An interaction method, system and storage medium for brain-computer interface
By collecting and analyzing EEG signals, especially alpha waves, beta waves, and P300 waves, an interest judgment index and amplitude analysis are generated, solving the problem of accurately judging consumer interest in marketing and improving personalized product recommendations and interactive experiences.
Patent Information
- Application Number
- CN202311620199.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-30
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2043-11-30
AI Technical Summary
Existing technologies struggle to accurately determine consumer interest in products during marketing, leading to a mismatch between product recommendations and user preferences.
By collecting and analyzing EEG signals, especially alpha waves, beta waves, and P300 waves, an interest judgment index and amplitude analysis are generated. Combined with fast Fourier transform and first-order difference, accurate judgment of consumer interests can be achieved.
It improves the accuracy of judging consumer interest, provides personalized product recommendations and interactive experiences, and enhances the accuracy of marketing and user satisfaction.
Smart Images

Figure CN117608404B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of brain-computer interface technology, and in particular to an interaction method, system and storage medium for brain-computer interfaces. Background Technology
[0002] Brain-computer interface (BCI) is a technology that enables interaction between the human brain and external devices by establishing direct communication between them. This technology can be applied in various fields, including medicine, entertainment, and industry.
[0003] With the continuous development of brain-computer interface technology, it can also be applied to neuromarketing. Neuromarketing refers to using neuroscience methods to study consumer behavior, explore the neural mechanisms of consumer decision-making, find the real driving force behind consumer behavior, and thus generate appropriate marketing strategies.
[0004] It actually emerged alongside the development of several fundamental disciplines supporting marketing theory in recent years. Among them, major breakthroughs in cognitive science and neuroscience played a key role. With the advancements in behavioral decision-making and cognitive science, marketing theory can borrow many psychological concepts to explain consumer behavior, such as implicit memory, automatic information processing, and the subconscious. Since the human brain controls all aspects of human behavior, understanding how the brain works not only helps explain human behavior but also helps marketers grasp the patterns of consumer behavior.
[0005] In the past, marketers have mostly relied on historical browsing data to understand consumer behavior patterns. However, the fragmented nature and difficulty in tracing historical data have made the analysis increasingly complex. Moreover, machine analysis has never been able to guarantee the accuracy of the analysis, and there are often cases where the recommended products do not match the user's preferences. Summary of the Invention
[0006] (a) Technical problems to be solved
[0007] To address the technical problems in the background art, this invention proposes an interaction method, system, and storage medium for brain-computer interfaces. By applying brain-computer interfaces to marketing, changes in brain neural signals can be used to determine the products that consumers are interested in, significantly improving the accuracy of the judgment results.
[0008] (II) Technical Solution
[0009] To achieve the above objectives, the present invention provides the following technical solution:
[0010] An interaction method for brain-computer interfaces, comprising:
[0011] The EEG signal data of the subjects is collected in real time by the EEG signal acquisition device, and the collected EEG signal data is transmitted wirelessly to the computer for preprocessing.
[0012] The frequency domain characteristics of each signal are obtained by using the real-time α-wave value and the real-time β-wave value through Fast Fourier Transform, thus obtaining the power magnitude P of the α-wave and β-wave. α ′ and P β Based on the magnitude of the alpha and beta wave power, an interest judgment index Xq is generated for the subject, thereby making a preliminary judgment on whether the subject is interested in the product;
[0013] Extract the P300 wave values from the time segments corresponding to the products identified as of the subject's interest, fit the discrete P300 wave values into a curve, and obtain the amplitude A at each time point in the time segment based on the characteristics of the curve. T Based on the amplitude, further determine whether the subject is interested in the product;
[0014] After determining that a subject is interested in a particular product, all alpha and beta wave power data for the corresponding time segment need to be stored in a second analysis table. This is done by calculating the normalized first-order difference between the alpha and beta wave power at two consecutive time points. and The system further analyzes the subject's thought processes during the browsing of the product to determine whether the subject chooses to continue browsing or browse the next product; based on the results of brain signal processing, it generates control commands for the computer; the computer then executes the corresponding operations according to the commands.
