface system based on high-frequency cold-start visual evoked potential are implemented.

[0019] In a fourth aspect, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the processing steps of a brain-computer interface system based on high-frequency cold-start visual evoked potential are implemented.

[0020] In the scheme implemented by the above-mentioned high-frequency cold-start visual evoked potential brain-computer interface system, device, equipment and medium, the instruction sequence generation module in the high-frequency cold-start visual evoked potential brain-computer interface system can determine the integer instruction sequence length that makes the flashing frequency within the first preset range according to the instruction target number and the device refresh rate, and determine the target instruction sequence according to the integer instruction sequence length. The local dynamic flashing module sorts the target instruction sequence by local dynamic flashing icons and determines the local dynamic flashing icon display sequence. The user display interface displays the local dynamic flashing icon display sequence to the target user. The decoding module receives the target EEG signal generated by the target user after watching the local dynamic flashing icon display sequence, and decodes the target EEG signal to obtain the target user's target intention. While improving the effectiveness of intention recognition, it also improves the comfort of the target user's use process and improves the practicality of the visual evoked potential brain-computer interface system. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0022] Figure 1 This is a schematic diagram of an application environment of a high-frequency cold-start visual evoked potential brain-computer interface system in one embodiment of the present invention;

[0023] Figure 2 It is a processing flow diagram of a high-frequency cold-start visual evoked potential brain-computer interface system in one embodiment of the present invention;

[0024] Figure 3 is a schematic diagram of a user display interface in one embodiment of the present invention;

[0025] Figure 4 2 is a schematic diagram of a high-frequency cold-start visual evoked potential brain-computer interface device according to an embodiment of the present invention;

[0026] Figure 5 is a schematic diagram of a structure of a computer device in one embodiment of the present invention;

[0027] Figure 6 It is another structural schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0030] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments.

[0031] The high-frequency cold-start visual evoked potential brain-computer interface system provided by the embodiment of the present invention can be applied in Figure 1 In the application environment. Exemplarily, taking the scenario in which the target user generates a target EEG signal after viewing a visual stimulation sequence and identifies the target intention of the target EEG signal as an example, the target user can view the visual stimulation sequence generated and selected by the high-frequency cold-start visual evoked potential brain-computer interface system through the user display interface (such as the target instruction sequence to be mentioned below), and the target EEG signal generated by the target user after viewing is collected by the high-frequency cold-start visual evoked potential brain-computer interface system, thereby decoding the target EEG signal through the decoding module to obtain the target intention of the target user. The high-frequency cold-start visual evoked potential brain-computer interface system provided by this application can improve the effectiveness of intention recognition while improving the comfort of the target user's use process, which is conducive to improving the practicality of the visual evoked potential brain-computer interface system. Among them, the target user can be a user of the high-frequency cold-start visual evoked potential brain-computer interface system, that is, a user who needs to identify the intention of his target EEG signal, and this application does not limit this.

[0032] The high-frequency cold-start visual evoked potential brain-computer interface system includes an instruction sequence generation module, a local dynamic flicker module, a user display interface and a decoding module. Among them, the instruction sequence generation module can be used to generate a visual stimulus sequence (i.e., an instruction sequence) presented to the user. Since different instruction sequences can trigger EEG signals of different characteristics, the present application can generate and select an instruction sequence that meets the requirements by the instruction sequence generation module, such as a high-frequency periodic instruction sequence can be generated, so that the required EEG signal can be more effectively stimulated to improve the accuracy and stability of the high-frequency cold-start visual evoked potential brain-computer interface system. The local dynamic flicker module can be used to guide the user's attention to focus on specific visual stimuli. The local dynamic flicker module can help the user accurately focus on key stimulus elements in a complex visual environment, thereby reducing the influence of surrounding interference information and improving the interaction effect of the brain-computer interface. The user display interface can be a direct window for the user to interact with the brain-computer interface system. The user display interface can show the user visual stimuli selected by the instruction sequence generation module and designed by the local dynamic flicker module. The decoding module can analyze and decode the collected EEG signals through a decoding algorithm to extract information related to the user's intention, thereby accurately determining the user's target intention.

[0033] See also Figure 2 As shown, Figure 2 A schematic diagram of a processing flow of a high-frequency cold-start visual evoked potential brain-computer interface system provided by an embodiment of the present invention includes the following steps:

[0034] S10: The instruction sequence generation module determines the integer instruction sequence length according to the instruction target number and the device refresh rate, so that the flickering frequency under the integer instruction sequence length is within a first preset range; wherein the flickering frequency is the quotient of the device refresh rate and the integer instruction sequence length.

[0035] The instruction target number may be used to indicate the number of display icons corresponding to the target instruction sequence. For example, the target instruction sequence may correspond to 12 display icons, and the instruction target number may be 12. Optionally, the instruction target number may be preset by the user or determined by the system by default, and this application does not limit this.

[0036] The device refresh rate can be used to indicate the refresh rate of the display device used at the time. It is understandable that a device with a high refresh rate can display subtle visual changes more clearly, which may be more beneficial for the presentation of high-frequency stimuli. For example, when the device refresh rate is very high, the high refresh rate device can accurately display high-frequency visual stimuli that flicker quickly. At this time, the instruction sequence generation module can select a relatively high flicker frequency value, and this application does not limit this.

[0037] The integer instruction sequence length can be used to indicate the length of the integer value to which the initial instruction sequence can correspond. The initial instruction sequence can be understood as a predetermined or system-default instruction sequence. For example, the instruction sequence generation module can determine that the integer instruction sequence length of the initial instruction sequence is 4 based on the instruction target number and the device refresh rate, and the present application does not limit this. Optionally, the initial instruction sequence can be one or more, and the integer instruction sequence length corresponding to each initial instruction sequence can be the same or different, and the present application does not limit this.

[0038] The initial command sequence can be a command sequence under the traditional Steady-State Visual Evoked Potential (SSVEP) stimulation paradigm. It is understandable that although the SSVEP stimulation paradigm is widely used in brain-computer interface research and some visual neuroscience research, it also has some limitations, such as visual fatigue and limitation on the stimulation frequency range. Therefore, this application makes corresponding adjustments to the initial command sequence to obtain a target command sequence that better meets the research needs, thereby achieving a brain-computer interaction process that is more comfortable, more convenient to use, and more accurate.

[0039] In the relevant field of high-frequency cold-start visual evoked potential brain-computer interface involved in this application, the flicker frequency mainly refers to a specific frequency that can effectively induce visual evoked potential (VEP) and has certain characteristics that can be used to distinguish different states or convey specific information. In other words, the instruction sequence generation module can determine a flicker frequency that has certain characteristics, can convey characteristic information, can effectively induce visual evoked potential, and meets the current device refresh rate based on relevant factors such as the number of instruction targets, the length of the integer instruction sequence, and the device refresh rate, so that in subsequent steps, a more accurate and appropriate target instruction sequence can be determined based on the flicker frequency.

[0040] Optionally, the instruction sequence generation module determines the flashing frequency according to the integer instruction sequence length and the device refresh rate, as shown in the following formula:

[0041]

[0042] Among them, f N Can be used to indicate flicker frequency, such as the basic flicker frequency, i.e. fundamental frequency; f r It can be used to indicate the device refresh rate; N1 can be used to indicate the length of an integer instruction sequence.

