32 results about "Visual evoked potentials" patented technology

The present application relates to the technical field of brain-computer interface, and provides a steady-state visual evoked potential brain-computer interface instruction classification method and device, wherein the method comprises: acquiring electroencephalogram data generated by a user under a frequency-semantic joint stimulation; inputting the electroencephalogram data into an electroencephalogram decoding model to obtain an instruction classification result; a frequency perception feature extractor and a semantic decoding feature extractor respectively extract frequency features and semantic features from the electroencephalogram data in parallel; and a joint decision module obtains the instruction classification result based on the frequency features and the semantic features. The present application expands the number of encodable instructions to the product of the frequency domain dimension and the semantic dimension by designing a frequency-semantic joint stimulation; the frequency perception feature extractor and the semantic decoding feature extractor can simultaneously extract frequency features and semantic features from the electroencephalogram signal, and then the joint decision module is used for fusion decision, so that the accuracy of instruction classification and the interaction efficiency are significantly improved under the premise of ensuring high comfort.

The application provides a steady-state visual evoked potential classification method and device and electronic equipment. The method comprises the following steps: obtaining a steady-state visual evoked potential signal; based on a plurality of different frequency band bandpass filters, extracting short time domain window data corresponding to each frequency band in the plurality of different frequency bands from the steady-state visual evoked potential signal; inputting the short time domain window data corresponding to the plurality of frequency bands into a pre-trained hybrid network model to obtain a classification result of the steady-state visual evoked potential signal; wherein the hybrid network model comprises a convolutional neural network and a bidirectional gated recurrent unit, the convolutional neural network is used to obtain a fusion feature map based on the short time domain window data corresponding to the plurality of frequency bands, and compress the fusion feature map into one-dimensional data; the bidirectional gated recurrent unit is used to obtain the classification result of the steady-state visual evoked potential signal based on the one-dimensional data. The application can accurately classify the short time domain window potential signal data and improve the classification performance.

The invention provides a steady-state visual evoked potential brain-computer interface system for snowflake point weak flicker coding, and the system comprises a stimulation presentation module which is used for presenting a stimulation interface of snowflake point weak flicker; each stimulation target comprises a corresponding frequency and initial phase combination, and the frequencies and initial phase combinations corresponding to different stimulation targets are different; the stimulation interface is used for guiding the sight line of the user to be transferred to a current stimulation target when each trial starts; the electroencephalogram collection module is used for collecting electroencephalogram signals of the occipital area of the user through a dry electrode head ring, and the dry electrode head ring comprises a plurality of movable electrodes, a reference electrode and a grounding electrode; the signal processing module is used for processing the electroencephalogram signal to identify a steady-state visual evoked potential and acquiring a frequency and phase combination of a stimulation target corresponding to the electroencephalogram signal; and the interaction control module is used for outputting a corresponding control instruction according to the frequency and phase combination of the stimulation target identified by the signal processing module to realize brain-computer interaction.

This invention relates to the field of brain-computer interface (BCI) data processing technology, and discloses a BCI data processing method that integrates steady-state motor visual evoked potentials and motor imagery. The method includes: generating a bimodal task configuration table; generating a bimodal synchronized stimulation sequence; acquiring and labeling multi-lead EEG signals; extracting visual response features and motor imagery response features; performing motor intention recognition; and generating rehabilitation feedback information. Compared to the single visual evoked paradigm in existing technologies, especially under conditions of significant differences in subjects' motor imagery abilities and low EEG noise ratios, this invention addresses the technical problem of failing to achieve stable motor intention recognition. By synchronously binding motor visual stimulation with motor imagery tasks and jointly extracting visual frequency response and motor imagery desynchronization features, stable recognition of motor intention is achieved, improving the recognition accuracy of rehabilitation BCI training.

