Intelligent control method and device of eye massager and storage medium
By using an electromyography sensor on an eye massager to determine the state of the eye muscles, combined with eye movement information and field of view angle, visual guidance training is achieved in a non-tension state, and the damage to the eye muscles of the eye movement training equipment is solved. The problem of inaccurate eye tracking improves the safety and accuracy of the equipment.
Patent Information
- Application Number
- CN202411362707.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-27
- Publication Date
- 2024-12-06
AI Technical Summary
Existing eye movement training equipment performs visual guidance training when eye muscles are tight, which can easily lead to irreversible damage. Eye movement tracking technology relies on cameras or infrared optical sensors, which is large in size, high in cost and susceptible to ambient light. The problem leads to inaccuracy.
By installing an electromyography sensor on the eye massager, collect the electromyography signals of the eye muscles to determine whether they are in a non-tension state. If they are not nervous, perform visual guidance training; if they are nervous, start the massage mode and then determine whether they are after the preset time. Training can be carried out, and eye movement information and field of view can be calculated through electromyography signals to evaluate the training completion degree.
It improves the safety of eye massagers, ensures training in appropriate conditions, reduces muscle damage, and improves the accuracy and reliability of training through accurate judgment of electromyography signals.
Smart Images

Figure CN119074497A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of massage equipment, and in particular to an intelligent control method and device, and a storage medium for an eye massager. Background Art
[0002] In the related art, eye movement training equipment directly enters into visual guided training. Due to the widespread overuse of eyes, if visual guided training is performed when some eye muscles are too tense, it will cause irreversible damage to the eye muscles. It is difficult for users to determine in what state visual training can be performed and in what state visual training cannot be performed. Therefore, current eye movement training equipment also increases the difficulty of use.
[0003] In addition, eye tracking technology in related technologies mainly relies on cameras or infrared optical sensors, but it has problems such as large size, high cost, and susceptibility to ambient light, which can also lead to inaccurate defects.
[0004] With respect to the above-mentioned problems existing in the related technologies, no efficient and accurate solutions have been found yet. Summary of the invention
[0005] The present invention provides an intelligent control method and device for an eye massager, and a storage medium to solve the above-mentioned technical problems existing in the related art.
[0006] According to one embodiment of the present invention, there is provided an intelligent control method for an eye massager, comprising: collecting a first electromyographic signal of an eye muscle by an electromyographic sensor of the eye massager, wherein the electromyographic sensor is arranged in an eye contact area between the eye massager and a target object; judging whether the eye muscle is in a non-tense state according to the first electromyographic signal; and if the eye muscle is in a non-tense state, controlling the eye massager to perform visual guidance training on the target object.
[0007] Optionally, after determining whether the eye muscles are in a non-tense state according to the electromyographic signal, the method further includes: if the eye muscles are in a tense state, controlling the eye massager to start a massage mode to massage the eye muscles; after the eye massager runs in the massage mode for a preset time, collecting a second electromyographic signal of the eye muscles through the electromyographic sensor of the eye massager; determining whether the eye muscles are in a non-tense state according to the second electromyographic signal; if the eye muscles are in a non-tense state, controlling the eye massager to perform visual guidance training on the target object.
[0008] Optionally, controlling the eye massager to perform visual guided training on the target object includes: determining the actual field of view angle of the target object, and collecting a third electromyographic signal of the eye muscle in real time through the electromyographic sensor of the eye massager; calculating the eye movement information of the target object based on the third electromyographic signal; and calculating the completion degree of the visual guided training based on the actual field of view angle and the eye movement information.
[0009] Optionally, determining the actual field of view angle of the target object includes: controlling the visual mark on the head-mounted display screen to move from a reference position in the center of the screen to an edge of the screen, and collecting the extreme movement positions of the target object's eyes when dynamically capturing the visual mark, wherein the extreme movement positions include horizontal extreme positions and vertical extreme positions; and calculating the actual field of view angle of the target object based on the extreme movement positions.
[0010] Optionally, calculating the actual field of view angle of the target object based on the extreme movement position includes: determining the screen distance from the eyeball of the target object to the head-mounted display screen; using the following formula to calculate the actual field of view angle of the target object α=arctan(Δx / d); wherein d is the screen distance, Δx=x1-x2, x1 is the extreme movement position, and x2 is the reference position.
[0011] Optionally, calculating the eye movement information of the target object based on the third electromyographic signal includes: extracting time domain features and frequency domain features of the third electromyographic signal; combining the time domain features and the frequency domain features into a feature vector; inputting the feature vector into a pre-trained eye movement prediction model to output the gaze coordinates of the target object; and calculating the eye movement direction of the target object based on the gaze coordinates, wherein the eye movement information includes the eye movement direction.
