Object prediction method and apparatus based on electroencephalography data

Abstract

The application discloses an object prediction method based on electroencephalogram data. Target electroencephalogram data of a user collected by an electroencephalogram collection device from a starting moment of a current collection period to a current moment is acquired, and the target electroencephalogram data is used to predict a target prediction object of a user's gaze from a plurality of preconfigured visual stimulation objects. The confidence of the target prediction object is determined, and in the case that the confidence of the target prediction object is lower than or equal to a preset confidence threshold and the time length from the starting moment to the current moment is less than an upper limit time length, the current moment is updated, and the step of acquiring the target electroencephalogram data of the user collected by the electroencephalogram collection device from the starting moment of the collection period to the current moment is executed until the target prediction object with the confidence higher than the confidence threshold is obtained as an output prediction object. The object prediction efficiency is improved by the object prediction method based on the electroencephalogram data.

Description

Technical Field

[0001] This invention relates to the field of human-computer interaction, and in particular to a method and apparatus for object prediction based on electroencephalogram (EEG) data. Background Technology

[0002] Brain-computer interfaces (BCIs) can express a user's intentions without relying on surrounding muscles and nerves, and can control external devices without physical activity. Visual evoked potentials (VEPs) are evoked EEG signals, primarily located in the parietal and occipital regions of the human brain, and are induced by visual stimuli. VEP-based BCIs are the main category of non-invasive BCIs. VEPs can be easily detected when the subject fixates on the stimulus; therefore, VEP-based BCIs are suitable and sufficiently stable as communication systems.

[0003] In related technologies, when BCI based on VEP predicts the visual stimulus object that the user is looking at, the fixed duration of EEG data corresponding to each visual stimulus object is different. Generally, the EEG data is collected based on the maximum fixed duration of EEG data corresponding to the visual stimulus object. Specifically, including visual stimuli A, B, and C, when predicting visual stimulus object A, at least 0.2 seconds of EEG data needs to be collected from the user; when predicting visual stimulus object B, at least 0.3 seconds of EEG data needs to be collected from the user; and when predicting visual stimulus object C, at least 0.25 seconds of EEG data needs to be collected from the user. Therefore, when the device predicts the visual stimulus object that the user is currently looking at, it generally collects at least 0.3 seconds of EEG data before making further predictions.

[0004] Because EEG data is collected for a fixed duration each time, and then the visual stimulus being predicted is further predicted, the prediction efficiency is low. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, apparatus, computer device, computer-readable storage medium, and computer program product based on EEG data that can improve object prediction in response to the above-mentioned technical problems.

[0006] Firstly, this application provides a method for object prediction based on electroencephalogram (EEG) data. The method includes:

[0007] Acquire the target EEG data of the user collected by the EEG acquisition device from the start of the current acquisition cycle to the current time;

[0008] Using the target EEG data, predict the target object that the user is looking at among a pre-configured set of visual stimuli.

[0009] Determine the confidence level of the target prediction object;

[0010] If the confidence level of the target prediction object is lower than or equal to a preset confidence threshold, and the duration from the start time to the current time is less than the upper limit duration, the current time is updated, and the process of acquiring the user's target EEG data collected by the EEG acquisition device from the start time of the acquisition cycle to the current time is returned. The steps of using the target EEG data to predict the target prediction object that the user is looking at among a number of pre-configured visual stimuli and determining the confidence level of the target prediction object are repeated until a target prediction object with a confidence level higher than the confidence threshold is obtained, which is then used as the output prediction object.

[0011] In one embodiment, the method further includes:

[0012] If the duration from the start time to the current time is greater than or equal to the upper limit duration, at the start of the next acquisition cycle, the step of acquiring the user's target EEG data acquired by the EEG acquisition device from the start time of the current acquisition cycle to the current time is executed.

[0013] In one embodiment, predicting the target object gazed upon by the user among a pre-configured set of visual stimuli using the target EEG data includes:

[0014] If the duration from the start time to the current time is greater than or equal to the lower limit duration, the target EEG data is used to predict the target object being gazed upon by the user among a pre-configured set of visual stimuli.

[0015] In one embodiment, the method further includes:

[0016] Based on the control instructions corresponding to the pre-configured visual stimulus objects, control instructions for the output prediction object are determined, and corresponding processing is performed according to the control instructions for the output prediction object.

[0017] In one embodiment, the step of predicting the target EEG data using a preset object prediction strategy to obtain the target prediction object includes:

[0018] The target EEG data is input into a spatial filter to obtain the denoised target EEG signal;

[0019] Based on the correlation analysis strategy, the correlation coefficient between the target EEG signal and the template EEG signal corresponding to each of the visual stimuli is determined;

[0020] The visual stimulus with the highest correlation coefficient is identified as the target prediction object.

[0021] In one embodiment, the method further includes:

[0022] During the calibration phase of the object prediction strategy, calibration EEG data acquired by the EEG acquisition device is received; the calibration EEG data is used to calibrate the parameters in the object prediction strategy.

[0023] Using a preset calibration strategy, the parameters in the object prediction strategy are calibrated based on the calibrated EEG data.

[0024] In one embodiment, calibrating the parameters in the object prediction strategy based on the calibrated EEG data using a preset calibration strategy includes:

[0025] Determine the availability of the calibrated EEG data;

[0026] If the availability rate is greater than a preset availability rate threshold, the availability rate will be displayed to the user.

[0027] Upon receiving the user's confirmation calibration instruction, the parameters in the object prediction strategy are calibrated based on the calibrated EEG data using a preset calibration strategy.

[0028] Upon receiving the user's recalibration instruction, the process returns to the step of receiving the calibration EEG data collected by the EEG acquisition device.

[0029] In one embodiment, when the target EEG data is based on coded modulation visual evoked potential data, receiving the calibrated EEG data acquired by the EEG acquisition device includes:

[0030] Select one visual stimulus object from the pre-configured plurality of visual stimulus objects as a calibration object, and control a preset visual stimulator to stimulate the calibration object.

[0031] Receive the calibration EEG data for the calibration object collected by the EEG acquisition device, and obtain the template EEG signal corresponding to the calibration object;

[0032] For each of the pre-configured visual stimulus objects, the template EEG signal corresponding to the calibration object is time-shifted according to a preset time-shifting strategy to obtain the template EEG signal corresponding to the other visual stimulus objects.