[0043] According to the flicker frequency, SSVEP can be divided into a low frequency band (8-15Hz), a medium frequency band (15-30Hz) and a high frequency band (≥30Hz). Therefore, in order to efficiently induce a more appropriate EEG signal, and accurately detect and interpret the EEG signal, the present application can control the flicker frequency under the integer instruction sequence length within a first preset range. Optionally, the first preset range can be 4Hz-30Hz, and the present application does not limit this.

[0044] S20: The instruction sequence generation module determines the target instruction sequence according to the integer instruction sequence length.

[0045] The present application proposes a new stimulation paradigm in which a pseudo-random binary sequence (PRBS) is superimposed within each cycle of the basic flickering frequency of the SSVEP stimulation paradigm, so that the resulting visual stimulation still has the periodic characteristics of the basic flickering frequency and has specific frequency characteristics in the frequency domain, which can facilitate subsequent decoding work. At the same time, the superimposed PRBS sequence in each cycle is visually presented as a high-frequency flickering signal to improve the comfort and safety of the system.

[0046] The essence of superimposing PRBS on SSVEP is to make the PRBS sequence appear periodically, so that the broadband characteristics of the PRBS sequence can be observed within one cycle, while the periodic characteristics are still exhibited over a long period of time. Finally, the frequency corresponding to the cycle and its multiples can be obtained from the frequency domain analysis, and this application does not impose any restrictions on this.

[0047] Therefore, after the instruction sequence generation module determines the integer instruction sequence length, the PRBS sequence to be superimposed can be optimized and designed according to the integer instruction sequence length to obtain the target instruction sequence.

[0048] This application selects the candidate PRBS sequences to ensure that each selected PRBS sequence has the best EEG response performance at this length. The stimulation paradigm of this study needs to maximize the amplitude at each frequency multiple, or maximize the signal-to-noise ratio corresponding to the EEG evaluation signal generated after the maximum EEG response evaluation, and use this as the characteristic value to traverse all possible PRBS sequences of the specified length, and finally obtain the optimal PRBS sequence, which is not limited by this application.

[0049] Based on the traditional steady-state visual evoked potential stimulation paradigm, a new stimulation paradigm that superimposes a PRBS sequence in each stimulation cycle can improve user comfort by introducing high-frequency components in each cycle. Moreover, since the stimulation paradigm still retains the periodicity similar to the steady-state visual evoked potential stimulation paradigm, the final induced potential is similar to the steady-state visual evoked potential and has obvious frequency characteristics, which is conducive to selecting a suitable template in the decoding process and directly using an unsupervised algorithm to decode and analyze it. The present application does not impose any restrictions on this.

[0050] S30: The local dynamic flashing module sorts the local dynamic flashing icons of the target instruction sequence to determine a local dynamic flashing icon display sequence.

[0051] Among them, the local dynamic flashing icon display sequence can be used to indicate that the icons corresponding to the target instruction sequence are localized and the instruction icon contents are gradually displayed, thereby obtaining an icon display sequence. It is understandable that after using the VEP brain-computer interface system for a long time, the user's vigilance (or attention) decreases due to fatigue. One of the important reasons is that the current visual stimulus image display form is too single, and the overall display of the icon content causes the user to not pay attention to the flashing content, but only to the flashing itself.

[0052] The present application improves the new stimulation paradigm, that is, adds a local dynamic flicker module (also referred to as a local dynamic micro-flicker module) to guide the user's attention point when viewing the instruction sequence, so that the user pays attention to the semantic information of the flickering content, thereby maintaining the user's vigilance at a high level, thereby effectively improving the final performance of the system, and the present application does not limit this. Optionally, the operation of the local dynamic flicker module to sort the target instruction sequence by local dynamic flickering icons can also be referred to as the operation of the local dynamic flicker module to process the target instruction sequence by local dynamic micro-flickering, and the present application does not limit this.

[0053] Optionally, the local dynamic flashing module generates a local dynamic flashing icon sequence for the target instruction sequence, and generates a local dynamic flashing icon display sequence corresponding to the target instruction sequence, which may be performed in the following steps:

[0054] 1. Segment each initial icon corresponding to the target instruction sequence to obtain a local dynamic flashing icon subset sequence corresponding to each initial icon.

[0055] Among them, the target instruction sequence may include one or more initial icons, for example, the target instruction sequence includes initial icon 1, initial icon 2, initial icon 3 and initial icon 4. Each initial icon can be divided into one or more local dynamic flashing icons, and each local dynamic flashing icon is a part of its corresponding initial icon. For example, the initial icon 1 can be divided into 4 local dynamic flashing icons, namely local dynamic flashing icon 1-1, local dynamic flashing icon 1-2, local dynamic flashing icon 1-3 and local dynamic flashing icon 1-4; the initial icon 2 can be divided into 4 local dynamic flashing icons, namely local dynamic flashing icon 2-1, local dynamic flashing icon 2-2, local dynamic flashing icon 2-3 and local dynamic flashing icon 2-4, etc., and this application does not limit this.

[0056] 2. Establish an associated index between each local dynamic flashing icon in the local dynamic flashing icon subset and the corresponding initial icon to obtain an associated index subset.

[0057] For example, based on the example in step 1, the server can establish association indexes for local dynamic flashing icon 1-1, local dynamic flashing icon 1-2, local dynamic flashing icon 1-3 and local dynamic flashing icon 1-4 with initial icon 1, such as establishing an association index for local dynamic flashing icon 1 to obtain local dynamic flashing icon 1-1, and establishing an association index for local dynamic flashing icon 2 to obtain local dynamic flashing icon 1-2, and this application does not limit this. Establishing association indexes for local dynamic flashing icon 2-1, local dynamic flashing icon 2-2, local dynamic flashing icon 2-3 and local dynamic flashing icon 2-4 with initial icon 2, etc., and this application does not limit this.

[0058] 3. Randomly shuffle each associated index in the associated index subset corresponding to each of the initial icons to obtain a random associated index subset corresponding to each of the initial icons.

[0059] Before displaying the instruction sequence to the target user, the server can use a random shuffling algorithm to randomly shuffle the associated indexes in each associated index subset to obtain a random associated index subset. Optionally, the server can also use other methods to randomly shuffle the associated indexes, which is not limited in this application.

[0060] 4. Determine a local dynamic flashing icon display sequence according to the random associated index subset corresponding to each of the initial icons.

[0061] For example, taking the random associated index subset of initial icon 1 as local dynamic flashing icon 1-3, local dynamic flashing icon 1-2, local dynamic flashing icon 1-4 and local dynamic flashing icon 1-1, and the random associated index subset of initial icon 2 as local dynamic flashing icon 2-4, local dynamic flashing icon 2-2, local dynamic flashing icon 2-1 and local dynamic flashing icon 2-3 as an example, the server can determine the corresponding local dynamic flashing icon display sequence, and display the local dynamic flashing icons to the target user in sequence according to the arrangement order of the random associated indexes in each random associated index subset. This application does not impose any restrictions on this.

