Control methods, devices, electronic devices, and storage media based on electroencephalogram (EEG) signals

Abstract

This invention discloses a control method, device, electronic device, and storage medium based on electroencephalogram (EEG) signals. The method includes: acquiring the EEG signal of a target object; extracting target category discrimination feature parameters corresponding to target category discrimination features in the EEG signal; determining the artifact identification result in the EEG signal based on the target category discrimination feature parameters and target category reference parameters, wherein the target category discrimination features and target category reference parameters are pre-determined based on the feature classification results of the sample EEG signal; determining the action state of the target object based on the artifact identification result; and executing the processing operation corresponding to the action state. By classifying signal artifacts in the EEG signal using pre-determined target category discrimination features and target category reference parameters, the complexity of the algorithm is reduced, the problem of existing electromyography (EMG) classification algorithms being unable to be embedded in chips is solved, and control operations based on EEG signal artifact categories are realized.

Description

Technical Field

[0001] This invention relates to the field of electroencephalogram (EEG) signal analysis and processing technology, and in particular to a control method, device, electronic device, and storage medium based on EEG signals. Background Technology

[0002] Electroencephalogram (EEG) signals are extremely weak compared to other types of physiological signals, with amplitudes on the order of microvolts. During EEG signal acquisition, environmental or human interference is inevitably introduced, affecting the quality of the EEG signals.

[0003] Currently, most common electromyography (EMG) classification algorithms rely on wireless surface EMG testing systems to perform detailed EMG classification tasks, such as determining the magnitude and direction of arm torque and distinguishing the shape of grasped targets. However, existing EMG analysis algorithms require a complete EMG signal acquisition system, which is costly. Furthermore, the development of EMG classification algorithms is mostly based on common machine learning or deep learning models, making it impossible to embed them into chips to meet computational requirements. Therefore, it is difficult to implement EMG-based control in computer devices. Summary of the Invention

[0004] One technical problem that this invention aims to solve is that existing electromyography (EMG) classification algorithms cannot be embedded in chips, making it difficult to achieve control based on EEG signals.

[0005] According to one aspect of the present invention, a control method based on electroencephalogram (EEG) signals is provided, comprising:

[0006] Acquire the electroencephalogram (EEG) signal of the target object and extract the target category discrimination feature parameters corresponding to the target category discrimination features in the EEG signal;

[0007] The result of violation identification in EEG signals is determined based on the target category identification feature parameters and the target category reference parameters, wherein the target category identification features and the target category reference parameters are determined in advance based on the feature classification results of the sample EEG signals;

[0008] Based on the violation identification results, the action state of the target object is determined, and the corresponding processing operation is executed.

[0009] According to another aspect of the present invention, a control device based on electroencephalogram (EEG) signals is provided, comprising:

[0010] The discriminative feature extraction module is used to acquire the EEG signal of the target object and extract the target category discriminative feature parameters corresponding to the target category discriminative features in the EEG signal;

[0011] The deliberate violation identification result determination module is used to determine the deliberate violation identification result in the EEG signal based on the target category identification feature parameters and the target category reference parameters. The target category identification features and the target category reference parameters are determined in advance based on the feature classification results of the sample EEG signal.

[0012] The processing operation execution module is used to determine the action state of the target object based on the violation identification result and execute the processing operation corresponding to the action state.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0014] At least one processor; and

[0015] A memory that is communicatively connected to at least one processor; wherein,

[0016] The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to perform the control method based on EEG signals according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the control method based on electroencephalogram (EEG) signals of any embodiment of the present invention.

[0018] The technical solution of this invention involves acquiring the electroencephalogram (EEG) signal of a target object and extracting target category identification feature parameters corresponding to the target category identification features in the EEG signal. Based on the target category identification feature parameters and target category reference parameters, the artifact identification result in the EEG signal is determined. The target category identification features and target category reference parameters are pre-determined based on the feature classification results of the sample EEG signal. Based on the artifact identification result, the action state of the target object is determined, and the corresponding processing operation is executed. By classifying signal artifacts in the EEG signal using pre-determined target category identification features and target category reference parameters, the complexity of the algorithm is reduced, the problem of existing electromyography (EMG) classification algorithms being unable to be embedded in chips is solved, and control operations based on EEG signal artifact categories are realized.

[0019] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0020] The accompanying drawings, which form part of this specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.

[0021] The invention will be more clearly understood with reference to the accompanying drawings and the following detailed description, wherein:

[0022] Figure 1 This is a flowchart of a control method based on electroencephalogram (EEG) signals provided in Embodiment 1 of the present invention;

[0023] Figure 2 This is a flowchart of the target feature extraction method and target category reference parameters provided in Embodiment 2 of the present invention;

[0024] Figure 3 This is a flowchart of a control method based on electroencephalogram (EEG) signals provided in Embodiment 2 of the present invention;

[0025] Figure 4 This is a schematic diagram of the structure of a control device based on electroencephalogram (EEG) signals provided in Embodiment 3 of the present invention;

[0026] Figure 5 This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention; Detailed Implementation

[0027] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention.