[0033] Secondly, this application also provides an object prediction system based on electroencephalogram (EEG) data. The system includes:

[0034] EEG acquisition equipment is used to collect users' EEG data;

[0035] The terminal is configured to acquire target EEG data of the user collected by the EEG acquisition device from the start of the current acquisition cycle to the current time; predict the target predicted object of the user's gaze among a pre-configured set of visual stimuli using the target EEG data; determine the confidence level of the target predicted object; and, if the confidence level of the target predicted object is lower than or equal to a preset confidence threshold and the duration from the start of the acquisition cycle to the current time is less than an upper limit duration, update the current time and return to execute the steps of acquiring the target EEG data of the user collected by the EEG acquisition device from the start of the acquisition cycle to the current time, predicting the target predicted object of the user's gaze among a pre-configured set of visual stimuli using the target EEG data, and determining the confidence level of the target predicted object, until a target predicted object with a confidence level higher than the confidence threshold is obtained, which is then used as the output predicted object.

[0036] Thirdly, this application also provides an object prediction device based on electroencephalogram (EEG) data. The device includes:

[0037] The receiving module is used to acquire the target EEG data collected by the EEG acquisition device from the start of the current acquisition cycle to the current time.

[0038] The prediction module is used to predict the target prediction object corresponding to the target EEG data among a preset number of objects using a preset object prediction strategy.

[0039] The determination module is used to determine the confidence level of the target prediction object;

[0040] The output module is used to update the current time and return to the steps of acquiring the user's target EEG data collected by the EEG acquisition device from the start time of the acquisition cycle to the current time, predicting the target prediction object that the user is looking at among a number of pre-configured visual stimuli using the target EEG data, and determining the confidence of the target prediction object, until a target prediction object with a confidence higher than the confidence threshold is obtained as the output prediction object.

[0041] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0042] Acquire the target EEG data of the user collected by the EEG acquisition device from the start of the current acquisition cycle to the current time;

[0043] Using the target EEG data, predict the target object that the user is looking at among a pre-configured set of visual stimuli.

[0044] Determine the confidence level of the target prediction object;

[0045] If the confidence level of the target prediction object is lower than or equal to a preset confidence threshold, and the duration from the start time to the current time is less than the upper limit duration, the current time is updated, and the process of acquiring the user's target EEG data collected by the EEG acquisition device from the start time of the acquisition cycle to the current time is returned. The steps of using the target EEG data to predict the target prediction object that the user is looking at among a number of pre-configured visual stimuli and determining the confidence level of the target prediction object are repeated until a target prediction object with a confidence level higher than the confidence threshold is obtained, which is then used as the output prediction object.

[0046] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0047] Acquire the target EEG data of the user collected by the EEG acquisition device from the start of the current acquisition cycle to the current time;

[0048] Using the target EEG data, predict the target object that the user is looking at among a pre-configured set of visual stimuli.

[0049] Determine the confidence level of the target prediction object;

[0050] If the confidence level of the target prediction object is lower than or equal to a preset confidence threshold, and the duration from the start time to the current time is less than the upper limit duration, the current time is updated, and the process of acquiring the user's target EEG data collected by the EEG acquisition device from the start time of the acquisition cycle to the current time is returned. The steps of using the target EEG data to predict the target prediction object that the user is looking at among a number of pre-configured visual stimuli and determining the confidence level of the target prediction object are repeated until a target prediction object with a confidence level higher than the confidence threshold is obtained, which is then used as the output prediction object.

[0051] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0052] Acquire the target EEG data of the user collected by the EEG acquisition device from the start of the current acquisition cycle to the current time;

[0053] Using the target EEG data, predict the target object that the user is looking at among a pre-configured set of visual stimuli.

[0054] Determine the confidence level of the target prediction object;

[0055] If the confidence level of the target prediction object is lower than or equal to a preset confidence threshold, and the duration from the start time to the current time is less than the upper limit duration, the current time is updated, and the process of acquiring the user's target EEG data collected by the EEG acquisition device from the start time of the acquisition cycle to the current time is returned. The steps of using the target EEG data to predict the target prediction object that the user is looking at among a number of pre-configured visual stimuli and determining the confidence level of the target prediction object are repeated until a target prediction object with a confidence level higher than the confidence threshold is obtained, which is then used as the output prediction object.

[0056] The aforementioned object prediction method, device, computer equipment, storage medium, and computer program product based on brain data simultaneously collects EEG data and performs object prediction based on the currently collected EEG data during each acquisition cycle. Compared with object prediction methods in related technologies, object prediction is more efficient.

[0057] 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

[0058] 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.

[0059] The invention will be more clearly understood with reference to the accompanying drawings and the following detailed description, wherein:

[0060] Figure 1 This is a schematic diagram of the structure of a visual brain-computer interface system in one embodiment;

[0061] Figure 2 This is a flowchart illustrating an object prediction method based on electroencephalogram (EEG) data in one embodiment.

[0062] Figure 3 This is a flowchart illustrating another object prediction method based on EEG data in one embodiment;

[0063] Figure 4 This is a structural block diagram of an object prediction device based on electroencephalogram (EEG) data in one embodiment.

[0064] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0065] 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. Furthermore, it should be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale.

[0066] 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.

[0067] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0068] 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.

[0069] In related technologies, the object prediction process based on VEP-based EEG data is generally as follows:

[0070] Step 1: At the start of the expected collection period, collect EEG data for a fixed duration;

[0071] Step 2: Preprocess the EEG data and extract its features;

[0072] Step 3: Classify the extracted features to obtain the object prediction results.

[0073] Visual evoked electroencephalogram (VEP) signals are primarily found in the parietal and occipital regions of the human brain and are induced by visual stimuli. In practical applications, visual brain-computer interfaces (BCIs) are typically modulated by providing different brightness stimuli at different time sequences (frames) based on the screen refresh rate on an LCD display. Building a VEP-based BCI mainly involves two parts: encoding and decoding. Encoding is the process of constructing visual stimuli, such as selecting the number of targets, encoding methods, and contrast settings. The decoding process depends on the encoding method; by decoding the VEP, the corresponding encoded information can be obtained, revealing the visual stimulus object currently being gazed upon by the subject.