[0012] Optionally, calculating the eye movement direction of the target object based on the gaze coordinates includes: determining a horizontal radius R of a model field of view of the eye movement prediction model H and vertical radius R V , and determining the horizontal viewing angle θ of the gaze coordinate in the model viewing angle H and vertical field angle θ V ; Use the following formula to calculate the horizontal rotation angle α H and the vertical rotation angle α V : Wherein, the gaze coordinate is (X i , Y i ), the eye movement direction includes the horizontal rotation angle and the vertical rotation angle.
[0013] Optionally, calculating the completion of the visual guided training based on the actual field of view angle and the eye movement information includes: converting the eye movement information into standard eye coordinates in a world coordinate system based on the actual field of view angle; generating an eye movement trajectory based on fitting of multiple sets of standard eye coordinates within a training cycle; calculating the offset of the eye movement trajectory relative to the visual guidance trajectory; and calculating the completion of the visual guided training based on the offset, wherein the completion is negatively correlated with the offset.
[0014] According to another embodiment of the present invention, there is provided an intelligent control device for an eye massager, comprising: a first acquisition module, used for acquiring a first electromyographic signal of an eye muscle through an electromyographic sensor of the eye massager, wherein the electromyographic sensor is arranged in an eye contact area between the eye massager and a target object; a first judgment module, used for judging whether the eye muscle is in a non-tense state according to the first electromyographic signal; and a first control module, used for controlling the eye massager to perform visual guidance training on the target object if the eye muscle is in a non-tense state.
[0015] Optionally, the device also includes: a second control module, which is used to control the eye massager to start a massage mode to massage the eye muscles after the first judgment module judges whether the eye muscles are in a non-tense state according to the electromyographic signal. If the eye muscles are in a tense state, the control module is used to control the eye massager to start a massage mode to massage the eye muscles; a second acquisition module, which is used to collect a second electromyographic signal of the eye muscles through the electromyographic sensor of the eye massager after the eye massager runs in the massage mode for a preset time; a second judgment module, which is used to judge whether the eye muscles are in a non-tense state according to the second electromyographic signal; and a third control module, which is used to control the eye massager to perform visual guidance training on the target object if the eye muscles are in a non-tense state.
[0016] Optionally, the first control module includes: a determination unit, used to determine the actual field of view angle of the target object, and to collect the third electromyographic signal of the eye muscle in real time through the electromyographic sensor of the eye massager; a first calculation unit, used to calculate the eye movement information of the target object according to the third electromyographic signal; and a second calculation unit, used to calculate the completion degree of the visual guided training according to the actual field of view angle and the eye movement information.
[0017] Optionally, the determination unit includes: a control subunit, used to control the sight mark on the head-mounted display screen to move from a reference position in the center of the screen to the edge of the screen, and to collect the extreme movement positions when the target object's eyes dynamically capture the sight mark, wherein the extreme movement positions include horizontal extreme positions and vertical extreme positions; and a calculation subunit, used to calculate the actual field of view angle of the target object based on the extreme movement positions.
[0018] The calculation subunit is also used to: determine the screen distance from the eyeball of the target object to the head display screen; calculate the actual field of view angle α of the target object using the following formula: α=arctan(Δx / d); wherein d is the screen distance, Δx=x1-x2, x1 is the extreme movement position, and x2 is the reference position.
[0019] Optionally, the first computing unit includes: an extraction subunit, used to extract time domain features and frequency domain features of the third electromyographic signal; a combination subunit, used to combine the time domain features and the frequency domain features into a feature vector; an output subunit, used to input the feature vector into a pre-trained eye movement prediction model, and output the gaze coordinates of the target object; and a computing subunit, used to calculate the eye movement direction of the target object based on the gaze coordinates, wherein the eye movement information includes the eye movement direction.
[0020] Optionally, the calculation subunit is further used to: determine the horizontal radius R of the model field of view of the eye movement prediction model H and vertical radius R V , and determining the horizontal viewing angle θ of the gaze coordinate in the model viewing angle H and vertical field angle θ V ; Use the following formula to calculate the horizontal rotation angle α H and the vertical rotation angle α V : Wherein, the gaze coordinate is (X i , Y i ), the eye movement direction includes the horizontal rotation angle and the vertical rotation angle.