[0028] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0029] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use.

[0030] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, they should be considered part of the specification.

[0031] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0032] Embodiments of this invention can be applied to computer systems / servers that can operate with a wide range of other general-purpose or special-purpose computing system environments or configurations. Examples of well-known computing systems, environments, and / or configurations suitable for use with computer systems / servers include, but are not limited to: personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputer systems, mainframe computer systems, and distributed cloud computing environments that include any of the above systems, etc.

[0033] Computer systems / servers can be described in the general context of computer system executable instructions (such as program modules) executed by the computer system. Typically, program modules can include routines, programs, object programs, components, logic, data structures, etc., which perform specific tasks or implement specific abstract data types. Computer systems / servers can be implemented in distributed cloud computing environments, where tasks are performed by remote processing devices linked through a communication network. In distributed cloud computing environments, program modules can reside on local or remote computing system storage media, including storage devices.

[0034] Example 1

[0035] Figure 1 This is a flowchart of a control method based on electroencephalogram (EEG) signals provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where control is performed based on EEG signals. The method can be executed by an EEG signal-based control device and / or an EEG signal processing system. The EEG signal-based control device and / or EEG signal-based control system can be implemented in hardware and / or software. The EEG signal-based control device can be configured in the electronic device provided in this embodiment of the invention. Figure 1 As shown, the method includes:

[0036] S110. Obtain the EEG signal of the target object and extract the target category discrimination feature parameters corresponding to the target category discrimination features in the EEG signal.

[0037] The target object can be the object performing the control operation. For example, if the movement of an actuator needs to be controlled based on the user's EEG signals, then the target object can be the user. The EEG signal can be an EEG signal collected by an EEG signal acquisition device containing signal artifacts. Signal artifacts are technical / biological artifacts generated during EEG signal acquisition due to errors in recording settings, the good conductivity of the scalp, etc. Examples include power line interference, electrical signals generated by blinking, and muscle activity, which are mixed with the EEG signal and interfere with its acquisition. It should be noted that this embodiment only classifies biological artifacts within the EEG signal.

[0038] The EEG signal acquisition device can be, but is not limited to, a brain-computer interface device; there is no limitation in this regard. In this embodiment, EEG signals are acquired using the EEG signal acquisition device to obtain EEG signals to be classified. Then, signal artifacts in the acquired EEG signals are classified to obtain the categories of signal artifacts in the EEG signals. By classifying the electromyographic artifacts in the acquired EEG signals, the user's facial expressions, emotional state, etc., can be determined based on the categories of electromyographic artifacts.

[0039] Target category discrimination features refer to pre-defined features that can clearly distinguish between target and non-target categories of deliberate artifacts. These features are used to identify the category of signal artifacts in EEG signals. Optionally, the target category can be at least one of body electromyography (EMG), facial EMG, and ocular EMG. Taking facial EMG as the target category as an example, facial category discrimination feature parameters corresponding to the facial category discrimination features in the EEG signal can be extracted, and the facial EMG category of the EEG deliberate artifact can be identified based on the extracted facial category discrimination feature parameters.

[0040] In this embodiment, the parameters corresponding to the target category discrimination features in the EEG signal can be directly extracted as target category discrimination feature parameters. The extracted parameters can also be further processed to obtain target category discrimination feature parameters.

[0041] In one implementation, target category discrimination feature parameters corresponding to target category discrimination features in the EEG signal are extracted, including: extracting target category discrimination features from the EEG signal to obtain extracted EEG features; and performing dimensionality reduction processing on the extracted EEG features to obtain target category discrimination feature parameters. Optionally, the extracted parameters can be used as the extracted EEG features for dimensionality reduction processing, reducing the extracted EEG features to a low-dimensional space to obtain target category discrimination feature parameters, and then using the target category discrimination parameters in the low-dimensional space to identify EEG violations. The low-dimensional space can be, but is not limited to, two-dimensional space, three-dimensional space, etc., and is not limited thereto. The dimensionality reduction processing methods can be projection, non-negative matrix factorization, correlation analysis, etc., and are not limited here.

[0042] S120. Determine the violation identification result in the EEG signal based on the target category identification feature parameter and the target category reference parameter, wherein the target category identification feature and the target category reference parameter are determined in advance based on the feature classification result of the sample EEG signal.

[0043] After determining the target category identification feature parameters, the distance between the target category identification feature parameters and the target category reference parameters is calculated. The multi-functional deception identification result is then determined based on the distance between the target category identification feature parameters and the target category reference parameters. Optionally, the target category reference parameters may include category reference parameters and non-category reference parameters. In this embodiment, the target category identification feature parameters are reduced to a low-dimensional space, and the deception identification result is determined based on the distances between the target category identification feature parameters and the category reference parameters and non-category reference parameters in the low-dimensional space, respectively.