[0074] Therefore, in order to obtain the corresponding encoded information by decoding VEP, related technologies generally require the acquisition of EEG data for all visual stimuli within a fixed duration. Specifically, this includes visual stimuli A, B, and C. When predicting visual stimuli A, at least 0.2 seconds of EEG data from the user is required; when predicting visual stimuli B, at least 0.3 seconds of EEG data is required; and when predicting visual stimuli C, at least 0.25 seconds of EEG data is required. Thus, when the device predicts the visual stimulus currently being viewed by the user, in order to acquire EEG data for all visual stimuli (or to avoid missing EEG data for some visual stimuli), at least 0.3 seconds of EEG data is typically acquired before further prediction.

[0075] While related technologies based on EEG data do not miss the EEG data of some visual stimuli, not all collected EEG data is helpful for the prediction of some visual stimuli. Therefore, collecting this part of the EEG data that is not useful for the prediction will reduce the efficiency of the prediction.

[0076] Based on this, this application proposes an object prediction method based on EEG data. The method involves acquiring the user's target EEG data collected by an EEG acquisition device from the start of the current acquisition cycle to the current time. Using this target EEG data, the method predicts the target object the user is looking at from a pre-configured set of visual stimuli. The method determines the confidence level of the target predicted object. If the confidence level of the target predicted object is lower than or equal to a preset confidence threshold, and the duration from the start of the acquisition cycle to the current time is less than an upper limit, the method updates the current time and returns to the previous steps of acquiring the user's target EEG data collected by the EEG acquisition device from the start of the acquisition cycle to the current time, predicting the target object the user is looking at from a pre-configured set of visual stimuli using the target EEG data, and determining the confidence level of the target predicted object. This process continues until a target predicted object with a confidence level higher than the confidence threshold is obtained, which is then used as the output predicted object.

[0077] The object prediction method based on EEG data proposed in this application collects EEG data and predicts objects based on the currently collected EEG data in each acquisition cycle. Therefore, when predicting visual stimulus objects, the collected target EEG data will not contain too much EEG data that is irrelevant to the final predicted visual stimulus object, thus improving the efficiency of instruction prediction.

[0078] The object prediction method based on EEG data provided in this application can be applied to, for example... Figure 1The illustrated object prediction system based on electroencephalogram (EEG) data includes a terminal 130 that communicates with an EEG acquisition device 110 via a communication connection. A visual stimulator 120 can be integrated with the terminal 130 and communicates with it via the same connection. The visual stimulator 120 can be any device capable of visual encoding, including cathode ray tube (CRT) displays, liquid crystal displays (LCDs), light-emitting diodes (LEDs), and organic light-emitting diodes (OLEDs). The terminal 130 can be, but is not limited to, various personal computers, laptops, smartphones, and tablets. The EEG acquisition device 110 can be, but is not limited to, wearable EEG acquisition devices.

[0079] To achieve precise EEG data synchronization, terminal 130 can use a photodiode as a capture device in the VEP BCI scenario, developed by Boruikang, utilizing the fast conduction characteristics of the PN (P-type semiconductor and N-type semiconductor) junction to achieve EEG signal synchronization. Terminal 130 can also establish a communication connection with EEG acquisition device 110 using Bluetooth Low Energy, reducing the EEG signal synchronization delay between EEG acquisition device 110 and terminal 130 to less than 1ms. The method for synchronizing EEG data between terminal 130 and EEG acquisition device 110 is not limited to the above methods; any method that achieves low-latency communication between the stimulator and the EEG acquisition device is acceptable, with a recommended one-way communication speed of less than 1ms.

[0080] The object prediction method based on EEG data provided in this application will be described in detail below.

[0081] In one embodiment, such as Figure 2 The image shows an object prediction method based on electroencephalogram (EEG) data, as illustrated in this application. This method is applied to… Figure 1 Taking terminal 130 as an example, the steps include:

[0082] Step 201: Acquire the target EEG data of the user collected by the EEG acquisition device from the start of the current acquisition cycle to the current time.

[0083] The EEG acquisition device is used to collect data from the user's head, such as electroencephalogram (EEG) data. It typically includes multiple channels, each used to collect EEG data from different areas of the head; for example, there might be two channels corresponding to the parietal and occipital regions of the head. Generally, each visual stimulus has a corresponding display cycle in the visual stimulator, and the corresponding EEG data also has a corresponding frequency cycle. Within one cycle, all visual stimuli have corresponding display cycles in the visual stimulator, and the EEG data corresponding to each command also has a frequency cycle. Therefore, one acquisition cycle corresponds to all targets flashing on the visual stimulator, and also to the user focusing on any target in the visual stimulator, allowing for the acquisition of complete EEG data.

[0084] In one embodiment, the EEG acquisition device synchronizes the acquired EEG data to the device in real time. If the device and the EEG acquisition device are connected by a cable, the device receives the EEG data acquired by the EEG acquisition device through the cable connected to the EEG acquisition device. If the device and the EEG data communicate via a wireless network, such as WiFi or Bluetooth, the device receives the EEG data acquired by the EEG acquisition device through a signal connection with the EEG acquisition device.

[0085] In one embodiment, the device continuously receives EEG data collected by the EEG acquisition device. At the beginning of each acquisition cycle, the device records the start time of that acquisition cycle and then uses the EEG data collected from the start time of that acquisition cycle to the current time as the target EEG data. For example, if the start time of one acquisition cycle is 17:02:41 and the current time is 17:03:41, then the acquisition time corresponding to the target EEG data is 17:02:41-17:03:41.

[0086] Step 203: Using the target EEG data, predict the target object that the user is looking at from a set of pre-configured visual stimuli.

[0087] When predicting the target object being gazed upon by the user from a pre-configured set of visual stimuli using target EEG data, object prediction strategies used in related technologies for predicting the visual stimuli being gazed upon can be employed. Examples include canonical correlation analysis (CCA) applied to frequency-modulated VEP BCI (f-VEP BCI), individual-template canonical correlation analysis (ITCCA) based on CCA, task-discriminant component analysis (TDCA) applied to code-modulated VEP BCI (c-VEP BCI), and task-related component analysis (TRCA) applied to f-VEP BCI. Detailed descriptions can be found in related technologies and will not be elaborated upon here. The pre-configured stimuli refer to the visual stimuli displayed on the visual stimulator. For example, the visual stimulator includes four visual stimuli, namely patterns A, B, C, and D. These four stimuli flash at different flashing frequencies. The pre-configured stimuli include A, B, C, and D. The device predicts the visual stimuli that the user will fixate on in A, B, C, and D based on the target EEG data, that is, it predicts the target predicted object that the user will fixate on.