[0021] Optionally, the second calculation unit includes: a conversion subunit, used to convert the eye movement information into standard eye coordinates in the world coordinate system based on the actual field of view angle; a generation subunit, used to generate an eye movement trajectory based on multiple sets of standard eye coordinates fitting within a training cycle; a first calculation subunit, used to calculate the offset of the eye movement trajectory relative to the visual guidance trajectory; and a second calculation subunit, used to calculate the completion of the visual guidance training based on the offset, wherein the completion is negatively correlated with the offset.
[0022] According to yet another embodiment of the present invention, a storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above-mentioned device embodiments when running.
[0023] According to another embodiment of the present invention, there is provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above-mentioned device embodiments.
[0024] Through the embodiment of the present invention, the first electromyographic signal of the eye muscle is collected by the electromyographic sensor of the eye massager, wherein the electromyographic sensor is arranged in the contact area between the eye massager and the eye of the target object; whether the eye muscle is in a non-tense state is judged according to the first electromyographic signal; if the eye muscle is in a non-tense state, the eye massager is controlled to perform visual guidance training on the target object. Before using the eye massager to perform visual guidance training on the target object, it is first judged by the first electromyographic signal whether the eye muscle is in a non-tense state, which solves the technical problem in the related art that the eye massager is easy to cause damage to the eye muscle, and improves the safety of the eye massager. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:
[0026] Figure 1 is a hardware structure block diagram of an eye massager according to an embodiment of the present invention;
[0027] Figure 2 is a flow chart of an intelligent control method of an eye massager according to an embodiment of the present invention;
[0028] Figure 3 Schematic diagram of the layout of the electromyographic sensor in an embodiment of the present invention;
[0029] Figure 4 is a flow chart of visual guidance training through electromyographic signals in an embodiment of the present invention;
[0030] Figure 5 It is a structural block diagram of an intelligent control device for an eye massager according to an embodiment of the present invention. DETAILED DESCRIPTION
[0031] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only embodiments of a part of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present application. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.
[0032] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0033] Example 1
[0034] The method embodiment provided in the first embodiment of the present application can be executed in an eye massager, a health device, a massager or a similar display device. Taking the operation on the eye massager as an example, Figure 1 1 is a hardware structure diagram of an eye massager according to an embodiment of the present invention. Figure 1 As shown, the eye massager may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data. Optionally, the above-mentioned eye massager may also include a transmission device 106 for communication functions and an input and output device 108. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned eye massager. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.
[0035] The memory 104 can be used to store eye massager programs, for example, software programs and modules of application software, such as an eye massager program corresponding to an intelligent control method for an eye massager in an embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the eye massager program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the eye massager via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0036] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the eye massager. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.
[0037] In this embodiment, an intelligent control method for an eye massager is provided. Figure 2 is a flow chart of an intelligent control method of an eye massager according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:
[0038] Step S202, collecting a first electromyographic signal of an eye muscle through an electromyographic sensor of the eye massager, wherein the electromyographic sensor is arranged in a contact area between the eye massager and the eye of the target object;
[0039] The eye massager of this embodiment is a health device or wearable device that acts on living bodies such as humans and animals. It can be an eye massager with an eye massage function, or it can be the following massagers with a head massage function: eye protectors, eye masks, virtual reality helmets, facial massagers, face massagers, etc.
[0040] sEMG electrodes (surface electromyography sensors) are placed in the outer eye muscle area designed according to regulations around the eyeball, and the upper and lower rectus muscles and left and right rectus muscle areas can be selected to capture the electromyography signals of horizontal and vertical eye movements. The electromyography sensors of this embodiment are arranged on the eye massager, including multiple ones. After the user wears the eye massager, each electromyography sensor is used to collect the electromyography signals of the corresponding part. Figure 3 Schematic diagram of the layout of the electromyographic sensors in the embodiment of the present invention, showing S1 to S8, a total of 8 electromyographic sensors, 4 electromyographic sensors corresponding to each eyeball, which are respectively used to collect electromyographic signals of the superior rectus muscle, inferior rectus muscle, lateral rectus muscle, and medial rectus muscle of the eyeball.
[0041] A high sampling frequency (e.g. 1000Hz) can be set to ensure that the electromyographic signals with sufficient details are collected. During the passive test, the subject does not need to actively move the eyeballs. The system induces eye movements through automation or external stimulation and collects the corresponding sEMG signals (electromyographic signals).