This invention discloses a brain-controlled cotton candy machine control method based on attention level and SSVEP (Steady State Visual Evoked Potential), belonging to the field of brain-computer interface and food processing equipment integration technology. This invention employs a NeuroSci wireless EEG acquisition system, using SSVEP steady-state visual evoked potentials to achieve flavor selection (original, strawberry, pineapple), and utilizes frontal electrodes to collect EEG signals to calculate attention level. An STM32F407ZGT6 is used as the main controller to establish a precise mapping relationship between attention level grading and the cotton candy machine's motor speed. This is combined with servo motors for automatic quantitative feeding, infrared monitoring for candy card reversal and obstacle clearance, and touch display, voice commands, and multimodal audio-visual feedback. A state machine logic is used to complete the entire process of automatic control, from initialization and preheating to interaction, production, and completion. This invention achieves closed-loop control of "EEG acquisition—signal analysis—device execution," solving the problems of cumbersome operation, poor interactivity, and limited functionality in traditional cotton candy machines. It features intelligent operation, high stability, and high safety, and can be used in parent-child interaction, brain science popularization, and attention training scenarios, possessing high application and promotion value.

A neurofeedback system includes an EEG apparatus, a presentation apparatus and a controller. The controller is configured to: (i) cause the presentation apparatus to display an overlaid image to the user that comprises a first image flickering at a first frequency and a second image flickering at a second frequency different than the first frequency, the first image being an affective distractor stimulus image and the second image being a task-relevant stimulus image, (ii) receive from the EEG apparatus a number of first steady-state visual evoked potential (SSVEP) signals generated in response the first image of the overlaid image and a number of second SSVEP signals generated in response the second image of the overlaid image, and (iii) calculate feedback indicative of how much attention of the was user allocated to the task-relevant stimulus image versus how much attention of the user was allocated to the affective distractor stimulus image.

The application discloses a brain-computer interface system and method, which comprises a visual stimulation module, a collection module and an analysis module. The visual stimulation module provides a plurality of visual stimulation targets, each of which comprises a background gray adjustable display area and a visual stimulation area formed in the display area. The visual stimulation area does not completely fill the display area, and the display area and the visual stimulation area jointly constitute a visual stimulation code. The collection module is used for collecting brain electrical signals generated by a user to the visual stimulation code. The analysis module extracts features in the brain electrical signals and identifies a visual stimulation target currently gazed by the user. The application adopts a contrast coding method and can be used on a common refresh rate display to realize a high-frequency multi-target steady-state visual evoked potential brain-computer interface system.

The invention discloses a target identification method based on a steady-state visual evoked potential brain-computer interface and a related device, and relates to the technical field of brain-computer interfaces, and the method comprises the steps: designing a trained identification model which comprises a segment coding module, a plurality of feature extraction modules and a classification module which are connected in sequence, the feature extraction module comprises a first normalization layer, a self-adaptive frequency spectrum module, an enhanced time delay neural network module, an inverse Fourier transform layer, a second normalization layer, a splicing layer, an interactive convolution module and a first addition layer, the trained recognition model is used for determining the recognition frequency corresponding to the electroencephalogram signals, the recognition frequency serves as the target frequency, and the recognition frequency is used as the target frequency. According to the method and the device, the three core application requirements of high-precision identification, cross-subject strong generalization ability, light weight and low delay of the SSVEP-BCI system under an ultra-short time window can be met at the same time.

The invention discloses a discriminative domain adaptation-based cross-subject SSVEP (Steady-State Visual Evoked Potential) signal decoding method, which is characterized in that an end-to-end double-branch shared weight neural network is constructed, joint training of source domain data and target domain data is realized, and cross-subject migration can be completed only by using a small amount of label-free target subject data. The model adopts a multi-loss joint optimization strategy including cross entropy loss, comparison pairing loss, minimum category confusion loss and maximum mean value difference loss, collaborative enhancement of intra-class compactness, suppression of prediction confusion and fine alignment of inter-domain feature distribution. According to the method, manual feature extraction is not needed, the network can directly learn the discriminative spatial-temporal features from the original SSVEP signals, and the target domain classification precision and generalization ability are remarkably improved while the stimulation frequency related structure is kept through a discriminative domain adaptation mechanism.

The embodiment of the invention discloses a control method based on an electroencephalogram signal, a medium, equipment and a product. The method comprises the steps of obtaining the electroencephalogram signal; the electroencephalogram signal comprises a steady-state visual evoked potential signal and a motor imagery signal; identifying the steady-state visual evoked potential signal to obtain a to-be-controlled object and an operation intention for the to-be-controlled object; and performing motor imagery classification on the motor imagery signal to obtain a target imagery mode; the operation intention corresponds to a preset imagination mode; if the preset imagination mode corresponding to the operation intention is consistent with the target imagination mode, generating a control instruction based on the operation intention; and controlling the to-be-controlled object to execute the control instruction. According to the embodiment of the invention, the control efficiency can be improved on the premise that the driving safety is guaranteed, the efficient human-vehicle interaction of'wanted control 'is realized, the control experience of a user is improved, and hidden potential safety hazards are avoided.