[0088] In one embodiment, after the device obtains the target EEG data, it inputs the target EEG data into a spatial filter to obtain a denoised target EEG signal. Then, according to a correlation analysis strategy, it determines the correlation coefficient between the target EEG signal and the template EEG signal corresponding to each visual stimulus object, obtains the predicted probability of each visual stimulus object, and then determines the visual stimulus object with the highest correlation coefficient as the target prediction object, that is, the visual stimulus object with the highest probability is determined as the target prediction object.

[0089] In one embodiment, the device makes predictions based on target EEG data, and the prediction results are multiple probabilities. Specifically, when the device predicts the visual stimulus object that the user is looking at based on the target EEG data, the prediction results are the probabilities of the user looking at each visual stimulus object. For example, if there are five visual stimuli, namely visual stimulus object A, visual stimulus object B, visual stimulus object C, visual stimulus object D, and visual stimulus object E, then the prediction results are as follows: based on the target EEG data, the probability that the visual stimulus object the user is looking at is visual stimulus object A is probability 1, the probability that the visual stimulus object the user is looking at is visual stimulus object B is probability 2, the probability that the visual stimulus object the user is looking at is visual stimulus object C is probability 3, the probability that the visual stimulus object the user is looking at is visual stimulus object D is probability 4, and the probability that the visual stimulus object the user is looking at is visual stimulus object E is probability 5. The visual stimulus object with the highest probability is the target prediction object corresponding to the target EEG data. The prediction result can also be the probability that the user is looking at a certain visual stimulus. For example, the probability that the visual stimulus being looked at by the user is visual stimulus D is 75%. The above are just some examples of object prediction results. In actual applications, prediction results may vary depending on the object prediction strategy used. Please refer to the relevant technologies for the object prediction results corresponding to each object prediction strategy.

[0090] In one embodiment, the device can also input the target EEG data into a model trained by a preset object prediction strategy, and obtain the target predicted object that the user is looking at based on the output of the model.

[0091] Step 205: Determine the confidence level of the target prediction object.

[0092] Specifically, after the device obtains the target prediction object corresponding to the target EEG data, it calculates the confidence level of the target prediction object based on the prediction results. For example, it calculates the entropy value of the prediction results, which is inversely proportional to the confidence level of the target prediction object. Or, it calculates the Gini index of the prediction results, which is inversely proportional to the confidence level of the target prediction object.

[0093] When using different object prediction strategies to predict target EEG data, the confidence determination process for the target prediction object will be different. The confidence determination process for different object prediction strategies in related technologies can be referred to, and will not be described in detail in this application.

[0094] Step 207: If the confidence level of the target prediction object is lower than or equal to the preset confidence threshold and the duration from the start time to the current time is less than the upper limit duration, update the current time and return to execute steps 201, 203, and 205 until a target prediction object with a confidence level higher than the preset confidence threshold is obtained, which is then used as the output prediction object.

[0095] The upper limit duration corresponds to the fixed duration in related technologies, which is the duration corresponding to a collection cycle.

[0096] In one embodiment, after the device determines the confidence level of the target predicted object corresponding to the target EEG data, if the confidence level of the target predicted object is lower than or equal to a preset confidence threshold, it indicates that the target EEG data does not contain complete EEG data corresponding to a certain visual stimulus object. The device further determines whether the duration from the start time of the current acquisition cycle to the current time is less than the upper limit duration. If the device determines that the duration from the start time of the current acquisition cycle to the current time is less than the upper limit duration, it indicates that the current target EEG data does not represent all the EEG data corresponding to the acquisition cycle, and no target predicted object with a confidence threshold higher than the current target EEG data has been obtained. The device continues to receive EEG data acquired by the EEG acquisition device.

[0097] In one embodiment, after the device determines the confidence level of the target prediction object corresponding to the target EEG data, if the confidence level of the target prediction object is higher than a preset confidence threshold, it indicates that the target EEG data is already the complete EEG data of the target prediction object, and the target prediction object is used as the output prediction object.

[0098] In one embodiment, the device determines that the confidence level of the target prediction object is lower than or equal to a preset confidence threshold, and the duration from the start of the current acquisition period to the current time is less than the upper limit duration. It then updates the current time and returns to step 201 until a target prediction object with a confidence level higher than the preset confidence threshold is obtained, which is then used as the output prediction object, i.e., the final prediction object. For example, if the acquisition period corresponding to the target EEG data is 17:02:41-17:03:41, meaning the start time of the current acquisition period is 17:02:41 and the current time is 17:03:41, as the device executes step 203, time passes, and the current time becomes 17:04:22. The device then updates the current time to 17:04:22 and acquires the EEG data acquired by the EEG acquisition device from 17:02:41 to 17:04:22 as the target EEG data.

[0099] The object prediction method based on EEG data provided in this application collects EEG data and performs object prediction based on the currently collected EEG data in each acquisition cycle. Compared with object prediction methods in related technologies, the object prediction efficiency is higher.

[0100] In one embodiment, the above-described object prediction method based on electroencephalogram (EEG) data may further include the following steps:

[0101] Step 207: If the duration from the start time to the current time is greater than or equal to the upper limit duration, return to step 201 at the start of the next acquisition cycle.

[0102] In one embodiment, if the device determines that the confidence level of the target prediction object is lower than or equal to a preset confidence threshold, and the duration from the start of the current acquisition cycle to the current time is greater than or equal to the upper limit duration, it indicates that the user has not looked at any visual stimulus object during the current acquisition cycle. The device then waits for the next acquisition cycle and returns to step 201 at the start of the next acquisition cycle.

[0103] In one embodiment, after obtaining a target prediction object with a confidence level higher than a preset confidence threshold as the output prediction object, the device waits for the next acquisition cycle. At the start of the next acquisition cycle, it returns to step 201. For example, in some application scenarios, after a user fixates on a visual stimulus object, they need to pause for 1 second before starting to fixate on the next visual stimulus object. Therefore, after the device has predicted the corresponding visual stimulus object based on the current target EEG data, it needs to pause for 3 seconds before starting the next acquisition cycle. Alternatively, after the device has predicted the corresponding visual stimulus object based on the current target EEG data, it directly enters the next acquisition cycle in conjunction with the visual stimulator.