[0015] Specifically, the EEG signal acquisition device is used to collect the consumer's EEG signal data in real time. The EEG signal acquisition device includes several electrodes, and the electrode positions are selected according to the 10-20 system.
[0016] The electroencephalogram (EEG) signal acquisition device was placed on the scalp of the subject and fixed with an electrode cap;
[0017] The data acquisition device must be successfully connected to Bluetooth. If it fails, no EEG signal data can be acquired. After successful connection, it is necessary to perform debugging. If the debugging results fluctuate within the normal range, signal acquisition can begin.
[0018] Specifically, the data in the real-time data table is segmented according to the time each product is transmitted to the subject. After a product is transmitted to the subject and real-time alpha, beta, delta, and theta values are generated, the corresponding frequency domain features of the alpha, beta, delta, and theta waves from that point onwards are obtained through Fast Fourier Transform.
[0019]
[0020] Among them, P i This represents the power of each frequency band, where N is the number of real-time sampling points, and f max with f min Let X(f) represent the maximum and minimum frequencies of the wave in each frequency band, respectively, and let X(f) represent the fast Fourier transform of the real-time data.
[0021] Combining the power of the four frequency bands, the total power P can be obtained as: The relative power P in the α band can be calculated using the following formula. α The relative power P of the ′ and β band EEG waves β ′:
[0022]
[0023] The calculated relative power P α ′ and relative power P β The interest judgment index Xq is formed by combining the following formulas:
[0024]
[0025] Where k1 and k2 are the relative power P of the α frequency band, respectively. α The relative power P of the ′ and β band EEG waves β The weight coefficients of ′, and 0 < k2 < k1 < 1, k1 + k2 = 1;
[0026] Furthermore, the calculated interest judgment index Xq is compared with the preset interest index threshold. When the interest judgment index Xq is greater than the set threshold, it indicates that the subject's concentration is strong at that time; otherwise, it is determined that the subject's consciousness is relatively relaxed at that time.
[0027] If there are time points in a product's time segment where the subject's concentration is strong, it is preliminarily determined that the user is interested in the product, and the corresponding time point T and the real-time EEG signal data of the time segment are stored in the first analysis table for further analysis; otherwise, it is determined that the user is not interested in the product.
[0028] Specifically, the time point T and time segment data are extracted from the first analysis table, and all P300 wave values within the corresponding segment are extracted from the real-time data table; the extracted discrete P300 wave values are interpolated using the linear interpolation method to obtain their continuous variation curve within the time segment.
[0029] The least squares method is used to fit the interpolated P300 curve into a smooth curve. A Cartesian coordinate system is established, and the curve is placed in this coordinate system. The horizontal axis represents the data acquisition time point, and the vertical axis represents the amplitude of the P300 wave vibration, i.e., amplitude A.
[0030] Plot the corresponding time point T and the amplitude A of the P300 curve corresponding to the previous time point T-1 in a Cartesian coordinate system. T and A T-1 The average of the two amplitudes represents the actual amplitude of the P300 wave at time point T.
[0031] Furthermore, an amplitude threshold is set, and the actual amplitude at time point T is compared with the amplitude threshold. If the calculated actual amplitude is greater than the amplitude threshold, it is determined that the change in wave value at that time point is caused by the subject's interest in the product, that is, the subject is interested in the product; otherwise, it is determined that the subject is not interested in the product. After the judgment is completed, the first analysis table is cleared.
[0032] Specifically, after determining that a subject is interested in a certain product, all alpha and beta power data for the time segment corresponding to that product are stored in a second analysis table, and the real-time data entry is monitored.