[0044] Specifically, the distance between the target category discrimination feature parameter and each target category reference parameter can be calculated. The category corresponding to the target reference parameter with the smallest distance between the target category discrimination feature and the target category reference parameter is taken as the violation identification result in the collected EEG signal. Violation identification results may include, but are not limited to, facial electromyography (EMG) categories, ocular EMG categories, body EMG categories, etc., and are not limited here.

[0045] In one embodiment of the present invention, the target category reference parameter may include a category reference parameter corresponding to the target category and a non-category reference parameter corresponding to the non-target category. Determining the deliberate violation identification result in the EEG signal based on the target category identification feature parameter and the target category reference parameter includes: determining a first distance between the target category identification feature parameter and the category reference parameter, and a second distance between the target category identification feature parameter and the non-category reference parameter; and taking the category corresponding to the smaller distance between the first distance and the second distance as the deliberate violation identification result.

[0046] The target category reference parameters include category reference parameters corresponding to the target category and non-category reference parameters corresponding to non-target categories. The category reference parameters corresponding to the target category can be either those corresponding to facial electromyography (EMG) or ocular EMG categories; this is not limited. The target category reference parameters can be facial EMG or ocular EMG reference parameters; correspondingly, the non-category reference parameters are non-facial EMG or non-ocular EMG reference parameters. In this embodiment, the distance between the target category discrimination feature parameter and the category reference parameter is calculated as the first distance, and the distance between the target category discrimination feature parameter and the non-category reference parameter is calculated as the second distance. The first distance and the second distance are compared. If the first distance is greater than the second distance, the category corresponding to the second distance is taken as the deception detection result; if the first distance is less than the second distance, the category corresponding to the first distance is taken as the deception detection result. It should be noted that each target category discrimination feature parameter corresponds to only one category; therefore, the first distance and the second distance are not necessarily equal. The first distance and the second distance can be calculated based on, but are not limited to, Euclidean distance; this is not limited.

[0047] In this embodiment, target category identification features and target category reference parameters are pre-determined based on the feature classification results of the sample EEG signals. When identifying the target category of the EEG signals, target category identification feature parameters are extracted from the target category identification features. These target category identification features can be directly selected from a large number of extracted features, or they can be extracted using different feature extraction methods. One or more of these methods are selected as the target feature extraction method, and the target category identification feature parameters of the EEG signals are extracted using this method. The target feature extraction method can be an EEG signal feature extraction method based on wavelet transform, an EEG signal feature extraction method based on fast Fourier transform, or an EEG signal feature extraction method based on frequency domain feature indices; no limitation is made here.

[0048] Based on the above embodiments, optionally, the determination of the target category identification features and the target category reference parameters includes:

[0049] The sample EEG signal and its label category are obtained, wherein the label category of the sample EEG signal is determined based on the category of signal artifacts in the sample EEG signal;

[0050] Extract sample category discrimination feature parameters from the sample EEG signal to obtain sample EEG feature points;

[0051] Based on the labeling category of each sample's EEG signal and the sample EEG feature points of each sample's EEG signal, determine the category reference point location information corresponding to each labeling category;

[0052] The distance between marker categories is determined based on the location information of the category reference points corresponding to each marker category;

[0053] When the distance between labeled categories is greater than a set threshold, the sample category identification feature is used as the target category identification feature, and the category reference point location information corresponding to each labeled category is used as the target category reference parameter.

[0054] When the distance between the labeled categories is not greater than the set threshold, repeat the above steps until the distance between each labeled category determined based on the extracted sample category identification feature parameters is greater than the set threshold. Then, use the sample category identification features as the target category identification features and use the category reference point location information corresponding to each current labeled category as the target category reference parameter.

[0055] Overall, by repeatedly extracting sample category discrimination feature parameters from a set of sample category discrimination features, the distance between labeled categories determined based on the extracted sample category discrimination feature parameters is determined. Based on the distance between labeled categories, a set of sample category discrimination features is selected from multiple sets of sample category discrimination features as the target category discrimination features. The parameters of the labeled categories corresponding to the target category discrimination features are used as the target category reference parameters. Optionally, sample category discrimination feature parameters for multiple sets of sample category discrimination features can be extracted using various different feature extraction methods to achieve the selection of different sample category discrimination features. When extracting multiple sets of sample category discrimination features using different feature extraction methods, after determining the target category discrimination features, the feature extraction method corresponding to the target category discrimination features is used as the target extraction method. When identifying EEG violation categories, the target category discrimination features can be directly extracted using the target extraction method.