The invention relates to the technical field of electroencephalogram control, and discloses an intelligent nursing bed self-adaptive control system and method based on multi-mode electroencephalogram intention recognition, and the system comprises a signal separation module, a state judgment module, an intention feature analysis module, a lateral feature extraction module, a hierarchical intention decision module and a control instruction generation module. Acquiring an occipital area steady-state visual evoked potential signal and a motion-related cortex potential signal based on the original electroencephalogram signal of the patient; analyzing the phase locking stability of the occipital area steady state visual evoked potential signal to judge an attention effective state; analyzing the motion-related cortex potential signal to obtain an energy change sequence; analyzing a contralateral dominating mode of the patient; performing hierarchical intention decision on the patient to obtain a preliminary motion intention of the patient; generating a final control instruction of the target equipment in combination with the physiological feedback signal of the patient; according to the invention, the efficiency of adaptive control of the intelligent nursing bed based on multi-mode electroencephalogram intention recognition can be improved.

This invention discloses a method for decoding EEG features for steady-state visual evoked potentials, including: the subject's fixation frequency being f n The original steady-state visual evoked potentials collected during visual stimulation are X. n Using N fb A filter bank for X n Preprocessing is performed to obtain N fb After preprocessing, the steady-state visual evoked potential pairs are filtered to obtain useful signal components and noise signal components, and then a spatial filter is obtained to obtain the steady-state visual evoked potential template signal. This is used to acquire the unknown steady-state visual evoked potential K and obtain N. fb K is a preprocessed unknown steady-state visual evoked potential. (m) By improving and K (m) The signal-to-noise ratio, calculate N fb The filtered sum K (m) The correlation coefficient of N fb The weighted summation yields ρ n ; We obtained K and steady-state visual evoked potentials [X1,…X i ,…,X n ,...,X Nf The correlation coefficient between ], ρ i If the maximum value is reached, then the visual stimulus frequency of K is f. i The technical solution in this embodiment achieves the effect of improving the accuracy of recognizing unknown visual stimuli using a small amount of training data.

The invention discloses a learning system and method based on steady-state visual evoked potential and transcranial magnetic regulation. The learning system comprises a display module which comprises a display interface and a flicker grid, the display interface is used for displaying a learned question, the flicker grid comprises a plurality of option boxes, each option box corresponds to an answer option, and the plurality of option boxes have different flicker frequencies; the electroencephalogram acquisition module is used for acquiring real-time electroencephalogram signals of a user and sending the real-time electroencephalogram signals to the data processing module; the data processing module is connected with the display module and the electroencephalogram acquisition module, and is used for analyzing the electroencephalogram signals acquired by the electroencephalogram acquisition module, obtaining SSVEP characteristic frequency, comparing the SSVEP characteristic frequency with flicker frequency of an option box, selecting answer options watched by a user, and displaying the options on the display module; a control instruction is output to the magnetic stimulation module according to the answering condition of the user; and the magnetic stimulation module is used for applying magnetic stimulation to the user.

The application discloses a visual evoked potential examination instrument based on refractive error compensation and a use method thereof, and relates to the technical field of visual evoked potential examination instruments, which comprises a human eye refractive error measurement system, a human eye refractive error compensation system and a visual evoked potential collection system. The human eye refractive error measurement system is used for objectively measuring human eye refractive error. The human eye refractive error compensation system is used for compensating the human eye refractive error. The visual evoked potential collection system is used for collecting visual evoked potential based on the compensation of the human eye refractive error. The visual evoked potential can be used for objective evaluation and examination of visual function and visual pathway disorders such as visual acuity and contrast sensitivity. In the application, a stimulation pattern is projected to the retina through an optical system. The human eye refractive error is objectively measured, and the human eye refractive error is accurately compensated by using optical focusing and a rotating cylindrical lens. In this case, the retina is stimulated by a flash or a pattern, the influence of the human eye refractive error on the projection of visual stimulation to the retinal fundus and the visual evoked potential is eliminated, and therefore, the accuracy, consistency and ease of use of the VEP examination are improved.