[0033] Whenever new alpha and beta power data are stored in the second analysis table, the normalized first-order difference of the alpha and beta power is calculated. and The change in the subjects' interest in the product was detected based on the differential value, specifically:
[0034]
[0035] Where, ρ α ρ β Let σ be the average of the first-order differences of the α-wave power and β-wave power within their respective time segments. α σ β These represent the standard deviations of α-wave power and β-wave power within their respective time segments:
[0036]
[0037]
[0038] Where i represents the α wave and β wave, N represents the total number of samples within the time segment, and P i (n) represents the α-wave power and β-wave power values calculated from the nth sample on a certain electrode.
[0039] Whenever new data enters the second analysis table, the above calculation is performed. The difference between the calculated result and the previous calculated result is calculated. If the difference between the normalized first-order difference of the two alpha wave powers is lower than the threshold and the difference between the normalized first-order difference of the two beta wave powers is also lower than the threshold, it indicates that the subject's interest in the product has decreased. After the judgment is completed, the second analysis table is cleared.
[0040] Specifically, when the system determines that a subject is likely interested in a product, it will send a product viewing instruction to the computer. After receiving the product viewing instruction, the computer will stay on the product description page for a certain period of time, giving the subject enough time to make a preliminary understanding.
[0041] When the system determines that a subject is interested in a certain product, it sends a command to the computer to open the product details page. After receiving the command, the computer will open the product details page.
[0042] When it is determined that the subject's interest in a certain product has decreased, a command to exit the product details page needs to be sent to the computer. After receiving the command, the computer will exit the product details page.
[0043] An interactive system for brain-computer interfaces, comprising:
[0044] Data acquisition module: Collects real-time EEG signal data of the subjects, specifically including real-time alpha wave value, real-time beta wave value, real-time delta wave value, real-time theta wave value, and real-time P300 wave value;
[0045] Data preprocessing module: The acquired real-time EEG signal data is filtered using an FIR digital filter;
[0046] Data storage module: Stores the preprocessed data into the real-time data table in the system database, and stores the real-time EEG signal data of the time point T and time segment corresponding to the product that the subject is suspected of being interested in into the first analysis table; stores all alpha and beta wave power data of the time segment corresponding to the product of interest into the second analysis table;
[0047] Data analysis module: By processing and analyzing the stored information, it determines whether the subject is interested in a particular product and whether the degree of interest has changed;
[0048] Instruction generation module: Used to generate the corresponding instructions and transmit them to the corresponding computer.
[0049] A storage medium storing a computer program that, when executed by a processor, implements the aforementioned brain-computer interface interaction method.
[0050] (III) Beneficial Effects
[0051] This invention provides an interaction method, system, and storage medium for brain-computer interfaces, which have the following beneficial effects:
[0052] 1. By collecting the subjects' electroencephalogram (EEG) signal data in real time and performing real-time preprocessing and analysis, the results of the subjects' judgment on the degree of interest in the products can be obtained in a timely manner, and corresponding decisions can be made quickly.
[0053] 2. By analyzing and processing EEG signals, a relatively objective indicator for judging the subject's level of interest can be provided. EEG signals are a direct reflection of the subject's brain activity. By performing frequency domain analysis and feature extraction on the signals, the subject's level of interest can be accurately characterized.
[0054] 3. By making two assessments, the subject's attitude towards the product of interest can be identified more accurately. The first assessment, based on the power spectrum of alpha and beta waves, can preliminarily determine whether the subject is interested in the product. The second assessment, based on the amplitude of the P300 wave, can further determine the degree of the subject's interest in the product. These two assessments corroborate each other, improving the accuracy of the judgment.
[0055] 4. By analyzing changes in EEG signals, a better user experience can be provided when subjects browse products. When a subject shows interest in a particular product, all wave and wave power data for the corresponding time segment can be stored. Furthermore, based on the wave power and the normalized first-order difference between two consecutive time points, the subject's thought processes during product browsing can be further analyzed. This helps to provide subjects with more personalized services. Attached Figure Description
[0056] Figure 1 A flowchart illustrating the steps of an interaction method for a brain-computer interface provided by the present invention;
[0057] Figure 2 This is a schematic diagram of the structure of an interactive system for brain-computer interfaces provided by the present invention. Detailed Implementation
[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0059] refer to Figure 1 This invention provides an interaction method for brain-computer interfaces, comprising the following steps:
[0060] S1. The EEG signal data of the subject is collected in real time through the EEG signal acquisition device, and the collected EEG signal data is transmitted wirelessly to the computer for preprocessing.