[0042] Step S204, judging whether the eye muscles are in a non-tense state according to the first electromyographic signal;
[0043] Whether the eye muscles are in a non-tense state can be determined by the signal strength of the first electromyographic signal, and whether the signal strength of the first electromyographic signal is less than a preset strength. If it is less than the preset strength, the eye muscles are in a non-tense state. When the first electromyographic signal includes multiple groups of electromyographic signals (each group of electromyographic signals corresponds to one electromyographic sensor), the average signal strength of the multiple groups of electromyographic signals is calculated, and whether the average signal strength is less than a preset strength is determined. If it is less than the preset strength, the eye muscles are in a non-tense state.
[0044] Step S206, if the eye muscles are in a non-tense state, control the eye massager to perform visual guidance training on the target object.
[0045] In the case of visually guided training, the training effect is also evaluated through the electromyographic signals of the eye muscles. If the training effect reaches the preset effect within a training cycle, the training can be ended. If the training does not achieve the preset effect after a training cycle, the training duration will be counted. If the training does not reach the preset duration, the next training will be carried out. When the accumulated time reaches a certain amount, the training will also be ended, because long-term ineffective training will only waste time and fail to achieve the purpose of training. It may be that the current user is not in the right state, and the training can be carried out at another time.
[0046] Through the above steps, the first electromyographic signal of the eye muscle is collected by the electromyographic sensor of the eye massager, wherein the electromyographic sensor is arranged in the contact area between the eye massager and the eye of the target object; whether the eye muscle is in a non-tense state is judged according to the first electromyographic signal; if the eye muscle is in a non-tense state, the eye massager is controlled to perform visual guidance training on the target object. Before using the eye massager to perform visual guidance training on the target object, it is first judged by the first electromyographic signal whether the eye muscle is in a non-tense state, which solves the technical problem in the related art that the eye massager is easy to cause damage to the eye muscle, and improves the safety of the eye massager.
[0047] In one implementation of the present embodiment, after determining whether the eye muscles are in a non-tense state based on the electromyographic signal, it also includes: if the eye muscles are in a tense state, controlling the eye massager to start a massage mode to massage the eye muscles; after the eye massager runs in the massage mode for a preset time, collecting a second electromyographic signal of the eye muscles through the electromyographic sensor of the eye massager; determining whether the eye muscles are in a non-tense state based on the second electromyographic signal; if the eye muscles are in a non-tense state, controlling the eye massager to perform visual guidance training on the target object.
[0048] When judging whether the eye muscles are in a non-tense state based on electromyographic signals, in addition to the above-mentioned judgment based on signal strength, the electromyographic signals of the eye muscles can also be analyzed by a muscle fatigue analysis algorithm to calculate the fatigue level, such as mild, moderate, and severe. The user can be divided into different states according to the fatigue level. For example, in the case of severe and moderate fatigue, it is the first state (tense state). In this case, it is not suitable for visual training. The user's eyes should be massaged to relax first. After massage and relaxation, it is checked again whether the fatigue state is relieved to a state where visual guidance can be performed. If it has reached a state where visual guidance can be performed, visual training is performed. If it is still in a tense state, the massage should continue to relax. If after a certain number of massages, it is still in a state that is not suitable for visual training, a reminder message should be sent to remind the user to pay attention to rest, reduce eye use and turn off the eye massager.
[0049] In one implementation of the present embodiment, controlling the eye massager to perform visual guidance training on the target object includes: determining the actual field of view angle of the target object, collecting the third electromyographic signal of the eye muscle in real time through the electromyographic sensor of the eye massager; calculating the eye movement information of the target object based on the third electromyographic signal; and calculating the completion degree of the visual guidance training based on the actual field of view angle and the eye movement information.
[0050] The actual field of view angle will be different for each user. For the same adult, the field of view angle is stable, but for minors, it will change as they grow. Therefore, for adult users, it is necessary to determine the field of view angle when using it for the first time; for minor users, in addition to determining it during the first use, it needs to be re-determined at regular intervals, such as every 3 months or half a year.
[0051] In one example, determining the actual field of view angle of the target object includes: controlling the visual mark on the head-mounted display screen to move from a reference position in the center of the screen to an edge of the screen, and collecting the extreme movement positions of the target object's eyes when dynamically capturing the visual mark, wherein the extreme movement positions include horizontal extreme positions and vertical extreme positions; and calculating the actual field of view angle of the target object based on the extreme movement positions.
[0052] The actual field of view angle of the user is measured by a dynamic sight mark. The system controls the sight mark on the screen to slowly move from the center of the screen to the edge. The user keeps his head still and tracks the sight mark with his eyes to capture the extreme angles. The process is repeated in all four directions (up, down, left, and right) to obtain the complete horizontal and vertical field of view angles (cycled twice for data calibration).