[0062] It is understandable that after the local dynamic flashing icons of the local dynamic flashing module are sorted, the target user can look at the position corresponding to the initial icon that needs to be focused on according to the guidance (i.e. the randomly lit local dynamic flashing icon). When a certain initial icon needs to be displayed, the corresponding local dynamic flashing icons will be displayed in sequence according to the order of the random association index. For example, the initial icon 1 includes 4 local dynamic flashing icons, and the random association indexes of the 4 local dynamic flashing icons are 1-3, 1-4, 1-2, and 1-1. When the initial icon 1 is displayed to the target user, it will be displayed in the order of local dynamic flashing icon 3, focus group 4, local dynamic flashing icon 2 and local dynamic flashing icon 1, and this application does not limit this.

[0063] Optionally, if all local dynamic flashing icons have appeared once, the local dynamic flashing module can use the random shuffling algorithm to randomly shuffle the associated index again to obtain a re-shuffled random associated index, so as to further achieve the random appearance of the user's focus point guide position. Optionally, if all possible index sequences corresponding to the local dynamic flashing icons have appeared once, the initial icon can be re-segmented and the associated index can be re-established to maximize the diversity of the focus point guide positions.

[0064] In order to address the defect that there is no focus guidance in the traditional icon display process, which leads to the user's unstable gaze position, the present application uses a local dynamic flashing module to sort the target instruction sequence into local dynamic flashing icons, which can effectively focus the target user's focus on the required position. During the stimulation process, the target user's vigilance can be improved, which is conducive to enhancing the performance of the high-frequency cold-start visual evoked potential brain-computer interface system.

[0065] S40: The user display interface displays a local dynamic flashing icon display sequence to the target user.

[0066] After the local dynamic flashing module determines the local dynamic flashing icon display sequence, the user display interface can obtain the local dynamic flashing icon display sequence and flash the local dynamic flashing icon display sequence in sequence according to the display order indicated in the local dynamic flashing icon display sequence.

[0067] For example, Figure 3 As shown, Figure 3 An exemplary user display interface arranged in a 4*3 manner is shown. After considering the equivalent neighbor principle, by Figure 3 The arrangement in can increase the distance between each stimulation target (initial icon of the target instruction sequence) as much as possible. Figure 3 In the example, a target instruction sequence includes 12 initial icons. The 12 icons can be arranged from large to small according to the selected frequency. The viewing angle of each icon can be a square of 6°×6° pixels. Under the premise of a distance of 50 cm from the screen, the viewing angle between rows can be 4.53°, and the viewing angle between columns can be 8°.

[0068] Studies have shown that the higher the contrast between the icon and the background displayed in the user display interface, the higher the visual evoked potential generated. Therefore, in order to increase the amplitude of the electroencephalogram (EEG) generated by the stimulation, the initial icon and the background can be selected in black and white, which is a color with a high contrast, and 0 and 1 in a binary sequence can be used to represent extinguishing or lighting, 0 represents completely extinguishing, and 1 represents completely lighting, and this application does not impose any restrictions on this.

[0069] S50: The decoding module receives the target EEG signal generated by the target user after watching the local dynamic flashing icon display sequence.

[0070] The target EEG signal can be collected by the acquisition system. The acquisition system can transmit the collected target EEG signal to the decoding module, and correspondingly, the decoding module can receive the target EEG signal from the acquisition system. Optionally, the acquisition system involved in this application can be composed of a Boricon EEG acquisition device and a personal computer with an LCD display. Among them, the personal computer can be used to present the stimulation paradigm, as well as to collect and analyze data. The screen refresh rate is 120Hz and the original resolution is 1920×1080, which is not limited in this application.

[0071] The stimulation paradigm of this application can be developed in the Matlab environment, and the monitor and the timing of the EEG acquisition program can be synchronized through a synchronizer. This application adopted a 32-bit electrode layout during the research process, and there are 5 usable electrode channels in the occipital lobe area, namely "PO3", "PO4", "Oz", "O1", and "O2". The sampling rate of the EEG amplifier is 1000Hz. The stimulation paradigm will send a trigger signal to the EEG acquisition program at the beginning of each round of stimulation, and the data of the sixth channel will be stored or sent together with the EEG data for locating the stimulation starting point in the subsequent decoding algorithm process, thereby achieving more accurate decoding and classification.

[0072] S60: The decoding module decodes the target EEG signal to obtain the target intention of the target user.

[0073] The target intention can be used to indicate the intention of the target user determined after the decoding module decodes the target EEG signal. Due to the particularity of the new stimulation paradigm in the selection of stimulation frequency and stimulation form, it is impossible to directly use the existing decoding algorithm for decoding, such as the traditional filter bank design and the traditional template. Therefore, the present application provides a decoding algorithm that can be modified on the basis of the traditional decoding algorithm to make it more compatible with the frequency characteristics and distribution form of the data obtained by the new stimulation paradigm.

[0074] Optionally, the present application can analyze and design multiple filters with different passbands according to the flickering frequency of the stimulation target to perform sub-band decomposition on the received EEG signal; after decomposing each sub-band, perform a correlation value calculation process on each sub-band to obtain its correlation value with each harmonic frequency reference signal; thereby further taking the weighted square of the correlation value of all sub-band components as the correlation eigenvalue of the stimulation target; and then outputting the stimulation target with the largest correlation eigenvalue in all sub-bands as the final output result (i.e., target intention), which is not restricted by the present application.

[0075] The present application can redesign the filter group subband division and reference signal weight calculation based on the unsupervised algorithm of filter group canonical correlation analysis, so as to obtain the calibration-time-free visual evoked potential brain-computer interface system provided by the present application, which can not only achieve cold start, but also improve the comfort of the user during use while maintaining the performance of the traditional paradigm.

[0076] It should be noted that, in the case of multiple initial instruction sequences, the instruction sequence generation module can determine a more appropriate target instruction sequence based on each initial instruction sequence, and the local dynamic flashing module can perform a local dynamic flashing icon sequence on each target instruction sequence, thereby displaying the local dynamic flashing icon display sequence corresponding to each target instruction sequence on the user display interface, so that the target user can generate a corresponding target EEG signal, and then decode each target EEG signal through the decoding module to further determine which local dynamic flashing icon corresponds to the target instruction sequence that the target user is currently watching, and then determine the target user's accurate target intention. This application does not impose any restrictions on this.

[0077] In this example, the instruction sequence generation module determines the integer instruction sequence length that makes the flashing frequency within the first preset range according to the instruction target number and the device refresh rate, and determines the target instruction sequence according to the integer instruction sequence length. The local dynamic flashing module sorts the target instruction sequence by local dynamic flashing icons and determines the local dynamic flashing icon display sequence. The user display interface displays the local dynamic flashing icon display sequence to the target user. The decoding module receives the target EEG signal generated by the target user after watching the local dynamic flashing icon display sequence, and decodes the target EEG signal to obtain the target user's target intention. While improving the effectiveness of intention recognition, it also improves the comfort of the target user during the use process, which is conducive to improving the practicality of the visual evoked potential brain-computer interface system.

[0078] In one implementation, the instruction sequence generation module determines the target instruction sequence according to the integer instruction sequence length, which may include the following steps:

[0079] A1: The instruction sequence generation module determines the set of pseudo-random sequences to be traversed according to the integer instruction sequence length;

[0080] A2: The instruction sequence generation module performs EEG maximum response evaluation on each pseudo-random sequence to be traversed in the set of pseudo-random sequences to be traversed, and obtains an EEG evaluation signal set;

[0081] A3: The instruction sequence generation module calculates the signal-to-noise ratio of each EEG evaluation signal in the EEG evaluation signal set to obtain a signal-to-noise ratio set;

[0082] A4: The instruction sequence generation module obtains the maximum signal-to-noise ratio from the signal-to-noise ratio set;

[0083] A5: The instruction sequence generation module determines the decimal number corresponding to the maximum signal-to-noise ratio as the optimal flicker frequency value;

[0084] A6: The instruction sequence generation module determines the optimal pseudo-random sequence according to the optimal flashing frequency value;

[0085] A7: The instruction sequence generation module performs periodic repetition processing on the optimal pseudo-random sequence to obtain a target instruction sequence.