[0104] In this embodiment, the device automatically waits for each acquisition cycle, and when the next acquisition cycle begins, it continues to acquire EEG data and perform object prediction.

[0105] In one embodiment, step 203 specifically includes:

[0106] If the duration from the start time to the current time is greater than or equal to the lower limit duration, the target prediction object that the user is looking at is predicted from a number of pre-configured visual stimuli using the target EEG data.

[0107] The lower limit duration refers to the acquisition time for collecting complete EEG data corresponding to at least one visual stimulus object. For example, if there are visual stimulus objects A, B, and C, when predicting visual stimulus object A, at least 0.2 seconds of EEG data needs to be collected from the user; when predicting visual stimulus object B, at least 0.3 seconds of EEG data needs to be collected from the user; and when predicting visual stimulus object C, at least 0.25 seconds of EEG data needs to be collected from the user. Therefore, the lower limit duration is 0.2 seconds.

[0108] In one embodiment, if the duration corresponding to the target EEG data is too short, i.e., the duration from the start of the current acquisition cycle to the current time is too short (e.g., the window duration corresponding to the target EEG data is only 0.1 seconds), the target EEG data will not contain any complete EEG data corresponding to any visual stimulus. Therefore, before predicting the target EEG data, the device compares the duration from the start of the current acquisition cycle to the current time with the lower limit duration. If the device determines that the duration from the start of the current acquisition cycle to the current time is greater than or equal to the lower limit duration, it indicates that the target EEG data may contain complete EEG data corresponding to a visual stimulus. The target EEG data is then used to predict the target prediction object that the user is looking at among a number of pre-configured visual stimuli.

[0109] In one embodiment, the device compares the duration from the start of the current acquisition cycle to the current time with the lower limit duration. If the device determines that the duration from the start of the current acquisition cycle to the current time is less than the lower limit duration, the device determines that the target EEG data will not contain any complete EEG data corresponding to any visual stimulus object, and more EEG data needs to be acquired until the duration of the target EEG data reaches the lower limit duration, and then proceeds to step 203.

[0110] In this embodiment, the device sets a lower limit duration. If the duration from the start of the current acquisition cycle to the current time is too short, object prediction will not be performed on the current target EEG data to avoid unnecessary waste of computing resources.

[0111] In one embodiment, the above method further includes:

[0112] Step 209: Based on the control instructions corresponding to the pre-configured visual stimulus objects, determine the control instructions for the output prediction object, and perform corresponding processing according to the control instructions for the output prediction object.

[0113] Specifically, the correspondence between visual stimuli and commands varies across different application scenarios. For example, if visual stimuli include A, B, and C, in some scenarios, visual stimulus A corresponds to the command to turn on the TV, visual stimulus C corresponds to the command to turn off the TV, and visual stimulus B corresponds to the command to change the channel. If the output prediction object is visual stimulus B, then based on the correspondence between the predicted visual stimulus and the command, the target command corresponding to the output prediction object is determined to be turning off the TV, and the device executes the command to turn off the TV. In other scenarios, visual stimulus A corresponds to the command to move the wheelchair forward, visual stimulus C corresponds to the command to move the wheelchair backward, and visual stimulus B corresponds to the command to stop the wheelchair. If the output prediction object is visual stimulus C, then based on the correspondence between the predicted visual stimulus and the command, the target command corresponding to the output prediction object is determined to be moving the wheelchair backward, and the device controls the wheelchair to move backward through an electrical connection with the wheelchair.

[0114] In one embodiment, step 203 specifically includes:

[0115] Step 203A: Input the target EEG data into the spatial filter to obtain the denoised target EEG signal.

[0116] In one embodiment, taking the application of the relevant component analysis algorithm to f-VEP BCI to predict the target object of user gaze among a number of pre-configured visual stimuli using target EEG data as an example, the device pre-acquires template EEG data corresponding to each visual stimulus object and generates a spatial filter that can filter the signal. The device inputs the target EEG data into the spatial filter to obtain the denoised target EEG signal.

[0117] Step 203B: Based on the correlation analysis strategy, determine the correlation coefficient between the target EEG signal and the template EEG signal corresponding to each visual stimulus object.

[0118] Specifically, the device inputs the target EEG data into a spatial filter to obtain the denoised target EEG signal. Then, based on a correlation analysis strategy, it calculates the correlation coefficient between the target EEG signal and the template EEG signal corresponding to each visual stimulus. The higher the correlation coefficient, the higher the probability that the visual stimulus corresponding to the target EEG data is the visual stimulus corresponding to the template, that is, the higher the probability that the user fixates on the visual stimulus corresponding to the template.

[0119] Step 203C: Identify the visual stimulus with the highest correlation coefficient as the target prediction object.

[0120] Specifically, after the device obtains the correlation coefficient between the target EEG data and each visual stimulus, it identifies the visual stimulus with the highest correlation coefficient as the target prediction object.

[0121] In one embodiment, the above-described object prediction method based on electroencephalogram (EEG) data may further include the following steps:

[0122] Step 211: During the calibration phase, receive the calibration EEG data collected by the EEG acquisition device; the calibration EEG data is used to calibrate the parameters in the object prediction strategy used for prediction.

[0123] The time curves of visual evoked responses vary significantly among different users. Therefore, some object prediction strategies require a calibration process to obtain a template corresponding to each visual stimulus. Then, using these templates, the user's gaze at the visual stimulus is predicted. Thus, calibrating EEG data is used to generate templates corresponding to different visual stimuli.

[0124] In one embodiment, the device prompts the user to enter the calibration phase and collects calibration EEG data, then receives the calibration EEG data collected by the EEG acquisition device. If the object prediction strategy used for prediction is the TRCA algorithm applied to the f-VEPBCI object prediction strategy, then it is necessary to collect the EEG data of the target corresponding to each visual stimulus object, and the user needs to repeatedly fixate on each target multiple times. If the object prediction strategy used for prediction is the TRCA algorithm applied to the c-VEP BCI object prediction strategy, then the device only needs to collect the EEG data corresponding to the user's fixation on a certain visual stimulus object; the templates corresponding to other visual stimuli objects can be obtained by time-shifting the templates of the calibration instructions.

[0125] Step 213: Using a preset calibration strategy, calibrate the parameters in the prediction strategy based on the calibrated EEG data.