[0053] Optionally, calculating the actual field of view angle of the target object based on the extreme movement position includes: determining the screen distance from the target object's eyeball to the head-mounted display screen; calculating the actual field of view angle α of the target object using the following formula: α=arctan(Δx / d); wherein d is the screen distance, Δx=x1-x2, x1 is the extreme movement position, and x2 is the reference position.
[0054] The actual field of view angle α includes the actual horizontal field of view angle H and the actual vertical field of view angle V. Taking H as an example, the actual horizontal field of view angle is the horizontal angle from the center position of the eyeball to the current gaze point. The eye tracking device calculates this angle based on the movement trajectory of the eyeball, H = arctan (Δx / d), d is the screen distance from the eye to the screen (the screen is equipped with an infrared ranging sensor), Δx is the horizontal movement distance of the eyeball, Δx = x horizontal limit movement position - x reference horizontal position, and the limit movement position can be determined by user feedback confirmation after reaching the limit point (for example, when reaching the limit point, the user feedbacks that the limit point has been reached through blinking or voice action).
[0055] In one example, calculating the target object's eye movement information based on a third electromyographic signal includes: extracting time domain features and frequency domain features of the third electromyographic signal; combining the time domain features and the frequency domain features into a feature vector; inputting the feature vector into a pre-trained eye movement prediction model to output the target object's gaze coordinates; and calculating the target object's eye movement direction based on the gaze coordinates, wherein the eye movement information includes the eye movement direction.
[0056] After obtaining the third electromyographic signal, it is necessary to preprocess the third electromyographic signal to obtain the corresponding feature vector. The preprocessing process includes filtering, extraction, vector construction, etc. Here is an explanation and description:
[0057] Filtering: Use a bandpass filter (e.g., 20-450 Hz) to remove low-frequency and high-frequency noise to retain the main frequency components of the sEMG signal.
[0058] Denoising: Use a sliding average filter or adaptive noise removal algorithm to reduce noise interference and improve the signal-to-noise ratio.
[0059] Baseline correction: Corrects signal offsets caused by poor electrode contact or changes in skin resistance by subtracting the signal's baseline drift.
[0060] Feature extraction: Extract features from the preprocessed third electromyographic signal for subsequent pattern recognition and coordinate determination, including time domain, frequency domain and time-frequency domain features.
[0061] Time domain characteristics:
[0062] Mean Absolute Value (MAV): N is the number of sampling points of the signal, and x(i) is the signal value at the i-th sampling point, which is used to measure the overall amplitude change of the signal.
[0063] Root Mean Square (RMS): RMS provides an estimate of the signal energy, reflecting the energy level of the signal, which is usually related to the intensity of muscle activity.
[0064] Zero Crossing Rate (ZC): Where Π is an indicator function that returns 1 when the expression is true and 0 otherwise. ZC reflects the frequency characteristics of the signal. It is used to identify high-frequency components in the signal, which are related to the muscle contraction frequency.
[0065] Waveform Length (WL): WL reflects the complexity and rate of change of the signal.
[0066] Frequency domain characteristics:
[0067] Power Spectral Density (PSD): Use Fast Fourier Transform (FFT) to calculate the power spectral density of a signal and analyze the energy distribution of different frequency components. X(f) is the Fourier transform of the signal, T is the time length of the signal;
[0068] Frequency Band Energy (FBE): Calculates the total energy of a signal within a specific frequency range, such as 20-50Hz, 50-150Hz, 150-250Hz, etc. It can be used to analyze EMG activity within a specific frequency band.
[0069] Computes the energy of a frequency band within a specific frequency range [f1,f2].
[0070] Time-frequency domain features:
[0071] Wavelet Transform Coefficients: Wavelet transform is used to extract the time-frequency characteristics of the signal, which is suitable for analyzing the local characteristics of non-stationary signals.
[0072] in is the mother wavelet, a is the scale factor, and b is the translation factor.
[0073] Short-Time Fourier Transform: Perform Fourier transform in segments through window functions to obtain the spectrum distribution of the signal at different time points.
[0074] Among them, ω(τ-t) is the window function and x(τ) is the signal.
[0075] The feature vector constructs a feature combination, combining the time domain, frequency domain, and time-frequency domain features of different dimensions to form a multi-dimensional feature vector. The dimension of the feature vector will be equal to the number of all extracted features.
[0076] You can also use feature selection methods (through principal component analysis PCA, linear discriminant analysis LDA, recursive feature elimination RFE) to reduce feature dimensions, remove redundant and irrelevant features, and improve the generalization ability and computational efficiency of the model.