[0056] Here, the sample EEG signals refer to a preset number of EEG signals containing different signal artifacts, used to determine the target feature extraction method and target category reference parameters. The preset number is set by those skilled in the art based on experience or needs, and is not limited here. In this embodiment, a preset number of EEG signals are collected using an EEG signal acquisition device to obtain sample EEG signals. The label category of each sample EEG signal is determined based on the category of signal artifacts in the sample EEG signals. Optionally, the label category can be facial electromyography (EMG) category and non-facial EMG category, ocular EMG category and non-ocular EMG category, or body EMG category and non-body EMG category; there is no limitation in this regard.

[0057] This embodiment determines the sample category discrimination feature by the distance between labeled samples, and then uses the feature value of the target category discrimination feature parameter of the target category discrimination feature corresponding to the labeled sample as the target category reference parameter. Correspondingly, the target category reference parameter is determined based on the spatial distribution region of the sample EEG feature points corresponding to each labeled category, including: for each labeled category, using the coordinate feature value of the sample EEG feature point corresponding to the labeled category as the target category reference parameter. For example, for each labeled category, the extracted discrimination feature parameter of the target category discrimination feature of the sample EEG signal in that labeled category is determined, and the extracted discrimination feature parameter is reduced to a low-dimensional space to obtain the target category discrimination feature parameter. Each target category discrimination feature parameter corresponds to a sample feature point, and then the feature values ​​of all sample feature points in that labeled category are used as the target category reference parameter. The coordinate feature value refers to the coordinate value that can represent the feature point feature corresponding to each labeled category. For example, the coordinate feature value can be the coordinate value of the centroid point, the coordinate value of the center point, the variance value, etc., of the sample EEG feature points corresponding to each labeled category.

[0058] In one implementation, the coordinate feature values ​​of the sample EEG feature points corresponding to the labeled category are used as the target category reference parameters, including: using the average coordinate value of the sample EEG feature points corresponding to the labeled category as the target category reference parameters. In the above example, the extracted discriminative feature parameters can be reduced to a three-dimensional space, the target category discriminative feature parameters can be used as coordinate points in the three-dimensional space, and the average coordinate value of the sample EEG feature points can be directly used as the target category reference parameters for that labeled category.

[0059] In another embodiment of the present invention, target category identification features and target category reference parameters can be determined based on the spatial distribution regions of different marker categories. Specifically, after mapping the sample feature points corresponding to each marker category to three-dimensional space, the spatial distribution region corresponding to each marker category is determined. When the spatial distribution regions corresponding to different marker categories meet the set conditions, the current sample category identification feature is used as the target category identification feature, and the corresponding target category identification parameters are determined.

[0060] Optionally, the distribution condition can be set such that there are no overlapping regions between the spatial distribution areas of EEG feature points corresponding to different label categories. One method for determining whether there are overlapping regions is to connect the edge EEG feature points of each spatial distribution area to form closed regions. The overlap between these closed regions is then used to determine whether there are overlapping regions between the spatial distribution areas. If the closed regions overlap, then there are overlapping regions between the spatial distribution areas; otherwise, there are no overlapping regions. Alternatively, the center of each spatial distribution area can be determined first, and then the distance from all sample EEG feature points to each center point can be calculated. The system can then determine whether there are sample EEG feature points of non-corresponding label categories whose distance to their corresponding center point is less than the distance from sample feature points of the corresponding label category to their corresponding center point. If such a distance exists, then there are overlapping regions between the spatial distribution areas; otherwise, there are no overlapping regions.

[0061] Optionally, the distribution condition can also be that the distance between the center points of the spatial distribution areas of EEG feature points corresponding to different label categories is greater than a preset center point distance. Specifically, the method for determining whether the distribution condition is met can be to first determine the center point of the spatial distribution area corresponding to each label category, and then calculate the distance between any two center points. When the distance between any two center points is greater than the preset center point distance, the spatial distribution area of ​​the EEG feature points corresponding to each label category meets the distribution condition. Alternatively, the distribution condition can be that the distance between any two feature points belonging to different label categories in the spatial distribution area is greater than a preset distance. Specifically, the method for determining whether the distribution condition is met can be to arbitrarily select two feature points belonging to different spatial distribution areas, calculate the distance between the two feature points, and when the distance between the two feature points is greater than the preset distance, the spatial distribution area of ​​the EEG feature points corresponding to each label category meets the distribution condition. It should be noted that the preset center point distance and the preset distance are set by those skilled in the art based on experience and needs, and are not limited here.

[0062] In one embodiment of the present invention, the target categories are body electromyography (EMG), facial EMG, and ocular EMG. Determining the violation identification result in the EEG signal based on target category identification features and target category reference parameters includes: determining the facial EMG violation identification result in the EEG signal based on facial EMG category identification features and facial EMG category reference parameters; when the facial EMG violation identification result is not a facial EMG violation category, determining the violation identification result as body EMG; when the facial EMG violation identification result is a facial EMG violation category, determining the ocular EMG violation identification result in the EEG signal based on ocular EMG category identification features and ocular EMG category reference parameters; when the ocular EMG violation identification result is not an ocular EMG violation category, determining the violation identification result as facial EMG; and when the ocular EMG violation identification result is an ocular EMG violation category, determining the violation identification result as ocular EMG. In other words, by sequentially identifying based on facial EMG category identification features and ocular EMG category identification features, any one of the target category identification results can be obtained.