The invention discloses an asynchronous BCI identification system based on multi-brain-region feature fusion of an action observation normal form. The asynchronous BCI identification system comprises a data preprocessing and anti-leakage data division unit, a multi-brain-region feature collaborative extraction and weighting unit, an integrated decoding unit and a dynamic decision and output unit. According to the method, the BCI is constructed by using the motion observation normal form, brain activities related to motion preparation can be induced, possibility is provided for fusing the features of the prefrontal lobe region reflecting the attention level, the motion is presented through the frame rate reduction technology, and meanwhile, the steady-state motion visual evoked potential can be induced in the visual region. By integrating multi-dimensional and complementary electroencephalogram characteristics of a prefrontal lobe (cognition / attention), a motion area (motion intention) and a visual area (visual stimulation phase-locked response), an asynchronous BCI with more comprehensive information, more stable judgment and stronger anti-interference capability is expected to be constructed, so that the reliability of the asynchronous BCI in medical application is improved.

The invention discloses an SSVEP (Steady-State Visual Evoked Potential) source separation system and method based on multivariate autoregressive cross power spectrum analysis. The method comprises the following steps: firstly, carrying out band-pass filtering, downsampling and principal component analysis on original multichannel electroencephalogram signals, and calculating a whitened data matrix; establishing a multiple autoregression MVAR model, and estimating a model parameter matrix; then calculating a cross power spectral density (CPSD) matrix among the channels; then manifold optimization decomposition is carried out, under a Stiefel manifold constraint condition, a Riemannian trust region optimization algorithm is adopted to maximize a CPSD objective function, a spatial filtering matrix is iteratively solved, and neural source components synchronized with stimulation are extracted; and finally, carrying out reverse reconstruction on the decomposed component signals, and outputting spatial and temporal distribution of the signals. According to the method, the multiple autoregression (MVAR) model and the cross power spectral density (CPSD) analysis are combined, the time delay dependency relationship and the frequency domain coherent structure of the neural signal can be captured at the same time, the neural source component of the steady state visual evoked potential (SSVEP) is efficiently extracted, and higher physiological rationality and higher signal-to-noise ratio are achieved.

This invention relates to a brain-computer interface (BCI) control method and system for a rehabilitation robot, comprising the following steps: S1: Using a target image stimulation paradigm, visual evoked signals are induced in the operator, and the operator's electroencephalogram (EEG) signals are simultaneously acquired. The EEG signals include two types: a first type is steady-state visual evoked potential (VEP) EEG signals, and a second type is hand fine motor intention EEG signals; S2: The steady-state VEP EEG signals are decoded to identify the target object and its spatial position in the image being viewed by the operator, generating a target selection command. By employing the above technical solution, this invention simultaneously acquires steady-state VEP EEG signals and hand fine motor intention EEG signals, which are used for target position selection and robot hand movement type determination, respectively. This achieves decision fusion of dual-modal EEG signals, overcoming the problems of low dimensionality and poor flexibility in single-mode control.

The application provides a dynamic adjustment micro-current stimulation and training method and system for ophthalmology. The application integrates various sensors through a wearable periocular device, collects physiological data such as corneal impedance, intraocular pressure signals, blinking behavior and visual evoked potential in real time, combines denoising, normalization, dimensionality reduction and multi-modal fusion processing, and constructs a standardized eye physiological state vector. The vector and incremental clustering are used to identify the micro-current stimulation tolerance inflection point, generate an individual safety upper limit initial value, combine a hierarchical fuzzy reasoning system and a dynamic comprehensive risk index, realize adaptive adjustment of the safety current limiting threshold, and finally smooth the micro-current output through a PID controller. The system can also update the model through cross-course memory data closed loop to improve the personalization and safety of the regulation. The application can dynamically adapt to individual physiological differences, effectively guarantee the biological safety of micro-current stimulation, and meet the needs of continuous and long-term adaptive fine intervention.