[0077] The mapping model selected by the eye movement prediction model of this embodiment can be a regression model or a classification model. Regression model: If the goal is to directly predict the coordinates of the gaze point, the model includes linear regression, support vector regression (SVR), neural network (such as multi-layer perceptron MLP). Classification model: If feature vector classification is performed first and then coordinate estimation is performed, models such as SVM and random forest can be used.
[0078] The eye movement prediction model of this embodiment is a prediction model obtained through a large amount of data training. The eye movement prediction model can be built into each eye massager, so that each eye massage device can be used offline to calculate feedback on the training effect. It can also be placed in the cloud, and the data collected by the eye massager is sent to the cloud for calculation, and the cloud feeds back the calculation results to the massage device or user APP, etc.
[0079] Optionally, calculating the eye movement direction of the target object based on the gaze coordinates includes: determining the horizontal radius R of the model field of view of the eye movement prediction model H and vertical radius R V , and determine the horizontal field of view angle θ of the gaze coordinate in the model field of view H and vertical field angle θ V ; Use the following formula to calculate the horizontal rotation angle α H and the vertical rotation angle α V : The gaze coordinate is (X i , Y i ), the eye movement direction includes horizontal rotation angle and vertical rotation angle.
[0080] In this embodiment, before the feature vector is input into the pre-trained eye movement prediction model, the eye movement prediction model may be obtained or trained. In the process of training the eye movement prediction model, the eye movement prediction model training includes: recording the electrical signals caused by eye movement at a known field of view position A (data under several fields of view), and using these data to train the mapping model to accurately predict the fixation point in the field of view in subsequent practical applications.
[0081] During the calibration process, the field of view A is first divided into multiple calibration points, such as grid points evenly distributed in the horizontal and vertical directions. Each calibration point corresponds to a known relative coordinate (x, y). Select multiple calibration points with known positions in the field of view. These points should be evenly distributed throughout the field of view to cover all possible eye movement directions and ranges. Record the electromyographic signal and the gaze point coordinates: When the user gazes at the calibration point (field of view coordinates x, y), record the corresponding sEMG signal.
[0082] Construct calibration dataset: Feature extraction: Extract time domain and frequency domain features from each sEMG signal segment, such as mean absolute value (MAV), root mean square value (RMS), power spectral density (PSD), etc. Construct feature vector: Combine the extracted features into feature vectors and pair them with the corresponding fixation point coordinates (x, y) to form a training dataset.
[0083] Given the relative coordinates of the calibration, the sEMG signals of the eyeballs looking at the calibration points. Based on the known positions of these calibration points, the horizontal and vertical rotation angles α when the eyeballs rotate from the middle position (initial position) to these calibration points are calculated. H , α V . Calculate the horizontal rotation angle α H and the vertical rotation angle α V : The horizontal viewing angle θ of the known viewing position A H 、Vertical field angle θ V , the 0 degree mark of the field of view is directly in front, look at the calibration point (X i ,Y i ), horizontal rotation angle α H The calculation uses the inverse mapping formula Where R H is the horizontal radius of the field of view, X i is the horizontal relative coordinate of the gaze point.
[0084] Vertical rotation angle θ V The calculation uses the inverse mapping formula Where R V is the vertical radius of the field of view, Y i is the vertical relative coordinate of the gaze point.
[0085] In the process of training the eye movement prediction model, the EMG signal is first preprocessed: the feature vector and coordinate data are normalized to ensure the stability and generalization ability of the model training, and the calibration data set is divided into a training set, a validation set, and a test set for segmentation. Model training and validation: The model is trained using the training set, and the model will learn how to map the sEMG signal features to the corresponding gaze point coordinates. The validation set is used to adjust the model hyperparameters to prevent overfitting and optimize the model's prediction ability. The data set is divided into a training set, a validation set, and a test set, with a division ratio of 70% (training set), 15% (validation set), and 15% (test set). The training set is used to train the model, and the validation set is used to adjust the hyperparameters and evaluate the model performance to avoid overfitting.
[0086] The initial model to be trained can be a machine learning model or a deep learning model.
[0087] Machine Learning: Support Vector Machine (SVM): Applicable to classification problems in high-dimensional space, using kernel functions (such as RBF kernel) to handle nonlinear classification. Random Forest: Uses an ensemble of multiple decision trees to improve classification accuracy and stability. K-Nearest Neighbor (k-NN): Classifies based on the majority of neighboring samples, suitable for simple classification tasks.