[0063] S130. Determine the action state of the target object based on the violation identification result, and execute the processing operation corresponding to the action state.

[0064] In this embodiment, the action state of the target object is determined by the violation detection results in the electroencephalogram (EEG) signal, and then the corresponding operation is executed based on the action state of the target object. It is understood that the operation corresponding to the action state can be determined according to business requirements. When the business requirement is to collect EEG signals in a set state, a prompt message can be generated to prompt the user to execute the corresponding state; when the business requirement is to perform control based on EEG signals, the operation corresponding to the action state can be executed.

[0065] In one embodiment of the present invention, determining the action state of a target object based on the evidence detection result includes: when the evidence detection result is a body electromyography (EMG) category, determining the target object's action state as an excited state; when the evidence detection result is a facial EMG category, determining the target object's action state as an unfocused state; and when the evidence detection result is an ocular EMG category, determining the target object's action state as a meditative state. When the evidence detection category is a body EMG category, indicating active body muscles, the target object's action state can be determined as an excited state; when the evidence detection category is a facial EMG category, indicating active facial muscles, the target object's action state can be determined as an unfocused state; and when the evidence detection category is an ocular EMG category, indicating active ocular muscles, the target object's action state can be determined as a meditative state.

[0066] Based on the above scheme, the processing operation corresponding to the action state is performed, including: determining posture adjustment prompts based on the movement state and outputting the posture adjustment prompts; collecting the EEG signal of the target object after performing the posture adjustment method corresponding to the posture adjustment prompts as the target EEG signal. In this embodiment, the business requirement is to collect EEG signals in a resting state. Therefore, posture adjustment prompts are generated and output (displayed or broadcast) based on the movement state of the target object. Then, after the user adjusts their posture according to the posture adjustment prompts, the resting EEG signal is collected. For example, assuming the movement state is an excited state, a posture adjustment prompt of "Please remain still" can be generated and output; assuming the movement state is an unfocused state, a posture adjustment prompt of "Please stay focused" can be generated and output; assuming the movement state is a meditative state, a posture adjustment prompt of "Please look straight ahead and remain still" can be generated and output.

[0067] Based on the above scheme, the processing operation corresponding to the action state is executed, including: determining the action execution mechanism corresponding to the action state, and controlling the action execution mechanism to perform the corresponding motion operation. In this embodiment, the business requirement is motion control based on EEG signals, such as controlling a brain-computer swing mechanism based on EEG signals. The opening and closing state corresponding to each action state can be preset, the target opening and closing state is determined based on the action state of the target object, and the opening and closing degree of the brain-computer swing mechanism is adjusted based on the target opening and closing state. For example, the opening and closing state corresponding to the excited state can be set to a large opening and closing state, the opening and closing state corresponding to the unfocused state to a small opening and closing state, and the opening and closing state corresponding to the meditative state to a swaying state. Then, when the motion state is excited, a large opening and closing operation is executed; when the motion state is unfocused, a small opening and closing operation is executed; and when the motion state is meditative, a swaying operation is executed. The specific execution methods of the large opening and closing operation, the small opening and closing operation, and the swaying operation can be preset and are not limited here.

[0068] It should be noted that the control method based on EEG signals provided in this embodiment can be executed by a brain-computer interface device or by a separate processing device. For example, the brain-computer interface device can collect EEG signals, and the processing device can execute the control method based on EEG signals provided in any embodiment of the present invention to generate control commands and perform corresponding control operations. The execution of the control operations can be performed by a separate device or by the brain-computer interface device. When the business requirement is motion control based on EEG signals, the brain-computer interface device can collect and process the EEG signals to obtain control commands, and then send the control commands to the execution mechanism to cause the execution mechanism to perform the corresponding operation.

[0069] The technical solution of this embodiment acquires the EEG signal of the target object and extracts the target category discrimination feature parameters corresponding to the target category discrimination features in the EEG signal. Based on the target category discrimination feature parameters and target category reference parameters, the artifact identification result in the EEG signal is determined. The target category discrimination features and target category reference parameters are pre-determined based on the feature classification results of the sample EEG signal. Based on the artifact identification result, the action state of the target object is determined, and the corresponding processing operation is executed. By classifying signal artifacts in the EEG signal using pre-determined target category discrimination features and target category reference parameters, the complexity of the algorithm is reduced, the problem of existing EMG classification algorithms being unable to be embedded in chips is solved, and control operations based on the category of EEG signal artifacts are realized.

[0070] Example 2

[0071] This embodiment provides a preferred embodiment. In this embodiment, facial electromyography, ocular electromyography, and body electromyography are used as examples of signal artifacts in electroencephalogram (EEG) signals to illustrate the target feature extraction method, the determination process of target category reference parameters, and the classification process of EEG signal artifacts.