[0086] Among them, the set of pseudo-random sequences to be traversed may include one or more pseudo-random sequences to be traversed, and the pseudo-random sequences to be traversed may be used to indicate pseudo-random sequences that need to be traversed and have a length of an integer instruction sequence. It is understandable that the pseudo-random sequence to be traversed may be understood as a PRBS sequence to be selected, and this application does not limit this.

[0087] For example, taking the integer instruction sequence length as N1, in order to compare all potential pseudo-random sequences of length N1, it is necessary to set the value of the decimal number D to be equal to 1 to All pseudo-random sequences corresponding to (excluding binary sequences of all 1s and all 0s) are compared. After the decimal number D is determined, the decimal number D can be converted into a binary pseudo-random sequence to obtain a set of pseudo-random sequences to be traversed. Optionally, the process of converting the decimal number D into a binary pseudo-random sequence can be referred to the following formula:

[0088]

[0089] Among them, d n It can be used to represent the value of the nth bit in a binary pseudo-random sequence. The nth number from right to left in the binary pseudo-random sequence is the nth bit of the sequence. For example, if the first number from right to left is 1, then the first bit of the binary pseudo-random sequence is 1, that is, d n =1; D can be used to represent a decimal number that needs to be converted; mod can be used to represent the operation of taking the remainder after a division operation; Dmod2 n Can be used to represent dividing D by 2 n Then take the remainder; It can be used to indicate that the number in the brackets is rounded down.

[0090] For example, taking D as 5, the first bit of the pseudo-random sequence to be traversed is calculated, and the remainder 1 is obtained after 5 is divided by 2, and the remainder 1 is divided by 1 to obtain 1, and 1 is rounded down to 1, so d1 = 1. Then d2 = 0, d3 = 1, d4 = 0 are calculated, and the pseudo-random sequence to be traversed is 1010, which is not limited in this application.

[0091] The EEG evaluation signal set may include one or more EEG evaluation signals, and the EEG evaluation signal may be understood as an EEG signal obtained after performing EEG maximum response evaluation on each pseudo-random sequence to be traversed in the pseudo-random sequence set to be traversed.

[0092] Optionally, the server may perform periodic repetition processing on each pseudo-random sequence to be traversed to obtain a complete set of instruction sequences. For example, the instruction sequence generation module may perform periodic repetition filling on the first pseudo-random sequence to be traversed, such as expanding the length of the first pseudo-random sequence to be traversed to 125. Optionally, if it is not enough to repeat completely, the sequence may be periodically repeated from high to low until the length of the first pseudo-random sequence to be traversed reaches the target length (such as 125), and this application does not limit this.

[0093] The first pseudo-random sequence to be traversed may be any pseudo-random sequence to be traversed in the set of pseudo-random sequences to be traversed. It is understandable that the set of pseudo-random sequences to be traversed may also include other pseudo-random sequences to be traversed, such as the second pseudo-random sequence to be traversed, the third pseudo-random sequence to be traversed, etc. The instruction sequence generation module may use the above steps to periodically repeat and fill other pseudo-random sequences to be traversed, thereby obtaining a complete instruction sequence set.

[0094] It can be understood that the EEG evaluation signal can be understood as the EEG signal generated by the target user for each complete instruction sequence in the complete instruction sequence set after each complete instruction sequence is displayed to the target user. That is to say, after the instruction sequence generation model obtains the complete instruction sequence corresponding to each pseudo-random sequence to be traversed, each complete instruction sequence can be displayed to the target user, thereby collecting the EEG evaluation signal generated by the target user after watching each complete instruction sequence, and then further analyzing the EEG evaluation signal to select the pseudo-random sequence with the best frequency domain characteristics of the EEG evaluation signal.

[0095] The signal-to-noise ratio set may include one or more signal-to-noise ratios, which can be understood as the signal-to-noise ratio obtained after calculating the signal-to-noise ratio of each EEG evaluation signal. The specific calculation process can be found in the detailed description of the embodiment below, and this application will not repeat it here.

[0096] The maximum signal-to-noise ratio can be used to indicate the maximum value in the signal-to-noise ratio set. The optimal flicker frequency value can be used to indicate the flicker frequency value corresponding to the maximum signal-to-noise ratio. The optimal pseudo-random sequence can be used to indicate the pseudo-random sequence corresponding to the maximum signal-to-noise ratio. Optionally, the instruction sequence generation module can further specify the optimal pseudo-random sequence as an instruction code, so that the instruction code can be repeated periodically to obtain a target instruction sequence, which is not limited in this application. Exemplarily, with an integer instruction sequence length of l i The instruction sequence generation module can divide the device refresh rate by the integer instruction sequence length (i.e., f r / l i) is processed periodically to achieve the purpose of superimposing high-frequency flicker on the low-frequency periodic signal.

[0097] It can be understood that the maximum signal-to-noise ratio can be the signal-to-noise ratio that is most advantageous and best meets specific requirements under the current given conditions, such as the current device refresh rate, the current flicker frequency and the current integer instruction sequence length. The pseudo-random sequence corresponding to the maximum signal-to-noise ratio (i.e., the optimal pseudo-random sequence) can be the most effective sequence for inducing specific EEG signals (related to the flicker frequency) at the current integer instruction sequence length and the best sequence at the current device refresh rate. This application does not impose any restrictions on this.

[0098] The instruction sequence generation module can determine the decimal number corresponding to the maximum signal-to-noise ratio as the optimal flicker frequency value, so as to determine the optimal pseudo-random sequence according to the optimal flicker frequency value, that is, the selected optimal PRBS sequence. Optionally, for initial instruction sequences with different integer instruction sequence lengths, the instruction sequence generation module can respectively determine the maximum signal-to-noise ratio that best matches and conforms to the current device refresh rate under each integer instruction sequence length, that is, obtain the optimal pseudo-random sequence corresponding to each integer instruction sequence length, so that each optimal pseudo-random sequence can be further periodically repeated and superimposed according to the total length required by each target instruction sequence to obtain each target instruction sequence.

[0099] Optionally, the instruction sequence generation module can combine the optimal flicker frequency value with a specific visual stimulation pattern generation algorithm to obtain a more appropriate and more satisfactory target instruction sequence. For example, the instruction sequence generation module can determine the flicker frequency of the visual stimulation, the icon change cycle, etc. according to the optimal flicker frequency value, so as to generate a more satisfactory target instruction sequence according to these parameters, and the present application does not limit this.