[0088] Deep learning: Convolutional neural network (CNN): extracts spatial features through convolutional layers, suitable for processing two-dimensional time-frequency graphs or one-dimensional time series data of sEMG signals. Long short-term memory network (LSTM): processes time series data, captures the time dependency and dynamic changes in the signal, and is particularly suitable for predicting eye movement trajectories.
[0089] After the training cycle is reached, the model is evaluated: the model performance is evaluated on the test set, and quantitative evaluation is performed using indicators such as accuracy, precision, recall, F1 score, etc. A confusion matrix can also be generated to analyze the classification performance of the model on different categories.
[0090] In one example, calculating the completion of vision-guided training based on actual field of view angle and eye movement information includes: converting eye movement information into standard eye coordinates in a world coordinate system based on the actual field of view angle; generating eye movement trajectories based on fitting of multiple sets of standard eye coordinates within a training cycle; calculating the offset of the eye movement trajectory relative to the vision-guided trajectory; and calculating the completion of vision-guided training based on the offset, wherein the completion is negatively correlated with the offset.
[0091] Application of eye movement prediction model: It can be used in eye massagers and other devices. Because different devices have different screen sizes.
[0092] Mapping of eyeball relative coordinates: Real-time prediction: Collect the user's electromyographic signal, convert it into a feature vector, input it into the model, and the model outputs the eyeball's coordinates (x, y) in the model's field of view. Then, the predicted coordinates (x, y) are normalized to the standard coordinates (x norm ,y norm ) to be applied on devices of different resolutions or sizes.
[0093] Figure 4 FIG. 1 is a flow chart of visual guidance training by electromyographic signals in an embodiment of the present invention, comprising: obtaining effective eye movement data (electromyographic signals), converting the electromyographic signals into feature vectors, obtaining gaze calibration coordinate points (X i , Y i ), calculate the movement direction according to the gaze calibration point coordinates and the field of view angle size (the field of view angle size preset by the prediction model): horizontal rotation angle α H and the vertical rotation angle α V ; The result is the user's field of view angle, calculate the relative coordinates, and finally combine the visual guidance coordinates to obtain the training effect.
[0094] When evaluating the training effect, the training effect can be divided into excellent, good, and general by calculating the deviation between the standard coordinates and the guide coordinates, and an eye movement track and a guide coordinate track can be displayed. The evaluation criteria for the training effect can be different according to different ages. For example, for young children, the completion standard can be set lower, and for older children, the completion standard can be set higher.
[0095] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.
[0096] Example 2
[0097] In this embodiment, an intelligent control device for an eye massager is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As used below, the term "module" may be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.
[0098] Figure 5 is a structural block diagram of an intelligent control device for an eye massager according to an embodiment of the present invention. Figure 5 As shown, the device comprises:
[0099] A first acquisition module 50, for acquiring a first electromyographic signal of an eye muscle through an electromyographic sensor of the eye massager, wherein the electromyographic sensor is arranged in a contact area between the eye massager and the eye of the target object;
[0100] A first judging module 52, configured to judge whether the eye muscles are in a non-tense state according to the first electromyographic signal;
[0101] The first control module 54 is used to control the eye massager to perform visual guidance training on the target object if the eye muscles are in a non-tense state.
[0102] Optionally, the device also includes: a second control module, which is used to control the eye massager to start a massage mode to massage the eye muscles after the first judgment module judges whether the eye muscles are in a non-tense state according to the electromyographic signal. If the eye muscles are in a tense state, the control module is used to control the eye massager to start a massage mode to massage the eye muscles; a second acquisition module, which is used to collect a second electromyographic signal of the eye muscles through the electromyographic sensor of the eye massager after the eye massager runs in the massage mode for a preset time; a second judgment module, which is used to judge whether the eye muscles are in a non-tense state according to the second electromyographic signal; and a third control module, which is used to control the eye massager to perform visual guidance training on the target object if the eye muscles are in a non-tense state.
[0103] Optionally, the first control module includes: a determination unit, used to determine the actual field of view angle of the target object, and to collect the third electromyographic signal of the eye muscle in real time through the electromyographic sensor of the eye massager; a first calculation unit, used to calculate the eye movement information of the target object according to the third electromyographic signal; and a second calculation unit, used to calculate the completion degree of the visual guided training according to the actual field of view angle and the eye movement information.
[0104] Optionally, the determination unit includes: a control subunit, used to control the sight mark on the head-mounted display screen to move from a reference position in the center of the screen to the edge of the screen, and to collect the extreme movement positions when the target object's eyes dynamically capture the sight mark, wherein the extreme movement positions include horizontal extreme positions and vertical extreme positions; and a calculation subunit, used to calculate the actual field of view angle of the target object based on the extreme movement positions.