[0126] In one embodiment, after acquiring the user's calibrated EEG data, the device uses a predicted calibration strategy to calibrate the parameters of the object prediction strategy based on the calibrated EEG data. For example, the device can use the user's calibrated EEG data to adjust the parameters of the filters corresponding to each visual stimulus object. It can also acquire multiple sets of the user's calibrated EEG data and train the parameters in the object prediction strategy using multiple sets of calibrated EEG data to obtain the calibrated object prediction strategy. For the specific calibration process, please refer to the description of the calibration process for different object prediction strategies in the related art, which will not be elaborated in this application.

[0127] In this embodiment, the parameters in the object prediction strategy used for prediction are calibrated by collecting the user's calibrated EEG data, which makes the device more accurate in predicting objects based on the collected EEG data.

[0128] In one embodiment, step 213 specifically includes:

[0129] Step 213a: Determine the availability of calibrated EEG data.

[0130] In one embodiment, when the EEG acquisition device was collecting the user's calibration EEG data, the user did not cooperate with the collection of calibration EEG data as prompted, so the usability of the calibration EEG data obtained by the device was not high.

[0131] In one embodiment, the device can acquire multiple sets of calibrated EEG data from a user, use one set of calibrated EEG data as test data, and use the remaining multiple sets of calibrated EEG data as training data. The device uses the training data to train an object prediction strategy, and then uses the trained object prediction strategy to predict objects on the test data. The prediction accuracy is proportional to the availability of the calibrated EEG data.

[0132] In one embodiment, the device collects training EEG data through calibration, then scores the calibrated EEG data and generates algorithm parameters. Afterward, the user decides whether to recalibrate or end the calibration based on the calibration availability rate reported by the device.

[0133] Step 213b: If the availability rate is greater than the preset availability rate threshold, display the availability rate to the user.

[0134] In one embodiment, after determining the availability of the calibrated EEG data, if the availability is greater than a preset availability threshold, the device uses a preset calibration strategy to calibrate the parameters in the target prediction strategy based on the calibrated EEG data. For an explanation of this step, please refer to the description of step 213 above.

[0135] In this embodiment, training EEG data is collected through calibration, and then algorithm parameters are generated based on the calibrated EEG data. The user determines whether to recalibrate or end the calibration based on the calibration score fed back by the device. After showing the user the availability of the collected calibrated EEG data, the device waits for and periodically detects the user's input commands.

[0136] Step 213d: Upon receiving a user confirmation calibration instruction, the parameters in the object prediction strategy are calibrated based on the calibration EEG data using a preset calibration strategy.

[0137] In one embodiment, the user detects a user input confirmation calibration command, confirms receipt of the user confirmation calibration command, and uses a preset calibration strategy to calibrate the parameters in the target prediction strategy based on the calibration EEG data. The explanation of "using a preset calibration strategy to calibrate the parameters in the target prediction strategy based on the calibration EEG data" can be found in the description of step 213 above.

[0138] Step 213f: Upon receiving a user recalibration instruction, return to step 211.

[0139] In one embodiment, the user detects a user input recalibration command, determines that the user recalibration command has been received, returns to step 211, and reacquires the user's calibration EEG data collected by the EEG acquisition device.

[0140] In this embodiment, the user can decide whether to perform calibration based on the current availability, according to their own wishes.

[0141] In one embodiment, when the target EEG data is based on coded modulation visual evoked potential data, i.e., c-VEP BCI, step 211 includes:

[0142] Step 211a: Select one visual stimulus object from a pre-configured set of visual stimulus objects as the calibration object, and control the preset visual stimulator to stimulate the calibration object.

[0143] Specifically, each visual stimulus has its corresponding flashing frequency. The device selects one of the pre-configured visual stimuli as the calibration object, and then controls the preset visual stimulator to stimulate the calibration object so that when the user looks at the target object, corresponding EEG data is generated.

[0144] Step 211b: Receive the calibration EEG data for the calibration object collected by the EEG acquisition device, and obtain the template EEG signal corresponding to the calibration object.

[0145] Specifically, the device controls a preset visual stimulator to stimulate the calibration object. The user gazes at the calibration object and generates corresponding EEG data. The EEG acquisition device collects the EEG data generated when the user gazes at the calibration object in the visual stimulator and synchronizes it to the device. After receiving the calibration EEG data for the calibration object collected by the EEG acquisition device, the device processes the calibration EEG data, such as noise reduction, to obtain the template EEG signal corresponding to the calibration object.

[0146] Step 211c: For each of the pre-configured visual stimulus objects, the template EEG signal corresponding to the calibration object is time-shifted according to a preset time-shifting strategy to obtain the template EEG signal corresponding to the other visual stimulus objects.

[0147] Specifically, the device acquires the template EEG signal corresponding to the calibration object. Since the EEG signals corresponding to different visual stimuli in c-VEP BCI have a cyclic displacement relationship, the templates for other visual stimuli can be obtained by time-shifting the template EEG signal corresponding to the calibration object. For example, the length of the template EEG signal is Ts = 63 / 60 = 1.05s, which is equal to the length of one stimulation cycle of the visual stimulator. Once the template EEG signal of the calibration object t0 is obtained, the template EEG signals of other visual stimuli can be easily obtained by cyclically shifting the template EEG signal of t0.

[0148] In this embodiment, during calibration, the user only needs to focus on one of the visual stimuli, without having to focus on each visual stimuli and collect the corresponding calibration EEG data.

[0149] like Figure 3 The diagram shown is a flowchart illustrating an object prediction method based on electroencephalogram (EEG) data as presented in this application.

[0150] Step 1: The terminal receives the EEG data collected by the EEG acquisition device.

[0151] Step 2: When the window duration corresponding to the target EEG data collected within the acquisition cycle (i.e., the duration from the start of the current acquisition cycle to the current time) reaches the lower limit, the terminal inputs the target EEG data into the pre-trained network model to obtain the target object prediction result, or uses a preset object prediction strategy to obtain the target predicted object.

[0152] Step 3: The terminal determines whether the confidence level of the target prediction object is higher than the preset confidence threshold. If it is higher than the preset confidence threshold, proceed to step 4; if it is lower than or equal to the preset confidence threshold, proceed to step 5.

[0153] Step 4: Use the target prediction object as the output prediction object and execute the instructions corresponding to the output prediction object.