[0072] Figure 2 This is a flowchart illustrating the target feature extraction method and target category reference parameters provided in Embodiment 2 of the present invention. Figure 2As shown, firstly, a large number of collected EEG signals mixed with different EMG artifacts are used as sample EEG signals. Feature extraction is performed on the sample EEG signals to obtain extracted features, and the extracted features are preprocessed. Then, based on the current facial EMG feature extraction method, the extracted features are filtered into facial EMG features and non-facial EMG features. Then, the filtered features are subjected to data dimensionality reduction processing to obtain the spatial distribution regions of facial EMG feature points and non-facial EMG feature points. It is determined whether the spatial distribution regions of facial EMG feature points and non-facial EMG feature points do not overlap. If there is no overlap, the current facial EMG feature extraction method is determined as Fa, and the centroid A of facial EMG feature points and the centroid B of non-facial EMG feature points are determined as facial EMG reference parameters. If there is overlap, a new facial EMG feature extraction method is obtained to re-filter facial EMG features and non-facial EMG features, and the above process continues until the spatial distribution regions of facial EMG feature points and non-facial EMG feature points do not overlap. Similarly, for the extracted features after data preprocessing, the ocular electromyography feature extraction method Fb is determined based on the same process, and the centroid C of ocular electromyography feature points and the centroid D of non-ocular electromyography feature points are determined as ocular electromyography reference parameters.

[0073] It should be noted that, Figure 2 The facial electromyography (EMG) feature extraction method Fa is used to distinguish between facial EMG features and non-facial EMG features, while the ocular EMG feature extraction method Fb is used to distinguish between ocular EMG features and non-ocular EMG features. Therefore, the facial EMG feature extraction method Fa differs from the ocular EMG feature extraction method Fb. Thus, to accurately determine the category of EEG violation, this embodiment uses the corresponding feature extraction method based on the target category.

[0074] Figure 3 This is a flowchart of a control method based on electroencephalogram (EEG) signals provided in Embodiment 2 of the present invention. Figure 3 As shown, based on the established facial electromyography (EMG) feature extraction method Fa and ocular EMG feature extraction method Fb, target category identification features (i.e., facial EMG category features and ocular EMG category features) are extracted. The target category identification features are preprocessed and the data dimensionality is reduced to obtain target category identification feature points. It is determined whether the distance between the target category identification feature point and the centroid B is less than the distance between the target category identification feature point and the centroid A. If so, the deliberate violation identification result is the body EMG category; if not, it is determined whether the distance between the target category identification feature point and the centroid C is less than the distance between the target category identification feature point and the centroid D. If so, the deliberate violation identification result is the ocular EMG category; if not, the deliberate violation identification result is the facial EMG category.

[0075] After determining the result of the violation identification, the corresponding operation is performed based on the violation identification structure. The specific operation can be referred to the above embodiment, and will not be repeated here.

[0076] The technical solution of this embodiment classifies electromyographic artifacts in EEG signals into facial electromyographic categories, ocular electromyographic categories, and body electromyographic categories, thereby facilitating the execution of corresponding operations based on the artifact category.

[0077] Example 3

[0078] Figure 4 This is a schematic diagram of a control device based on electroencephalogram (EEG) signals provided in Embodiment 3 of the present invention. Figure 4 As shown, the device includes:

[0079] The feature extraction module 310 is used to acquire the electroencephalogram (EEG) signal of the target object and extract the target category discrimination feature parameters corresponding to the target category discrimination features in the EEG signal.

[0080] The intrusion detection result determination module 320 is used to determine the intrusion detection result in the EEG signal based on the target category identification feature parameters and the target category reference parameters, wherein the target category identification features and the target category reference parameters are determined in advance based on the feature classification results of the sample EEG signal;

[0081] The processing operation execution module 330 is used to determine the action state of the target object based on the violation identification result and execute the processing operation corresponding to the action state.

[0082] Based on the above embodiments, optionally, the feature extraction module 310 is specifically used for:

[0083] Feature extraction of target category discrimination features is performed on EEG signals to obtain extracted EEG features;

[0084] The extracted EEG features are then subjected to dimensionality reduction processing to obtain target category discrimination feature parameters.

[0085] Based on the above embodiments, optionally, the target category reference parameters include category reference parameters corresponding to the target category and non-category reference parameters corresponding to non-target categories. The violation identification result determination module 320 is specifically used for:

[0086] Determine the first distance between the target category discrimination feature parameter and the category reference parameter, and the second distance between the target category discrimination feature parameter and the non-category reference parameter;

[0087] The category corresponding to the smaller distance between the first and second distances is taken as the result of the violation identification.