The invention discloses an individual specific feature decoupling method based on steady-state visual evoked potential, which comprises the following steps: S1, acquiring steady-state visual evoked potential signal data, preprocessing the steady-state visual evoked potential signal data as input, and using Euclidean to align the data to reduce differences between sessions; s2, respectively extracting task features and individual specificity features by using two double-branch feature extraction frameworks with the same parameters; s3, decoupling the extracted features through decorrelation constraint and a gradient inversion layer, and subtracting individual specific features by using feature subtraction to obtain pure task features; and S4, inputting the pure task features into an attention-based feature enhancement and classifier model, and realizing accurate decoding of the steady-state visual evoked potential in combination with a multi-loss supervision optimization model. The invention further discloses an individual specific feature decoupling system based on the steady-state visual evoked potential. According to the method, feature decoupling and enhancement can be realized, and the decoding accuracy, the information transmission rate and the robustness are remarkably improved.

This invention discloses a control system and method for Internet of Things (IoT) devices based on steady-state visual evoked potentials (SSVEP) and augmented reality (AR). The system is applied to wearable AR devices and includes an SSVEP signal acquisition module, an AR display and stimulus presentation module, a processing and recognition module, a wireless communication and control module, and an AR feedback presentation module. The AR module renders a virtual control object with a specific flickering frequency in the user's field of vision, while the acquisition module acquires EEG signals from the user's occipital cortex in real time. The processing and recognition module decodes the user's gaze intent and generates control commands using CCA or deep learning algorithms. Furthermore, the system integrates eye tracking for dual intent verification and utilizes a context-aware module to adaptively switch scene modes. This invention achieves "what you see is what you get" hands-free, silent interaction, effectively solving the control challenges of IoT devices under conditions of hand-occupancy and environmental noise interference, and possesses the advantages of high robustness and low cognitive load.

This invention provides a real-time interactive method and system for a brain-controlled rehabilitation robot based on a brain-computer interface. It synchronously acquires the user's electroencephalogram (EEG) signals and records the steady-state visual evoked potential (SSVEP) stimulation frequencies associated with different rehabilitation tasks. The system then processes the EEG signals in parallel to extract features: extracting motor imagery-related frequency band signals, obtaining tangent space feature vectors by calculating the Riemann covariance matrix, and calculating the spectral Shannon entropy; calculating response features for each SSVEP stimulation frequency; constructing a composite feature vector, inputting the vector into an intention classification model, and outputting preliminary decoding results including candidate motor intentions and confidence levels; setting a rejection threshold; if the highest confidence level is below the threshold, it is determined as an invalid instruction and feature extraction is repeated; if it is above the threshold, the corresponding candidate intention is determined as a control instruction; inputting the control instruction confidence level, SSVEP response features, and spectral Shannon entropy into a dynamic mapping model to calculate motion parameters and drive the rehabilitation robot to perform actions.

This invention discloses a method for decoding electroencephalogram (EEG) signals based on the fusion of time-domain and frequency-domain signals, relating to the fields of brain-computer interfaces and EEG signal processing. The method includes: establishing a brain-computer interface data acquisition environment based on encoded modulation visual evoked potentials; acquiring the raw EEG observation signal set of the subject under multi-class encoded stimuli using a multi-channel EEG acquisition device; and adaptively filtering non-stationary noise, fully exploiting time-frequency complementary features, preserving high-order statistical information and fine-grained temporal structure of the signal, and significantly improving the accuracy and robustness of EEG signal decoding.

The application discloses a training method of a single-time spatial-temporal filter based on a periodic characteristic and a discriminant analysis method, and comprises the following steps: collecting original steady-state visual evoked potentials of a subject when a video frequency is f n ; pre-processing the original steady-state visual evoked potentials to obtain a sub-band; dividing the sub-band according to a length to obtain a number of data segments with the length, and the value of the length increases with the increase of the frequency f n ; averaging the data segments to obtain, and connecting multiple to obtain ; expanding and enhancing to obtain ; projecting onto a subspace spanned by an ideal reference signal to obtain ; re-enhancing to obtain ; calculating an inter-class difference matrix and an intra-class difference matrix; calculating a scatter matrix and ; obtaining a projection subspace capable of effectively classifying all classes by using generalized eigenvalue decomposition, and obtaining the spatial-temporal filter according to the projection subspace. The application achieves the technical effect of reducing the training cost of the SSVEP spatial-temporal filter.