[0100] In this example, the instruction sequence generation module determines the set of pseudo-random sequences to be traversed according to the length of the integer instruction sequence, so as to perform EEG maximum response evaluation on each pseudo-random sequence to be traversed in the set of pseudo-random sequences to be traversed, and obtain an EEG evaluation signal set, thereby performing signal-to-noise ratio calculation on each EEG evaluation signal in the set of EEG evaluation signals to obtain a signal-to-noise ratio set, so as to obtain the maximum signal-to-noise ratio from the signal-to-noise ratio set, and further determine the decimal number corresponding to the maximum signal-to-noise ratio as the optimal flickering frequency value, so as to determine the optimal pseudo-random sequence according to the optimal flickering frequency value, and then perform periodic repetition processing on the optimal pseudo-random sequence to obtain a target instruction sequence that better meets the requirements, which is conducive to more effectively inducing the desired EEG signals and achieving the expected brain-computer interaction effect.

[0101] In one implementation, taking the case where an EEG maximum response evaluation is performed on a first pseudo-random sequence to be traversed in a set of pseudo-random sequences to be traversed to obtain a first EEG evaluation signal as an example, the instruction sequence generation module calculates the signal-to-noise ratio of the first EEG evaluation signal to obtain the signal-to-noise ratio corresponding to the first EEG evaluation signal, which may include the following steps:

[0102] B1: the instruction sequence generation module obtains a first EEG evaluation signal from the EEG evaluation signal set;

[0103] B2: The instruction sequence generation module performs discrete Fourier transform processing on the first EEG evaluation signal to obtain a target spectrum;

[0104] B3: The instruction sequence generation module determines the signal-to-noise ratio corresponding to the first EEG evaluation signal according to the amplitude of the target spectrum, the sampling rate of the EEG device, the flickering frequency and the length of the first EEG evaluation signal.

[0105] Among them, the first EEG evaluation signal can be the EEG evaluation signal obtained after the EEG maximum response evaluation is performed on the first pseudo-random sequence to be traversed in the set of pseudo-random sequences to be traversed. It can be understood that the instruction sequence generation module can perform EEG maximum response evaluation on other pseudo-random sequences to be traversed, such as the second pseudo-random sequence to be traversed, the third pseudo-random sequence to be traversed, etc., and other EEG evaluation signals can be obtained, such as the second EEG evaluation signal, the third EEG evaluation signal, etc., and this application does not limit this.

[0106] The target spectrum is the spectrum obtained after the first EEG evaluation signal is subjected to discrete Fourier transform processing. It can be understood that the instruction sequence generation module can realize the conversion of the first EEG evaluation signal from the time domain to the frequency domain by performing discrete Fourier transform on the first EEG evaluation signal. In other words, the target spectrum can be the frequency domain expression corresponding to the first EEG evaluation signal.

[0107] Optionally, the instruction sequence generation module performs discrete Fourier transform processing on the first EEG evaluation signal to obtain a target spectrum, which can be seen in the following formula:

[0108]

[0109] Among them, X k can be used to represent the target spectrum, k can be used to represent the frequency index in the frequency domain; N1 can be used to represent the integer instruction sequence length of the first instruction sequence; n can be used to represent the index of the integer instruction sequence length; x n can be used to represent the first EEG evaluation signal; e can be used to represent a natural constant; and j can be used to represent an imaginary unit.

[0110] Through the above formula, we can clearly see the relative strength or proportion of different frequency components in the first EEG evaluation signal. For example, if the amplitude of a certain frequency in the frequency domain is large, it means that the frequency is relatively important in the first EEG evaluation signal and contributes more to the sequence; conversely, if the amplitude of a certain frequency is small, its proportion in the first EEG evaluation signal is relatively small. When considering factors such as flicker frequency, frequency domain information is very critical. By converting the sequence to the frequency domain, we can more directly observe the relationship between the sequence and the flicker frequency, as well as the signal-to-noise ratio distribution at different frequencies, which is conducive to better determining the maximum signal-to-noise ratio in subsequent steps.

[0111] Optionally, the instruction sequence generation module determines the signal-to-noise ratio corresponding to the first EEG evaluation signal according to the amplitude of the target spectrum, the sampling rate of the EEG device, the flicker frequency and the length of the first EEG evaluation signal, as shown in the following formula:

[0112]

[0113] Among them, SNR1 can be used to represent the signal-to-noise ratio corresponding to the first EEG evaluation signal; log 10 It can be used to represent logarithmic operations with base 10; y(f) can be used to represent the amplitude of the flicker frequency in the target spectrum; f can be used to represent the flicker frequency, which can be the basic flicker frequency, that is, the fundamental frequency; k can be used to represent the index of the frequency point; f s It can be used to indicate the sampling rate of the EEG device; N2 can be used to indicate the length of the first EEG evaluation signal.

[0114] Optionally, the instruction sequence generation module obtains a second EEG evaluation signal from the EEG evaluation signal set, and performs discrete Fourier transform processing on the second EEG evaluation signal to obtain a second spectrum, so as to determine a second signal-to-noise ratio corresponding to the second EEG evaluation signal according to the second amplitude of the second spectrum, the sampling rate of the EEG device, the flickering frequency, and the length of the second EEG evaluation signal. It can be understood that the instruction sequence generation module can determine the signal-to-noise ratio corresponding to each EEG evaluation signal by calculating the signal-to-noise ratio of each EEG evaluation signal to obtain a signal-to-noise ratio set.

[0115] It should be noted that the instruction sequence generation module can perform EEG maximum response evaluation on the fundamental frequency and the multiple frequency (such as the first multiple frequency, the second multiple frequency, etc.) corresponding to the first pseudo-random sequence to be traversed in the set of pseudo-random sequences to be traversed, so as to obtain the signal-to-noise ratio corresponding to the fundamental frequency (that is, the signal-to-noise ratio corresponding to the above-mentioned first EEG evaluation signal) and the signal-to-noise ratio corresponding to the multiple frequency, and then further perform a weighted sum operation on the signal-to-noise ratio corresponding to the fundamental frequency and the signal-to-noise ratio corresponding to the multiple frequency to obtain a fused signal-to-noise ratio that integrates the features at the fundamental frequency and the features at the multiple frequency, and then compare the fused signal-to-noise ratio with the fused signal-to-noise ratio of the second pseudo-random sequence to be traversed, the fused signal-to-noise ratio of the third pseudo-random sequence to be traversed, etc., to obtain the maximum fused signal-to-noise ratio, so as to find the optimal pseudo-random sequence from the set of pseudo-random sequences to be traversed, and determine a target instruction sequence that better meets the requirements. Among them, the instruction sequence generation module can adopt the existing weighted summation method to perform a weighted summation operation on the signal-to-noise ratio corresponding to the base frequency and the signal-to-noise ratio corresponding to the multiplier, such as using filter bank canonical correlation analysis (Filter Bank CanonicalCorrelation Analysis, FBCCA) to perform a weighted summation operation, and this application does not impose any restrictions on this.

[0116] In this example, the instruction sequence generation module obtains the first EEG evaluation signal from the EEG evaluation signal set, and performs discrete Fourier transform processing on the first EEG evaluation signal to obtain the target spectrum, so as to further determine the signal-to-noise ratio corresponding to the first EEG evaluation signal according to the amplitude of the flickering frequency in the target spectrum, the sampling rate of the EEG device, the flickering frequency and the length of the first EEG evaluation signal, which is conducive to comparing and screening the signal-to-noise ratios of EEG evaluation signals corresponding to other pseudo-random sequences to be traversed, so as to find the optimal pseudo-random sequence from the set of pseudo-random sequences to be traversed, and to determine a target instruction sequence that better meets the requirements.