[0061] While the subjects browsed products online, an EEG signal acquisition device was used to collect the consumers' EEG signal data in real time. The EEG signal acquisition device included several electrodes, and the electrode positions were selected according to the 10-20 system. The EEG signal acquisition device was placed on the subject's scalp and fixed with an electrode cap. The acquisition device must be able to connect successfully to Bluetooth. If it fails, no EEG signal data can be collected. After successful connection, it is necessary to adjust it. If the adjustment results fluctuate within the normal range, signal acquisition can begin.
[0062] The EEG signal acquisition module transmits the real-time acquired EEG signal data to the computer via Bluetooth, and stores the EEG data in a specific database through a designed program, and marks it as a product, i.e., the time when the computer transmits the product to the subject; the specific acquired EEG signal data includes real-time alpha wave value, real-time beta wave value, real-time delta wave value, real-time theta wave value, and real-time P300 wave value;
[0063] Alpha waves can reflect consumers' psychological state, including attention, relaxation level, and cognitive level. By collecting and analyzing alpha waves, we can understand consumers' perception and reaction to products, and thus conduct more precise interaction design. Alpha waves exhibit individual differences and are related to each person's physiological and psychological state. By collecting and analyzing alpha waves, we can understand the characteristics and needs of different consumers, thereby providing more targeted product recommendations and interactive experiences for each consumer.
[0064] Beta waves are typically associated with cognitive and alert states in the brain. Collecting and analyzing beta waves when processing and recognizing product information can reveal a consumer's overall cognitive state regarding the product, including whether they are alert or relaxed. This helps designers better understand consumers' cognitive processes, such as attention, comprehension, and reaction speed, allowing for product designs or recommendations that better align with consumers' cognitive levels and needs. Variations in the frequency and amplitude of beta waves may also be related to certain product characteristics. For example, specific visual or auditory features of a product can influence a consumer's beta wave response. Collecting and analyzing beta waves further reveals the relationship between product features and consumer cognition and response, providing more useful feedback for product design and optimization.
[0065] Among them, delta waves and theta waves are EEG signals of different frequencies, which are used in this method to evaluate the power of alpha waves and beta waves;
[0066] The P300 wave represents the brain's response to specific events or stimuli, particularly those related to cognitive and decision-making processes. Collecting and analyzing P300 waves during consumer-product interactions can reveal consumers' level of interest and cognitive responses. Higher P300 wave amplitudes and shorter latency generally indicate greater consumer interest or a stronger cognitive response. Analyzing P300 waves provides insights into consumers' subjective feelings and reactions to products, offering more targeted data for product recommendations and interaction design.
[0067] The P300 waveform can also reflect consumers' decision-making process. During product selection and purchase, consumers' decisions may be influenced by various factors, such as product features, price, and brand. Collecting and analyzing the P300 waveform can help understand consumers' comparison and selection processes of different products, as well as their purchase intentions and decision-making speed. This contributes to a better understanding of consumers' buying motivations and preferences, providing valuable feedback for product design and marketing strategies.
[0068] After the collected real-time EEG data is transmitted to the real-time data table in the designated database of the computer via Bluetooth, the real-time data is extracted and preprocessed. During the EEG signal acquisition process, there will be a lot of noise interference, such as power frequency interference, electromyography interference, and external environmental factors. Most of these are noises caused by human factors. To deal with the above noise, the EEG signal is filtered using an FIR digital filter, and the preprocessed data is put back into the corresponding position.