The application discloses a kind of space positioning systems based on steady-state visual evoked potential.The stimulation module in the system utilizes the spatial coding strategy of four flicker arrays, the SSVEP signal induced is amplified and collected after signal pretreatment is carried out by electroencephalogram signal collection module, then the amplitude, phase, correlation coefficient of each stimulation frequency component and the ratio of the correlation coefficient corresponding to each coding stimulation frequency or phase are used as characteristic information to decode spatial information to determine the coordinates of visual fixation, finally the coordinate information is converted into instruction and given to controlled operation module, to realize visual feedback.The application can effectively solve the problem that the area of divided region and the number of target in each partition are limited in existing spatial coding, and the edge information of each partition cannot be distinguished, realizes the BCI of low visual load and accurate positioning in screen, and provides the possibility for realizing high-integration embedded BCI system.

This invention discloses a graded obstacle avoidance brain-controlled wheelchair and its control method based on steady-state visual evoked potentials (SVPs). By collecting EEG signals from the user's occipital lobe region and combining them with visual stimulation to induce steady-state visual evoked potentials, wheelchair movement commands are generated. Simultaneously, an environmental perception module acquires obstacle information in real time, and the wheelchair movement control module determines the obstacle avoidance status accordingly. Furthermore, the decoding process of the steady-state visual evoked potentials is constrained during the brain-controlled command generation stage, achieving coordinated control of obstacle avoidance warnings and brain-controlled commands. Through graded obstacle avoidance processing and corresponding feedback mechanisms, unsafe control commands are restricted at the generation stage, thereby improving the safety, stability, and applicability of the brain-controlled wheelchair in complex environments.

A steady-state visual evoked potential classification method based on a multistage time-frequency collaborative channel attention mechanism comprises the steps that 1, tested electroencephalogram signals under different frequencies are collected to serve as sample signals, the sample signals are preprocessed to obtain a sample data set, and the sample data set is divided into a training set, a verification set and a test set according to a set proportion; constructing an MSFDCA-Net network which comprises a branch-level time-frequency collaborative channel attention convolution layer, a multi-branch feature fusion layer, a final feature extraction convolution layer and an output layer in sequence from an input layer; the training set is input into the MSFDCA-Net network for training, and a trained MSFDCA-Net network is obtained; and verifying the trained MSFDCA-Net network by using the verification set to obtain a final MSFDCA-Net network. And inputting the test set into the final MSFDCA-Net network to complete identification of different electroencephalogram signals in the test set. Through the learnable adaptive filter bank and the multi-level attention fusion mechanism, the limitation of time-frequency cooperation deficiency and feature fusion roughness of an existing method is overcome, the accuracy and robustness of SSVEP classification are remarkably improved, and the method is particularly suitable for a high-speed brain-computer interface system under a short time window.

The application discloses a control strategy generation method and device for brain-controlled rehabilitation equipment, an equipment and a storage medium, relates to the technical field of signal processing, and comprises the following steps: acquiring multi-channel electroencephalogram signals and performing preprocessing to obtain target electroencephalogram signals comprising steady-state visual evoked potential signals and motor imagery signals, wherein the target electroencephalogram signals correspond to preset action categories; decoding the steady-state visual evoked potential signals by using a filter group task-related component analysis to obtain a correlation score vector; decoding the motor imagery signals by using a Mamba dynamic routing spatiotemporal network model to obtain a classification score vector; performing posterior probability distribution conversion and weighted fusion on the correlation score vector and the classification score vector to obtain a fusion probability distribution; and generating a target control strategy according to an action category corresponding to a highest probability value in the fusion probability distribution. The control strategy obtained by the application can take into account both intention recognition accuracy and rehabilitation neural activation effect.

The application discloses a mechanical arm grabbing method based on stimulus control adaptive updating, which can be used for matching a steady-state visual evoked potential (SSVEP) based rehabilitation auxiliary brain-computer interface system for different application requirements. The method process is as follows: a mask region model for the task is obtained by training a self-built data set, target recognition and mask segmentation of the experimental scene image are carried out by using the mask region model, the classification, the bounding box and the mask of the object in the image are obtained, adaptive frequency and phase distribution are carried out based on the same, the stimulus control is updated in real time, the user induces the SSVEP signal by staring at the corresponding stimulus control of the target object, the recognition result is intuitively fed back to the user through display mask and the like after classification analysis, and the mechanical arm is controlled to assist the user to complete the grabbing and the like.