[0117] In one implementation, the decoding module decodes the target EEG signal to obtain the target intention of the target user, which may include the following steps:

[0118] C1: The decoding module filters the target EEG signal through a filter bank to obtain a set of sub-band EEG signals;

[0119] C2: The decoding module calculates the correlation value between each sub-band EEG signal in the sub-band EEG signal set and the reference signal corresponding to each reference intent in the reference intent set to obtain a correlation value vector set;

[0120] C3: The decoding module performs weighted summation processing on the correlation value vectors corresponding to each reference intent in the correlation value vector set to obtain a correlation feature value set;

[0121] C4: The decoding module obtains the maximum relevant eigenvalue from the relevant eigenvalue set;

[0122] C5: The decoding module takes the reference intention corresponding to the maximum relevant eigenvalue as the target intention of the target user.

[0123] Among them, the sub-band EEG signal set may include one or more sub-band EEG signals, which can be used to indicate the EEG signal obtained after filtering the target EEG signal through the filter group. Each sub-band EEG signal retains the EEG signal within a specific frequency range, which can more carefully analyze the relationship between different frequency components and user intentions, and help to more accurately identify the user's target intention in the subsequent intention recognition process.

[0124] The correlation value vector set may include one or more correlation value vectors, which may be used to indicate the correlation between the reference signal corresponding to the reference intention and the sub-band EEG signal. The reference intention set may include one or more reference intentions, which may be used to indicate the intention corresponding to the stimulation target (i.e., the display icon of the target instruction sequence).

[0125] Optionally, each display icon in the target instruction sequence can flash at different frequencies, and the acquired target EEG signal can also have a corresponding flashing frequency. Therefore, the decoding module can display the sine function of the frequency corresponding to the icon as a reference signal corresponding to the reference intention. Optionally, the process of determining the reference signal by the decoding module can refer to the following formula:

[0126]

[0127] in, Can be used to represent the reference signal; f k It can be used to represent a specific frequency in the frequency domain; k can be used to represent the frequency index in the frequency domain; m can be used to represent the sub-band index in the sub-band EEG signal set, which corresponds to the multiple of the flicker frequency, such as the first sub-band corresponds to the single multiple of the flicker frequency, the second sub-band corresponds to the double multiple of the flicker frequency, and so on; t can be used to represent time.

[0128] It is understandable that when the decoding module generates a reference signal, it needs to generate a reference signal with the same length as the target EEG signal in the time domain to facilitate subsequent correlation value calculations. Therefore, the decoding module can generate a reference signal with a specific frequency (such as a flicker frequency) in the time domain according to the above formula.

[0129] Optionally, the decoding module calculates the correlation value of each sub-band EEG signal in the sub-band EEG signal set and the reference signal corresponding to each reference intent in the reference intent set to obtain a correlation value vector set, which can be seen in the following formula:

[0130]

[0131] in, can be used to represent a correlation value vector in a correlation value vector set, which can be used to represent the correlation value between the first sub-band EEG signal and the one-fold frequency of the reference signal corresponding to the k-th reference intention; k can be used to represent the index of the reference signal, which corresponds to the frequency index in the frequency domain; ρ(x,y) can be used to represent the operation of performing a typical correlation analysis on x and y, that is, and Perform canonical correlation analysis operations; It can be used to represent the first sub-band EEG signal; Can be used to represent the transpose of the first sub-band EEG signal; W X Can be used to represent the weight vector for the input signal; It can be used to represent the frequency doubling of the reference signal corresponding to the k-th reference intention; It can be used to represent the transpose of the reference signal corresponding to the kth reference intention at one frequency; W Y Can be used to represent the weight vector for the output signal; can be used to represent a correlation value vector in a correlation value vector set, and the correlation value vector can be used to represent the correlation value between the first sub-band EEG signal and the double frequency of the reference signal corresponding to the kth reference intention; It can be used to represent the second sub-band EEG signal; It can be used to represent the transpose of the second sub-band EEG signal; It can be used to represent the double frequency of the reference signal corresponding to the kth reference intention; It can be used to represent the transpose of the double frequency of the reference signal corresponding to the kth reference intention; can be used to represent a correlation value vector in a correlation value vector set, and the correlation value vector can be used to represent the correlation value between the Z-th sub-band EEG signal and the Z-frequency multiple of the reference signal corresponding to the k-th reference intention; Z can be used to represent the index of the sub-band EEG signal; It can be used to represent the Zth sub-band EEG signal; It can be used to represent the transpose of the Zth sub-band EEG signal; It can be used to represent the Z frequency multiple of the reference signal corresponding to the k-th reference intention; It can be used to represent the Z-fold transpose of the reference signal corresponding to the k-th reference intent.

[0132] The relevant feature value set may include one or more relevant feature values, which may be used to indicate the relevant feature value obtained after weighted summing the relevant value vectors in the relevant value vector set. Optionally, the decoding module performs weighted summing on the relevant value vectors corresponding to each reference intent in the relevant value vector set to obtain the relevant feature value set, as shown in the following formula:

[0133]

[0134] w(z)=z -a +b,z∈[1,Z]

[0135] in, can be used to represent a relevant eigenvalue in the relevant eigenvalue set, and the relevant eigenvalue can be used to represent the relevant eigenvalue obtained after weighted summation of the relevant value vectors corresponding to the kth reference intention in the relevant value vector set; Z can be used to represent the index of the sub-band EEG signal; w(z) can be used to represent the operation of calculating the weight component of the zth sub-band EEG signal; can be used to represent the correlation value between the z-th sub-band EEG signal and the reference signal corresponding to the k-th reference intention in the correlation value vector set; a and b can be used to represent constants that maximize the classification performance.

[0136] By performing weighted summation processing on the correlation value vectors in the correlation value vector set, the correlation between multiple sub-band EEG signals and a reference intention can be comprehensively considered to obtain a correlation feature value that more comprehensively reflects the correlation between the reference intention and the current EEG signal.

[0137] Furthermore, the decoding module obtains the maximum relevant eigenvalue from the relevant eigenvalue set, and uses the reference intention corresponding to the maximum relevant eigenvalue as the target intention of the target user. It can be understood that the maximum relevant eigenvalue can represent the intention eigenvalue with the strongest correlation with the current EEG signal among all reference intentions. At this time, the reference intention corresponding to the maximum relevant eigenvalue can be used as the target intention of the target user, that is, the decoding module decodes the user's target intention.

[0138] In this example, the decoding module filters the target EEG signal through a filter group to obtain a sub-band EEG signal set, and calculates the correlation value of each sub-band EEG signal in the sub-band EEG signal set with the reference signal corresponding to each reference intention in the reference intention set to obtain a correlation value vector set, and further performs weighted summation on the correlation value vectors corresponding to each reference intention in the correlation value vector set to obtain a correlation eigenvalue set, so that the maximum correlation eigenvalue can be obtained from the correlation eigenvalue set, and then the reference intention corresponding to the maximum correlation eigenvalue can be used as the target intention of the target user, which improves the effectiveness of the target intention recognition process and is conducive to improving the practicality of the visual evoked potential brain-computer interface system.