[0069] S2. Obtain the frequency domain characteristics of each signal by performing a Fast Fourier Transform on the real-time α-wave and real-time β-wave values, and obtain the power magnitude P of the α-wave and β-wave. α ′ and P β Based on the magnitudes of alpha and beta wave power, an interest index Xq is generated for the subject, thereby making a preliminary judgment on whether the subject is interested in the product; specifically including:
[0070] S201. Divide the data in the real-time data table into segments according to the time each product is transmitted to the subject, and input the segment information into the first processing unit. After a product is transmitted to the subject and real-time alpha wave, real-time beta wave, real-time delta wave, and real-time theta wave values are generated, obtain the corresponding frequency domain features from the alpha wave, beta wave, delta wave, and theta wave starting from this time through a fast Fourier transform. Specifically:
[0071]
[0072] Among them, P i This represents the power of each frequency band, where N is the number of real-time sampling points, and f max with f minLet X(f) represent the maximum and minimum frequencies of each frequency band, respectively, and let X(f) represent the fast Fourier transform of the real-time data.
[0073] Given that the frequency range of alpha waves is 8-13 Hz, beta waves are 14-30 Hz, delta waves are 0.5-3 Hz, and theta waves are 4-7 Hz, the specific powers of the corresponding alpha, beta, delta, and theta waves are as follows:
[0074]
[0075]
[0076]
[0077]
[0078] S202. Combining the power of the four frequency bands, the total power P can be obtained as follows: The relative power P in the α band can be calculated using the following formula. α The relative power P of the ′ and β band EEG waves β ′:
[0079]
[0080] S203, calculate the relative power P α ′ and relative power P β The interest judgment index Xq is formed by combining the following formulas:
[0081]
[0082] Where k1 and k2 are the relative power P of the α frequency band, respectively. α The relative power P of the ′ and β band EEG waves β The weight coefficients of ′ are given, and 0 < k2 < k1 < 1, k1 + k2 = 1.
[0083] S204. Compare the calculated interest judgment index Xq with the preset interest index threshold. When the interest judgment index Xq is greater than the set threshold, it indicates that the subject's concentration is strong at that time; otherwise, it is determined that the subject's consciousness is relatively relaxed at that time. If there is a time point in a product's time segment where the subject's concentration is strong, it is preliminarily determined that the user is interested in the product, and the corresponding time point T and the real-time EEG signal data of the time segment are stored in the first analysis table for further analysis; otherwise, it is determined that the user is not interested in the product.
[0084] By combining the power of alpha and beta waves to form an interest index, we can understand a user's reaction when they see a product, and thus make a preliminary judgment on whether the user is interested in the product.
[0085] S3. Extract the P300 wave values from the time segments corresponding to the products identified as of the subject's interest, fit the discrete P300 wave values into a curve, and obtain the amplitude A at each time point in the time segment based on the characteristics of the curve. T Based on the amplitude, further determination is made as to whether the subject is interested in the product; specifically including:
[0086] S301. Extract time point T and time segment data from the first analysis table, and extract all P300 wave values within the corresponding segment from the real-time data table.
[0087] S302. Use linear interpolation to interpolate the extracted discrete P300 wave values to obtain its continuous variation curve within the time segment.
[0088] S303. Use the least squares method to fit the interpolated P300 curve into a smooth curve, establish a plane rectangular coordinate system, put the curve into the coordinate system, the horizontal axis represents the data acquisition time point, and the vertical axis represents the amplitude of the P300 wave vibration, i.e., amplitude A.
[0089] S304. Plot the corresponding time point T and the amplitude A of the P300 curve corresponding to the previous time point T-1 in a Cartesian coordinate system. T and A T-1 The average of the two amplitudes represents the actual amplitude of the P300 wave at time point f.
[0090] S305. Set an amplitude threshold. Compare the calculated actual amplitude at time T with the amplitude threshold. If the calculated actual amplitude is greater than the amplitude threshold, it is determined that the change in wave value at that time point is caused by the subject's interest in the product, i.e., the subject is interested in the product; otherwise, it is determined that the subject is not interested in the product. After the judgment is completed, clear the first analysis table.