[0154] Step 5: The terminal determines whether the duration from the start of the current acquisition cycle to the current time has reached the upper limit. If the upper limit has not been reached, it returns to step 1 and continues to receive EEG data collected by the EEG acquisition device; if the upper limit has been reached, it proceeds to step 6.

[0155] Step 6: Clear the target EEG data and begin the next trial judgment, that is, begin the object prediction for the next collection cycle.

[0156] This application also provides an object prediction system based on EEG data. The system includes a terminal and an EEG acquisition device. The EEG acquisition device is used to collect the user's EEG data, and the terminal is used to execute any of the above-described object prediction methods based on EEG data. The EEG acquisition device and the terminal can be electrically connected or wirelessly connected, that is, a wireless communication connection can be established through WiFi, Bluetooth, or other means.

[0157] This application also provides an object prediction system based on electroencephalogram (EEG) data. The system includes a terminal, an EEG acquisition device, and a controlled device. The EEG acquisition device is used to collect the user's EEG data. The terminal is used to execute any of the aforementioned object prediction methods based on EEG data. The EEG acquisition device and the terminal can be electrically connected or wirelessly connected, i.e., a wireless communication connection can be established via WiFi, Bluetooth, or other methods. The controlled device executes corresponding actions according to the instructions predicted by the terminal, such as the wheelchair mentioned above.

[0158] 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.

[0159] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0160] Based on the same inventive concept, this application also provides an object prediction device based on electroencephalogram (EEG) data for implementing the above-described object prediction method based on EEG data. The solution provided by this device is similar to the implementation described in the above-described method. Therefore, the specific limitations of one or more object prediction device embodiments based on EEG data provided below can be found in the limitations of the object prediction method based on EEG data described above, and will not be repeated here.

[0161] In one embodiment, such as Figure 4 As shown, an object prediction device based on electroencephalogram (EEG) data is provided, comprising:

[0162] The receiving module 401 is used to acquire the target EEG data collected by the EEG acquisition device from the start of the current acquisition cycle to the current time.

[0163] The prediction module 403 is used to predict the target prediction object corresponding to the target EEG data among a preset number of objects using a preset object prediction strategy.

[0164] The determination module 405 is used to determine the confidence level of the target prediction object;

[0165] The output module 407 is used to update the current time and return to the steps of acquiring the user's target EEG data collected by the EEG acquisition device from the start time of the acquisition cycle to the current time, predicting the target prediction object that the user is looking at among a number of pre-configured visual stimuli using the target EEG data, and determining the confidence of the target prediction object, until a target prediction object with a confidence higher than the confidence threshold is obtained as the output prediction object.

[0166] In one embodiment, the above-mentioned apparatus further includes:

[0167] Return module 409 (not shown in the figure) is used to perform the step of acquiring the user target EEG data acquired by the EEG acquisition device from the start time of the current acquisition cycle to the current time when the duration from the start time to the current time is greater than or equal to the upper limit duration, at the beginning of the next acquisition cycle.

[0168] In one embodiment, the prediction module 403 is specifically used for:

[0169] If the duration from the start time to the current time is greater than or equal to the lower limit duration, the target EEG data is used to predict the target object being gazed upon by the user among a pre-configured set of visual stimuli.

[0170] In one embodiment, the above-mentioned apparatus further includes:

[0171] The execution module 411 (not shown in the figure) is used to determine the control instructions for the output prediction object according to the control instructions corresponding to the pre-configured visual stimulus objects, so as to perform corresponding processing according to the control instructions for the output prediction object.

[0172] In one embodiment, the prediction module 403 is specifically used for:

[0173] The target EEG data is input into a spatial filter to obtain the denoised target EEG signal;

[0174] Based on the correlation analysis strategy, the correlation coefficient between the target EEG signal and the template EEG signal corresponding to each of the visual stimuli is determined;

[0175] The visual stimulus with the highest correlation coefficient is identified as the target prediction object.

[0176] In one embodiment, the above-mentioned apparatus further includes:

[0177] The calibration data receiving module 413 (not shown in the figure) is used to receive the calibration EEG data collected by the EEG acquisition device;

[0178] The calibration module 415 (not shown in the figure) is used to calibrate the parameters in the object prediction strategy based on the calibrated EEG data using a preset calibration strategy.

[0179] In one embodiment, the calibration module 415 specifically includes:

[0180] Availability determination unit 415A (not shown in the figure) is used to determine the availability of the calibrated EEG data;

[0181] Display unit 415B (not shown in the figure) is used to display the availability rate to the user when the availability rate is greater than a preset availability rate threshold.

[0182] The calibration unit 415C (not shown in the figure) is used to calibrate the parameters in the object prediction strategy based on the calibration EEG data using a preset calibration strategy when the user confirms the calibration instruction.

[0183] The return unit 415D (not shown in the figure) is used to return to the step of receiving the calibration EEG data collected by the EEG acquisition device when the user recalibrates the instruction.

[0184] In one embodiment, when the target EEG data is based on encoded modulated visual evoked potential data, the calibration data receiving module 413 includes:

[0185] The selection unit 413A (not shown in the figure) is used to select one visual stimulus object from the pre-configured plurality of visual stimulus objects as a calibration object, and control the preset visual stimulator to stimulate the calibration object.

[0186] The template acquisition unit 413B (not shown in the figure) is used to receive the calibration EEG data for the calibration object collected by the EEG acquisition device, and obtain the template EEG signal corresponding to the calibration object.

[0187] The shifting unit 413C (not shown in the figure) is used to shift the template EEG signal corresponding to the calibration object in time according to a preset time shifting strategy for each of the pre-configured visual stimulus objects and other visual stimulus objects, so as to obtain the template EEG signal corresponding to the other visual stimulus objects.

[0188] The modules in the aforementioned object prediction device based on electroencephalogram (EEG) data can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0189] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 5As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements an object prediction method based on electroencephalogram (EEG) data. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0190] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0191] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the above-described object prediction methods based on electroencephalogram (EEG) data.

[0192] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-described object prediction methods based on electroencephalogram (EEG) data.

[0193] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of any of the above-described object prediction methods based on electroencephalogram (EEG) data.