[0139] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0140] In one embodiment, a high-frequency cold-start visual evoked potential brain-computer interface device is provided, and the high-frequency cold-start visual evoked potential brain-computer interface device corresponds one-to-one with the high-frequency cold-start visual evoked potential brain-computer interface system in the above embodiment. Figure 4 As shown, the high-frequency cold-start visual evoked potential brain-computer interface device includes a first determination module 101, a second determination module 102, a third determination module 103, a display module 104, a receiving module 105 and a decoding module 106. The functional modules are described in detail as follows:

[0141] The first determination module 101 is used to determine the integer instruction sequence length according to the instruction target number and the device refresh rate, so that the flickering frequency under the integer instruction sequence length is within a first preset range; wherein the flickering frequency is the quotient of the device refresh rate and the integer instruction sequence length;

[0142] A second determination module 102 is used to determine a target instruction sequence according to the length of the integer instruction sequence;

[0143] The third determination module 103 is used to sort the local dynamic flashing icons of the target instruction sequence and determine the local dynamic flashing icon display sequence;

[0144] Display module 104, used to display a local dynamic flashing icon display sequence to a target user;

[0145] The receiving module 105 is used to receive the target EEG signal generated by the target user after watching the local dynamic flashing icon display sequence;

[0146] The decoding module 106 is used to decode the target EEG signal to obtain the target intention of the target user.

[0147] In one embodiment, the second determining module 102 is specifically configured to:

[0148] Determine a set of pseudo-random sequences to be traversed according to the length of the integer instruction sequence;

[0149] Performing an EEG maximum response evaluation on each pseudo-random sequence to be traversed in the set of pseudo-random sequences to be traversed, to obtain an EEG evaluation signal set;

[0150] Calculating the signal-to-noise ratio of each electroencephalogram evaluation signal in the electroencephalogram evaluation signal set to obtain a signal-to-noise ratio set;

[0151] Get the maximum signal-to-noise ratio from the signal-to-noise ratio set;

[0152] The decimal number corresponding to the maximum signal-to-noise ratio is determined as the optimal flicker frequency value;

[0153] According to the optimal flickering frequency value, an optimal pseudo-random sequence is determined;

[0154] The optimal pseudo-random sequence is subjected to periodic repetition processing to obtain a target instruction sequence.

[0155] In one embodiment, the second determining module 102 is specifically configured to:

[0156] Acquire a first EEG evaluation signal from the EEG evaluation signal set;

[0157] Performing discrete Fourier transform processing on the first EEG evaluation signal to obtain a target frequency spectrum;

[0158] The signal-to-noise ratio corresponding to the first EEG evaluation signal is determined according to the amplitude of the flicker frequency in the target spectrum, the sampling rate of the EEG device, the flicker frequency and the length of the first EEG evaluation signal.

[0159] In one embodiment, the third determining module 103 is specifically configured to:

[0160] Segment each initial icon corresponding to the target instruction sequence to obtain a local dynamic flashing icon subset corresponding to each initial icon;

[0161] Establishing an associated index between each local dynamic flashing icon in the local dynamic flashing icon subset and the corresponding initial icon to obtain an associated index subset;

[0162] Randomly shuffle each associated index in the associated index subset corresponding to each initial icon to obtain a random associated index subset corresponding to each initial icon;

[0163] According to the random associated index subset corresponding to each initial icon, a local dynamic flashing icon display sequence is determined.

[0164] In one embodiment, the decoding module 106 is specifically configured to:

[0165] The target EEG signal is filtered through a filter bank to obtain a sub-band EEG signal set;

[0166] Calculate the correlation value of each sub-band EEG signal in the sub-band EEG signal set and the reference signal corresponding to each reference intention in the reference intention set to obtain a correlation value vector set;

[0167] Performing weighted sum processing on the correlation value vector corresponding to each reference intention in the correlation value vector set to obtain a correlation feature value set;

[0168] Obtain the maximum relevant eigenvalue from the relevant eigenvalue set;

[0169] The reference intention corresponding to the maximum relevant eigenvalue is taken as the target intention of the target user.

[0170] The present invention provides a high-frequency cold-start visual evoked potential brain-computer interface device. According to the number of instruction targets and the device refresh rate, the length of an integer instruction sequence that makes the flashing frequency within a first preset range is determined, and according to the integer instruction sequence length, a target instruction sequence is determined, so as to further sort the target instruction sequence into local dynamic flashing icons, determine a local dynamic flashing icon display sequence, and display the local dynamic flashing icon display sequence to a target user, so as to receive a target electroencephalogram signal generated by the target user after watching the local dynamic flashing icon display sequence, and then decode the target electroencephalogram signal to obtain the target user's target intention. While improving the effectiveness of intention recognition, the comfort of the target user during use is improved, which is conducive to improving the practicality of the visual evoked potential brain-computer interface system.

[0171] For the specific limitations of the high-frequency cold-start visual evoked potential brain-computer interface device, please refer to the limitations of the high-frequency cold-start visual evoked potential brain-computer interface system above, which will not be repeated here. Each module in the above-mentioned high-frequency cold-start visual evoked potential brain-computer interface device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0172] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 5As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client through a network connection. When the computer program is executed by the processor, it realizes a function or step on the server side of a high-frequency cold-start visual evoked potential brain-computer interface system.

[0173] In one embodiment, a computer device is provided. The computer device may be a client, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server via a network connection. When the computer program is executed by the processor, it realizes a function or step on the client side of a high-frequency cold-start visual evoked potential brain-computer interface system.

[0174] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0175] Determine the integer instruction sequence length according to the target number of instructions and the device refresh rate, so that the flickering frequency under the integer instruction sequence length is within a first preset range; wherein the flickering frequency is the quotient of the device refresh rate and the integer instruction sequence length;

[0176] Determine a target instruction sequence according to the length of the integer instruction sequence;

[0177] Sorting the local dynamic flashing icons of the target instruction sequence to determine the local dynamic flashing icon display sequence;

[0178] Display a partial dynamic flashing icon display sequence to the target users;

[0179] receiving a target electroencephalogram signal generated by a target user after viewing a local dynamic flashing icon display sequence;

[0180] Decode the target EEG signal to obtain the target user's target intention.

[0181] The present invention provides a computer device, which determines the length of an integer instruction sequence that makes a flashing frequency within a first preset range according to the target number of instructions and the device refresh rate, and determines the target instruction sequence according to the integer instruction sequence length, so as to further sort the target instruction sequence into local dynamic flashing icons, determine the local dynamic flashing icon display sequence, and display the local dynamic flashing icon display sequence to a target user, so as to receive the target electroencephalogram signal generated by the target user after watching the local dynamic flashing icon display sequence, and then decode the target electroencephalogram signal to obtain the target user's target intention, thereby improving the effectiveness of intention recognition and the comfort of the target user during use, which is conducive to improving the practicality of the visual evoked potential brain-computer interface system.

[0182] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0183] Determine the integer instruction sequence length according to the target number of instructions and the device refresh rate, so that the flickering frequency under the integer instruction sequence length is within a first preset range; wherein the flickering frequency is the quotient of the device refresh rate and the integer instruction sequence length;

[0184] Determine a target instruction sequence according to the length of the integer instruction sequence;

[0185] Sorting the local dynamic flashing icons of the target instruction sequence to determine the local dynamic flashing icon display sequence;

[0186] Display a partial dynamic flashing icon display sequence to the target users;

[0187] receiving a target electroencephalogram signal generated by a target user after viewing a local dynamic flashing icon display sequence;

[0188] Decode the target EEG signal to obtain the target user's target intention.