[0091] After initially determining whether a user is interested in a product, a second interest assessment is conducted using the amplitude of the P300 wave. This allows for a more comprehensive understanding of the user's reaction to the product and improves the overall accuracy of the assessment.
[0092] S4. After determining that the subject is interested in a certain product, all alpha and beta power data of the time segment corresponding to the product need to be stored in the second analysis table. By calculating the normalized first-order difference of the alpha and beta power at the two time points before and after, the subject's thought changes during the browsing process of the product can be further determined to determine whether the subject chooses to continue browsing or browse the next product.
[0093] After determining that a subject is interested in a particular product, all alpha and beta power data for the corresponding time segment need to be stored in a second analysis table, and the input of real-time data needs to be monitored. Whenever new alpha and beta power data is stored in the second analysis table, the normalized first-order difference of the alpha and beta power needs to be calculated in real time. and The change in the subjects' interest in the product was detected based on the differential value, specifically:
[0094]
[0095] Where, ρ α ρ β Let σ be the average of the first-order differences of the α-wave power and β-wave power within their respective time segments. α σ β These represent the standard deviations of α-wave power and β-wave power within their respective time segments:
[0096]
[0097]
[0098] Where i represents the α wave and β wave, N represents the total number of samples within the time segment, and P i (n) represents the α-wave power and β-wave power values calculated from the nth sample on a certain electrode.
[0099] Whenever new data enters the second analysis table, the above calculation is performed. The difference between the calculated result and the previous calculated result is calculated. If the difference between the normalized first-order difference of the two alpha wave powers is lower than the threshold and the difference between the normalized first-order difference of the two beta wave powers is also lower than the threshold, it indicates that the subject's interest in the product has decreased. After the judgment is completed, the second analysis table is cleared.
[0100] S5. Based on the results of brain signal processing, generate control commands for the computer; the computer executes the corresponding operations according to the commands.
[0101] When the system determines that a subject is likely interested in a product, it will send a product dwell instruction to the computer. After receiving the product dwell instruction, the computer will stay on the product introduction page for a certain period of time, giving the subject enough time to make an initial understanding.
[0102] When the system determines that a subject is interested in a product, it sends a command to the computer to open the product details page. Upon receiving the command, the computer opens the product details page. When the system determines that the subject's interest in a product decreases, it sends a command to the computer to exit the product details page. Upon receiving the command, the computer exits the product details page, thus achieving the interactive effect of the brain-computer interface.
[0103] refer to Figure 2 The present invention also provides an interactive system for brain-computer interfaces, specifically comprising:
[0104] Data acquisition module: Collects real-time EEG signal data of the subjects, specifically including real-time alpha wave value, real-time beta wave value, real-time delta wave value, real-time theta wave value, and real-time P300 wave value;
[0105] Data preprocessing module: The acquired real-time EEG signal data is filtered using an FIR digital filter;
[0106] Data storage module: Stores the preprocessed data into the real-time data table in the system database, and stores the real-time EEG signal data of the time point T and time segment corresponding to the product that the subject is suspected of being interested in into the first analysis table; stores all alpha and beta wave power data of the time segment corresponding to the product of interest into the second analysis table;
[0107] Data analysis module: By processing and analyzing the stored information, it determines whether the subject is interested in a particular product and whether the degree of interest has changed;
[0108] Instruction generation module: Used to generate the corresponding instructions and transmit them to the corresponding computer.
[0109] Furthermore, the storage medium of the present invention stores a computer program that, when executed by a processor, implements the above-described brain-computer interface interaction method.
[0110] The storage medium may include, for example, a memory card for a smartphone, a storage component for a tablet computer, a hard disk for a personal computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disc read-only memory (CD-ROM), a USB memory, or any combination of the above storage media.