[0088] Based on the above embodiments, optionally, the device further includes a feature parameter determination module, used for:

[0089] The sample EEG signal and its label category are obtained, wherein the label category of the sample EEG signal is determined based on the category of signal artifacts in the sample EEG signal;

[0090] Extract sample category discrimination feature parameters from the sample EEG signal to obtain sample EEG feature points;

[0091] Based on the labeling category of each sample's EEG signal and the sample EEG feature points of each sample's EEG signal, determine the category reference point location information corresponding to each labeling category;

[0092] The distance between marker categories is determined based on the location information of the category reference points corresponding to each marker category;

[0093] When the distance between labeled categories is greater than a set threshold, the sample category identification feature is used as the target category identification feature, and the category reference point location information corresponding to each labeled category is used as the target category reference parameter.

[0094] When the distance between the labeled categories is not greater than the set threshold, repeat the above steps until the distance between each labeled category determined based on the extracted sample category identification feature parameters is greater than the set threshold. Then, use the sample category identification features as the target category identification features and use the category reference point location information corresponding to each current labeled category as the target category reference parameter.

[0095] Based on the above embodiments, optionally, the target category reference parameter determination unit is used to use the coordinate feature value of the sample EEG feature point corresponding to the label category as the target category reference parameter for each label category.

[0096] Based on the above embodiments, optionally, the target category reference parameter determination unit is used to use the average coordinate of the sample EEG feature points corresponding to the labeled category as the target category reference parameter corresponding to the labeled category.

[0097] Based on the above embodiments, optionally, the target category is body electromyography (EMG) category, facial EMG category, and ocular EMG category, and the violation identification result determination module 320 is specifically used for:

[0098] The facial electromyography (EMG) violation identification results in EEG signals are determined based on facial EMG category identification features and facial EMG category reference parameters.

[0099] When the facial electromyography (EMG) deliberation result is classified as a non-facial EMG deliberation category, the deliberation result is determined to be a body EMG category.

[0100] When the facial electromyography (EMG) deliberation result is classified as a facial EMG deliberation category, the EMG deliberation result in the EEG signal is determined based on the EMG category identification characteristics and EMG category reference parameters. When the EMG deliberation result is classified as a non-EMG deliberation category, the deliberation result is determined to be a facial EMG category. When the EMG deliberation result is classified as an EMG deliberation category, the deliberation result is determined to be an EMG deliberation category.

[0101] Based on the above embodiments, optionally, the processing operation execution module 330 is specifically used for:

[0102] When the violation identification result is the body electromyography category, the target object's action state is determined to be an excited state;

[0103] When the violation identification result is facial electromyography category, the target object's movement state is determined to be an unfocused state;

[0104] When the violation identification result is in the ocular electromyography category, the target object's movement state is determined to be a meditative state.

[0105] Based on the above embodiments, optionally, the processing operation execution module 330 is specifically used for:

[0106] Determine and output posture adjustment prompts based on the motion state;

[0107] The target EEG signal is collected after the target object performs the posture adjustment method corresponding to the posture adjustment prompt information.

[0108] Based on the above embodiments, optionally, the processing operation execution module 330 is specifically used for:

[0109] Determine the action execution mechanism corresponding to the action state, and control the action execution mechanism to perform the corresponding motion operation.

[0110] The control device based on electroencephalogram (EEG) signals provided in the embodiments of the present invention can execute the control method based on EEG signals provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0111] Example 4

[0112] Figure 5This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0113] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0114] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0115] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as control methods based on electroencephalogram (EEG) signals.

[0116] In some embodiments, the EEG-based control method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the EEG-based control method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the EEG-based control method by any other suitable means (e.g., by means of firmware).

[0117] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0118] Computer programs for implementing the EEG-based control method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0119] Example 5

[0120] Embodiment 5 of the present invention also provides a computer-readable storage medium storing computer instructions for causing a processor to execute a control method based on electroencephalogram (EEG) signals, the method comprising:

[0121] Acquire the electroencephalogram (EEG) signal of the target object and extract the target category discrimination feature parameters corresponding to the target category discrimination features in the EEG signal;

[0122] The result of violation identification in EEG signals is determined based on the target category identification feature parameters and the target category reference parameters, wherein the target category identification features and the target category reference parameters are determined in advance based on the feature classification results of the sample EEG signals;

[0123] Based on the violation identification results, the action state of the target object is determined, and the corresponding processing operation is executed.

[0124] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0125] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0126] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0127] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0128] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0129] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For system embodiments, since they largely correspond to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0130] The methods and systems of the present invention may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the methods is for illustrative purposes only, and the steps of the methods of the present invention are not limited to the order specifically described above unless otherwise specifically stated. Furthermore, in some embodiments, the present invention may also be implemented as a program recorded on a recording medium, the program comprising machine-readable instructions for implementing the methods according to the present invention. Thus, the present invention also covers recording media storing programs for performing the methods according to the present invention.

[0131] The description of this invention is given for illustrative and descriptive purposes only and is not intended to be exhaustive or to limit the invention to the forms disclosed. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of the invention and to enable those skilled in the art to understand the invention and to design various embodiments with various modifications suitable for a particular purpose.