[0194] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0195] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0196] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0197] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

[0198] 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 method for object prediction based on electroencephalogram (EEG) data, characterized in that, The method includes: During the calibration phase, calibration EEG data collected by the EEG acquisition device is received; the calibration EEG data is used to calibrate the parameters in the object prediction strategy used for prediction. Using a pre-defined calibration strategy, parameters in the prediction strategy of the target are calibrated based on the calibrated EEG data; Acquire the target EEG data of the user collected by the EEG acquisition device from the start of the current acquisition cycle to the current time; Using the target EEG data and the object prediction strategy, the target object being gazed upon by the user is predicted from among a pre-configured set of visual stimuli. Determine the confidence level of the target prediction object; If the confidence level of the target prediction object is lower than or equal to a preset confidence threshold, and the duration from the start time to the current time is less than the upper limit duration, update the current time, and return to execute the steps of acquiring the user's target EEG data collected by the EEG acquisition device from the start time of the acquisition cycle to the current time, using the target EEG data to predict the target prediction object that the user is looking at among a number of pre-configured visual stimuli, and determining the confidence level of the target prediction object, until a target prediction object with a confidence level higher than the confidence threshold is obtained, which is then used as the output prediction object; The target EEG data is based on coded modulation visual evoked potential data, and the calibrated EEG data acquired by the receiving EEG acquisition device includes: Select one visual stimulus object from a pre-configured set of visual stimulus objects as a calibration object, and control a preset visual stimulator to stimulate the calibration object. Receive calibration EEG data for the calibration object collected by the EEG acquisition device, and obtain the template EEG signal corresponding to the calibration object; For each of the pre-configured visual stimulus objects, the template EEG signal corresponding to the calibration object is time-shifted according to a preset time-shifting strategy to obtain the template EEG signal corresponding to the other visual stimulus objects.

2. The method according to claim 1, characterized in that, The method further includes: If the duration from the start time to the current time is greater than or equal to the upper limit duration, at the start of the next acquisition cycle, the step of acquiring the user's target EEG data acquired by the EEG acquisition device from the start time of the current acquisition cycle to the current time is executed.

3. The method according to claim 1, characterized in that, The step of predicting the target object being gazed upon by the user among a pre-configured set of visual stimuli using the target EEG data and the object prediction strategy includes: If the duration from the start time to the current time is greater than or equal to the lower limit duration, the target EEG data and the object prediction strategy are used to predict the target object that the user is looking at among a number of pre-configured visual stimuli.

4. The method according to any one of claims 1-3, characterized in that, The method further includes: Based on the control instructions corresponding to the pre-configured visual stimulus objects, control instructions for the output prediction object are determined, and corresponding processing is performed according to the control instructions for the output prediction object.

5. The method according to claim 1, characterized in that, The step of predicting the target object being gazed upon by the user among a pre-configured set of visual stimuli using the target EEG data and the object prediction strategy includes: The target EEG data is input into a spatial filter to obtain the denoised target EEG signal; Based on the correlation analysis strategy, the correlation coefficient between the target EEG signal and the template EEG signal corresponding to each of the visual stimuli is determined; The visual stimulus with the highest correlation coefficient is identified as the target prediction object.

6. An object prediction system based on electroencephalogram (EEG) data, characterized in that, The system includes: EEG acquisition equipment is used to collect users' EEG data; A terminal is used to receive calibration EEG data collected by an EEG acquisition device during the calibration phase. The calibration EEG data is used to calibrate parameters in the object prediction strategy used for prediction. Based on the calibration EEG data, the parameters in the object prediction strategy are calibrated using a preset calibration strategy. The terminal acquires the user's target EEG data collected by the EEG acquisition device from the start of the current acquisition cycle to the current time. Using the target EEG data and the object prediction strategy, the terminal predicts the target predicted object the user is looking at from a set of pre-configured visual stimuli. The terminal determines the confidence level of the target predicted object. If the confidence level of the target predicted object is lower than or equal to a preset confidence threshold, and the duration from the start of the acquisition cycle to the current time is less than an upper limit, the terminal updates the current time and returns to the previous steps of acquiring the user's target EEG data collected by the EEG acquisition device from the start of the acquisition cycle to the current time, predicting the target predicted object the user is looking at from a set of pre-configured visual stimuli using the target EEG data, and determining the confidence level of the target predicted object, until a target predicted object with a confidence level higher than the confidence threshold is obtained as the output predicted object. The target EEG data is based on coded modulation visual evoked potential data. The terminal is specifically used to: select one visual stimulus object as a calibration object from a pre-configured set of visual stimulus objects, and control a preset visual stimulator to stimulate the calibration object; receive calibration EEG data for the calibration object collected by an EEG acquisition device to obtain a template EEG signal corresponding to the calibration object; and for other visual stimulus objects in each set of pre-configured visual stimulus objects, perform time shifting on the template EEG signal corresponding to the calibration object according to a preset time shifting strategy to obtain template EEG signals corresponding to other visual stimulus objects.

7. An object prediction device based on electroencephalogram (EEG) data, characterized in that, The device includes: The calibration data receiving module is used to receive calibration EEG data collected by the EEG acquisition device; the calibration EEG data is used to calibrate the parameters in the object prediction strategy used for prediction. The calibration module is used to calibrate the parameters in the prediction strategy of the target based on the calibration EEG data using a preset calibration strategy. The receiving module is used to acquire the target EEG data of the user collected by the EEG acquisition device from the start of the current acquisition cycle to the current time. The prediction module is used to predict the target object that the user is looking at from a pre-configured set of visual stimuli using the target EEG data and the object prediction strategy. The determination module is used to determine the confidence level of the target prediction object; The output module is used to update the current time and return to the steps of acquiring the user's target EEG data collected by the EEG acquisition device from the start time of the acquisition cycle to the current time, predicting the target prediction object that the user is looking at among a number of pre-configured visual stimuli using the target EEG data, and determining the confidence of the target prediction object, until a target prediction object with a confidence higher than the confidence threshold is obtained as the output prediction object; The target EEG data is based on coded modulated visual evoked potential data. The calibration data receiving module is specifically used to select one visual stimulus object as the calibration object from a pre-configured set of visual stimulus objects, and control a preset visual stimulator to stimulate the calibration object; receive the calibration EEG data for the calibration object collected by the EEG acquisition device to obtain the template EEG signal corresponding to the calibration object; for other visual stimulus objects in each set of pre-configured visual stimulus objects, according to a preset time shifting strategy, the template EEG signal corresponding to the calibration object is time-shifted to obtain the template EEG signal corresponding to the other visual stimulus objects.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images