[0111] This design based on EEG signal analysis can assess subjects' interest in products in real time, objectively, and with precision, and make corresponding adjustments and optimizations based on changes in subjects' thoughts, providing enterprises with more valuable market analysis and optimization solutions.
[0112] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
[0113] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. An interaction method for brain-computer interfaces, characterized in that: include: The EEG signal data of the subjects is collected in real time by the EEG signal acquisition device, and the collected EEG signal data is transmitted wirelessly to the computer for preprocessing. The frequency domain characteristics of each signal are obtained by using the real-time α-wave value and the real-time β-wave value through Fast Fourier Transform, thus obtaining the power magnitude P of the α-wave and β-wave. α ′ and P β Based on the magnitude of the alpha and beta wave power, an interest judgment index Xq is generated for the subjects to make a preliminary judgment on whether the subjects are interested in the product. Extract the P300 wave values from the time segments corresponding to the products identified as of the subject's interest, fit the discrete P300 wave values into a curve, and obtain the amplitude A at each time point in the time segment based on the characteristics of the curve. T Based on the amplitude, further determine whether the subject is interested in the product; After determining that a subject is interested in a particular product, all alpha and beta wave power data for the corresponding time segment need to be stored in a second analysis table. This is done by calculating the normalized first-order difference between the alpha and beta wave power at two consecutive time points. and The system analyzes the subject's thought processes while browsing the product to determine whether the subject chooses to continue browsing or move on to the next product; based on the results of brain signal processing, it generates control commands for the computer. The computer executes the corresponding operation based on the command.
2. The interaction method for brain-computer interfaces as described in claim 1, characterized in that: The EEG signal acquisition device is used to collect consumers' EEG signal data in real time. The EEG signal acquisition device includes several electrodes, and the electrode positions are selected according to the 10-20 system. The EEG signal acquisition device is placed on the scalp surface of the subject and fixed by electrode caps. The data acquisition device must be successfully connected to Bluetooth; otherwise, it will be impossible to acquire EEG signal data. After a successful connection, debugging is still required. If the debugging results fluctuate within the normal range, signal acquisition will begin.
3. The interaction method for brain-computer interfaces as described in claim 2, characterized in that: The data in the real-time data table is segmented according to the time each product is transmitted to the subject. After a product is transmitted to the subject and real-time alpha, beta, delta, and theta values are generated, the corresponding frequency domain features of the alpha, beta, delta, and theta waves from that point onwards are obtained through Fast Fourier Transform. Specifically: Among them, P i This represents the power of each frequency band, where N is the number of real-time sampling points, and f max with f min Let X(f) represent the maximum and minimum frequencies of the wave in each frequency band, respectively, and let X(f) represent the fast Fourier transform of the real-time data. Combining the power of the four frequency bands, the total power P can be obtained as: The relative power P in the α band can be calculated using the following formula. α The relative power P of the ′ and β band EEG waves β ′: The calculated relative power P α ′ and relative power P β The interest judgment index Xq is formed by combining the following formulas: Where k1 and k2 are the relative power P of the α frequency band, respectively. α The relative power P of the ′ and β band EEG waves β The weight coefficient of ′, and 0 <k2<k1<1,k1+k2=1。 4. The interaction method for brain-computer interface as described in claim 3, characterized in that: The calculated interest judgment index Xq is compared with the preset interest index threshold. When the interest judgment index Xq is greater than the set threshold, it indicates that the subject's concentration is strong; otherwise, it is determined that the subject's consciousness is relatively relaxed. If there are time points in a product's time segment where the subject's concentration is strong, it is preliminarily determined that the user is interested in the product, and the corresponding time point T and the real-time EEG signal data of the time segment are stored in the first analysis table for further analysis; otherwise, it is determined that the user is not interested in the product.