Claims

1. A control method based on electroencephalogram (EEG) signals, characterized in that, include: The brainwave signals of the target object are acquired, and the target category discrimination feature parameters corresponding to the target category discrimination features in the brainwave signals are extracted; wherein, the target category is at least one of body electromyography category, facial electromyography category and ocular electromyography category; The artifact identification result in the EEG signal is determined based on the target category identification feature parameter and the target category reference parameter, wherein the target category identification feature and the target category reference parameter are determined in advance based on the feature classification result of the sample EEG signal; wherein the sample EEG signal is the EEG signal containing different signal artifacts; Based on the artifact identification result, determine the action state of the target object and execute the processing operation corresponding to the action state; The extraction of target category discrimination feature parameters corresponding to the target category discrimination features in the electroencephalogram (EEG) signal includes: The target category discrimination features are extracted from the EEG signal to obtain the extracted EEG features; The extracted EEG features are subjected to dimensionality reduction processing to obtain the target category discrimination feature parameters; The target category reference parameters include category reference parameters corresponding to the target category and non-category reference parameters corresponding to non-target categories. Determining the artifact identification result in the EEG signal based on the target category discrimination feature parameters and the target category reference parameters includes: Determine a first distance between the target category discrimination feature parameter and the category reference parameter, and a second distance between the target category discrimination feature parameter and the non-category reference parameter; The category corresponding to the smaller distance between the first distance and the second distance is taken as the artifact identification result.

2. The method according to claim 1, characterized in that, The determination of the target category identification features and the target category reference parameters includes: Acquire sample EEG signals and the labeling categories of the sample EEG signals, wherein the labeling categories of the sample EEG signals are determined based on the categories of signal artifacts in the sample EEG signals; Extract the sample category discrimination feature parameters from the sample EEG signal to obtain the sample EEG feature points of the sample EEG signal; Based on the labeling category of each sample EEG signal and the sample EEG feature points of each sample EEG signal, determine the category reference point location information corresponding to each labeling category; The distance between the marker categories is determined based on the category reference point location information corresponding to each of the marker categories; When the distance between the labeled categories is greater than a set threshold, the sample category identification feature is used as the target category identification feature, and the category reference point position information corresponding to each labeled category is used as the target category reference parameter. When the distance between the labeled categories is not greater than a set threshold, repeat the above steps until the distance between each labeled category determined based on the extracted sample category identification feature parameters is greater than the set threshold. Then, use the sample category identification feature as the target category identification feature and use the category reference point location information corresponding to each of the current labeled categories as the target category reference parameter.

3. The method according to claim 2, characterized in that, The determination of the target category reference parameters based on the spatial distribution region of sample EEG feature points corresponding to each of the labeled categories includes: For each of the labeled categories, the coordinate feature values ​​of the sample EEG feature points corresponding to the labeled category are used as the target category reference parameters corresponding to the labeled category.

4. The method according to claim 1, characterized in that, The processing operation corresponding to the action state includes: Determine posture adjustment prompts based on the action state and output the posture adjustment prompts. The target EEG signal is collected after the target object performs the posture adjustment method corresponding to the posture adjustment prompt information.

5. The method according to claim 1, characterized in that, The processing operation corresponding to the action state includes: Determine the action execution mechanism corresponding to the action state, and control the action execution mechanism to perform the corresponding motion operation.

6. A control device based on electroencephalogram (EEG) signals, characterized in that, include: The feature extraction module is used to acquire the electroencephalogram (EEG) signal of the target object and extract the target category identification feature parameters corresponding to the target category identification features in the EEG signal; wherein, the target category is at least one of body electromyography (EMG), facial EMG, and ocular EMG. The artifact identification result determination module is used to determine the artifact identification result in the EEG signal based on the target category identification feature parameter and the target category reference parameter, wherein the target category identification feature and the target category reference parameter are determined in advance based on the feature classification result of the sample EEG signal; wherein the sample EEG signal is the EEG signal containing different signal artifacts; The processing operation execution module is used to determine the action state of the target object based on the artifact identification result and execute the processing operation corresponding to the action state. The discriminative feature extraction module is specifically used for: Feature extraction of target category discrimination features is performed on EEG signals to obtain extracted EEG features; The extracted EEG features are subjected to dimensionality reduction processing to obtain target category discrimination feature parameters; The target category reference parameters include category reference parameters corresponding to the target category and non-category reference parameters corresponding to non-target categories. The artifact identification result determination module is specifically used for: Determine the first distance between the target category discrimination feature parameter and the category reference parameter, and the second distance between the target category discrimination feature parameter and the non-category reference parameter; The category corresponding to the smaller distance between the first and second distances is taken as the artifact identification result.

7. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the control method based on EEG signals as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the control method based on EEG signals as described in any one of claims 1-5.

Images