[0189] The present invention provides a computer-readable storage medium, which determines the length of an integer instruction sequence that makes the flashing frequency within a first preset range according to the target number of instructions and the refresh rate of a device, and determines the target instruction sequence according to the length of the integer instruction sequence, so as to further sort the target instruction sequence into local dynamic flashing icons, determine the local dynamic flashing icon display sequence, and display the local dynamic flashing icon display sequence to a target user, so as to receive the target electroencephalogram signal generated by the target user after watching the local dynamic flashing icon display sequence, and then decode the target electroencephalogram signal to obtain the target user's target intention, thereby improving the effectiveness of intention recognition and the comfort of the target user during use, which is conducive to improving the practicality of the visual evoked potential brain-computer interface system.

[0190] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can refer to the relevant descriptions on the server side and the client side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0191] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0192] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0193] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A high-frequency cold-start visual evoked potential brain-computer interface system, characterized in that: The high-frequency cold-start visual evoked potential brain-computer interface system includes an instruction sequence generation module, a local dynamic flickering module, a user display interface and a decoding module, wherein: The instruction sequence generation module determines the integer instruction sequence length according to the instruction target number and the device refresh rate, so that the flickering frequency under the integer instruction sequence length is within a first preset range; wherein the flickering frequency is the quotient of the device refresh rate and the integer instruction sequence length; The instruction sequence generation module determines a target instruction sequence according to the integer instruction sequence length; The local dynamic flashing module sorts the target instruction sequence by local dynamic flashing icons to determine a local dynamic flashing icon display sequence; The user display interface displays the local dynamic flashing icon display sequence to the target user; The decoding module receives a target electroencephalogram signal generated by the target user after viewing the local dynamic flashing icon display sequence; The decoding module decodes the target EEG signal to obtain the target intention of the target user.

2. The high-frequency cold-start visual evoked potential brain-computer interface system according to claim 1, characterized in that: in, The instruction sequence generation module determines a set of pseudo-random sequences to be traversed according to the integer instruction sequence length; The instruction sequence generation module performs an EEG maximum response evaluation on each pseudo-random sequence to be traversed in the set of pseudo-random sequences to be traversed to obtain an EEG evaluation signal set; The instruction sequence generation module calculates the signal-to-noise ratio of each EEG evaluation signal in the EEG evaluation signal set to obtain a signal-to-noise ratio set; The instruction sequence generation module obtains a maximum signal-to-noise ratio from the signal-to-noise ratio set; The instruction sequence generation module determines the decimal number corresponding to the maximum signal-to-noise ratio as the optimal flicker frequency value; The instruction sequence generation module determines an optimal pseudo-random sequence according to the optimal flickering frequency value; The instruction sequence generation module performs periodic repetition processing on the optimal pseudo-random sequence to obtain a target instruction sequence.

3. The high-frequency cold-start visual evoked potential brain-computer interface system according to claim 2, characterized in that: in, The instruction sequence generation module obtains a first EEG evaluation signal from the EEG evaluation signal set; The instruction sequence generation module performs discrete Fourier transform processing on the first EEG evaluation signal to obtain a target spectrum; The instruction sequence generation module determines the signal-to-noise ratio corresponding to the first EEG evaluation signal according to the amplitude of the flicker frequency in the target spectrum, the sampling rate of the EEG device, the flicker frequency and the length of the first EEG evaluation signal.

4. The high-frequency cold-start visual evoked potential brain-computer interface system according to claim 3, characterized in that: in, The local dynamic flashing module performs segmentation processing on each initial icon corresponding to the target instruction sequence to obtain a local dynamic flashing icon subset corresponding to each initial icon; The local dynamic flashing module establishes an association index between each local dynamic flashing icon in the local dynamic flashing icon subset and the corresponding initial icon to obtain an association index subset; The local dynamic flashing module randomly shuffles the order of each associated index in the associated index subset corresponding to each of the initial icons to obtain a random associated index subset corresponding to each of the initial icons; The local dynamic flashing module determines a local dynamic flashing icon display sequence according to the random associated index subset corresponding to each of the initial icons.

5. The high-frequency cold-start visual evoked potential brain-computer interface system according to claim 4, characterized in that: in, The decoding module filters the target EEG signal through a filter bank to obtain a sub-band EEG signal set; The decoding module calculates the correlation value between each sub-band EEG signal in the sub-band EEG signal set and a reference signal corresponding to each reference intent in the reference intent set to obtain a correlation value vector set; The decoding module performs weighted sum processing on the correlation value vector corresponding to each reference intent in the correlation value vector set to obtain a correlation feature value set; The decoding module obtains a maximum relevant eigenvalue from the relevant eigenvalue set; The decoding module uses the reference intention corresponding to the maximum relevant eigenvalue as the target intention of the target user.

6. A high-frequency cold-start visual evoked potential brain-computer interface device, characterized in that: The high-frequency cold-start visual evoked potential brain-computer interface device comprises: A first determination module is used to determine the integer instruction sequence length according to the instruction target number and the device refresh rate, so that the flickering frequency under the integer instruction sequence length is within a first preset range; wherein the flickering frequency is the quotient of the device refresh rate and the integer instruction sequence length; A second determination module, configured to determine a target instruction sequence according to the length of the integer instruction sequence; A third determination module is used to sort the local dynamic flashing icons of the target instruction sequence and determine a local dynamic flashing icon display sequence; A display module, used to display the local dynamic flashing icon display sequence to a target user; A receiving module, used for receiving a target electroencephalogram signal generated by the target user after watching the local dynamic flashing icon display sequence; The decoding module is used to decode the target EEG signal to obtain the target intention of the target user.

7. The high-frequency cold-start visual evoked potential brain-computer interface device according to claim 6, characterized in that: The second determination module is specifically used for: Determining a set of pseudo-random sequences to be traversed according to the length of the integer instruction sequence; Performing an EEG maximum response evaluation on each pseudo-random sequence to be traversed in the set of pseudo-random sequences to be traversed to obtain an EEG evaluation signal set; Calculating the signal-to-noise ratio of each electroencephalogram evaluation signal in the electroencephalogram evaluation signal set to obtain a signal-to-noise ratio set; Acquire a maximum signal-to-noise ratio from the signal-to-noise ratio set; Determine the decimal number corresponding to the maximum signal-to-noise ratio as the optimal flicker frequency value; Determining an optimal pseudo-random sequence according to the optimal flickering frequency value; The optimal pseudo-random sequence is subjected to a sequence periodic repetition process to obtain a target instruction sequence.

8. The high-frequency cold-start visual evoked potential brain-computer interface device according to claim 7, characterized in that: The second determination module is specifically used for: Acquire a first EEG evaluation signal from the EEG evaluation signal set; Performing discrete Fourier transform processing on the first EEG evaluation signal to obtain a target frequency spectrum; The signal-to-noise ratio corresponding to the first EEG evaluation signal is determined according to the amplitude of the flicker frequency in the target spectrum, the sampling rate of the EEG device, the flicker frequency and the length of the first EEG evaluation signal.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method based on the high-frequency cold-start visual evoked potential brain-computer interface system as described in any one of claims 1 to 5 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method of the high-frequency cold-start visual evoked potential brain-computer interface system according to any one of claims 1 to 5 is implemented.

Images