5. The interaction method for brain-computer interface as described in claim 1, characterized in that: Extract time point T and time segment data from the first analysis table, and extract all P300 wave values within the corresponding segment from the real-time data table; use linear interpolation to interpolate the extracted discrete P300 wave values to obtain their continuous variation curve within the time segment. The interpolated P300 curve was fitted to a smooth curve using the least squares method. A Cartesian coordinate system was established, and the curve was placed in this system. The horizontal axis represents the data acquisition time point, and the vertical axis represents the amplitude of the P300 wave vibration, i.e., amplitude A. The corresponding time point T and the amplitude A of the P300 curve corresponding to the previous time point T-1 were marked on the Cartesian coordinate system. T and A T-1 The average of the two amplitudes represents the actual amplitude of the P300 wave at time point T.
6. The interaction method for brain-computer interface as described in claim 5, characterized in that: The actual amplitude at time point T is calculated and compared with the amplitude threshold. If the calculated actual amplitude is greater than the amplitude threshold, it is determined that the change in wave value at that time point is caused by the subject's interest in the product, that is, the subject is interested in the product; otherwise, it is determined that the subject is not interested in the product. After the judgment is completed, the first analysis table is cleared.
7. The interaction method for brain-computer interface as described in claim 1, characterized in that: After determining that a subject is interested in a particular product, all alpha and beta power data for the corresponding time segment are stored in a second analysis table, and the input of real-time data is monitored. Whenever new alpha and beta power data are stored in the second analysis table, the normalized first-order difference of the alpha and beta power is calculated. and The change in the subjects' interest in the product was detected based on the differential value, specifically: Where, ρ α ρ β Let σ be the average of the first-order differences of the α-wave power and β-wave power within their respective time segments. α σ β These represent the standard deviations of α-wave power and β-wave power within their respective time segments: Where i represents the α wave and β wave, N represents the total number of samples within the time segment, and P i (n) represents the α-wave power and β-wave power values calculated from the nth sampling on a certain electrode; Whenever new data enters the second analysis table, the above calculation is performed. The difference between the calculated result and the previous calculated result is calculated. If the difference between the normalized first-order difference of the two alpha wave powers is lower than the threshold and the difference between the normalized first-order difference of the two beta wave powers is also lower than the threshold, it indicates that the subject's interest in the product has decreased. After the judgment is completed, the second analysis table is cleared.
8. The interaction method for brain-computer interface as described in claim 1, characterized in that: When the system determines that a subject is likely interested in a product, it will send a product dwell instruction to the computer. After receiving the product dwell instruction, the computer will stay on the product introduction page for a certain period of time, giving the subject enough time to make an initial understanding. When the system determines that a subject is interested in a certain product, it sends a command to the computer to open the product details page. After receiving the command, the computer will open the product details page. When it is determined that the subject's interest in a certain product has decreased, a command to exit the product details page needs to be sent to the computer. After receiving the command, the computer will exit the product details page.
9. An interactive system for brain-computer interfaces, used to implement the method according to any one of claims 1 to 8, characterized in that: include: Data acquisition module: Collects real-time EEG signal data of the subjects, specifically including real-time alpha wave value, real-time beta wave value, real-time delta wave value, real-time theta wave value, and real-time P300 wave value; Data preprocessing module: The acquired real-time EEG signal data is filtered using an FIR digital filter; Data storage module: Stores the preprocessed data into the real-time data table in the system database, and stores the real-time EEG signal data of the time point T and time segment corresponding to the product that the subject is suspected of being interested in into the first analysis table; All alpha and beta power data for the time segment corresponding to the product of interest are stored in the second analysis table; Data analysis module: By processing and analyzing the stored information, it determines whether the subject is interested in a particular product and whether the degree of interest has changed; Instruction generation module: Used to generate the corresponding instructions and transmit them to the corresponding computer.
10. A storage medium storing a computer program, characterized in that, When the program is executed, it implements the brain-computer interface interaction method as described in any one of claims 1 to 8.
Citation Information
Patent Citations
Ultra-dense electrode-based brain imaging system
CN108024752A
Shopping decision-making method based on brain-computer interaction
CN108932511A