[0105] The calculation subunit is also used to: determine the screen distance from the eyeball of the target object to the head display screen; calculate the actual field of view angle α of the target object using the following formula: α=arctan(Δx / d); wherein d is the screen distance, Δx=x1-x2, x1 is the extreme movement position, and x2 is the reference position.
[0106] Optionally, the first computing unit includes: an extraction subunit, used to extract time domain features and frequency domain features of the third electromyographic signal; a combination subunit, used to combine the time domain features and the frequency domain features into a feature vector; an output subunit, used to input the feature vector into a pre-trained eye movement prediction model, and output the gaze coordinates of the target object; and a computing subunit, used to calculate the eye movement direction of the target object based on the gaze coordinates, wherein the eye movement information includes the eye movement direction.
[0107] Optionally, the calculation subunit is further used to: determine the horizontal radius R of the model field of view of the eye movement prediction model H and vertical radius R V , and determining the horizontal viewing angle θ of the gaze coordinate in the model viewing angle H and vertical field angle θ V ; Use the following formula to calculate the horizontal rotation angle α H and the vertical rotation angle α V : Wherein, the gaze coordinate is (X i , Y i ), the eye movement direction includes the horizontal rotation angle and the vertical rotation angle.
[0108] Optionally, the second calculation unit includes: a conversion subunit, used to convert the eye movement information into standard eye coordinates in the world coordinate system based on the actual field of view angle; a generation subunit, used to generate an eye movement trajectory based on multiple sets of standard eye coordinates fitting within a training cycle; a first calculation subunit, used to calculate the offset of the eye movement trajectory relative to the visual guidance trajectory; and a second calculation subunit, used to calculate the completion of the visual guidance training based on the offset, wherein the completion is negatively correlated with the offset.
[0109] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.
[0110] Example 3
[0111] An embodiment of the present invention further provides a storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.
[0112] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:
[0113] S1, collecting a first electromyographic signal of an eye muscle through an electromyographic sensor of an eye massager, wherein the electromyographic sensor is arranged in a contact area between the eye massager and the eye of a target object;
[0114] S2, judging whether the eye muscles are in a non-tense state according to the first electromyographic signal;
[0115] S3, if the eye muscles are in a non-tense state, controlling the eye massager to perform visual guidance training on the target object.
[0116] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.
[0117] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.
[0118] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.
[0119] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:
[0120] S1, collecting a first electromyographic signal of an eye muscle through an electromyographic sensor of an eye massager, wherein the electromyographic sensor is arranged in a contact area between the eye massager and the eye of a target object;
[0121] S2, judging whether the eye muscles are in a non-tense state according to the first electromyographic signal;
[0122] S3, if the eye muscles are in a non-tense state, controlling the eye massager to perform visual guidance training on the target object.
[0123] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.
[0124] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.
[0125] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.
[0126] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.
[0127] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0128] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0129] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.
[0130] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.
Claims
1. An intelligent control method for an eye massager, characterized in that: include: Collecting a first electromyographic signal of an eye muscle through an electromyographic sensor of the eye massager, wherein the electromyographic sensor is arranged in a contact area between the eye massager and the eye of the target object; determining whether the eye muscles are in a non-tense state according to the first electromyographic signal; If the eye muscles are in a non-tense state, the eye massager is controlled to perform visual guidance training on the target object.
2. The method according to claim 1, characterized in that: After determining whether the eye muscles are in a non-tense state according to the electromyographic signal, the method further includes: If the eye muscles are in a tense state, controlling the eye massager to start a massage mode to massage the eye muscles; After the eye massager operates in the massage mode for a preset time, collecting a second electromyographic signal of the eye muscle through the electromyographic sensor of the eye massager; determining whether the eye muscles are in a non-tense state according to the second electromyographic signal; If the eye muscles are in a non-tense state, the eye massager is controlled to perform visual guidance training on the target object.
3. The method according to claim 1, characterized in that Controlling the eye massager to perform visual guidance training on the target object includes: Determine the actual field of view angle of the target object, and collect the third electromyographic signal of the eye muscle in real time through the electromyographic sensor of the eye massager; Calculating eye movement information of the target object according to the third electromyographic signal; The degree of completion of the visual guidance training is calculated according to the actual field of view angle and the eye movement information.
4. The method according to claim 3, characterized in that Determining the actual field of view angle of the target object includes: Controlling the sight mark on the head display screen to move from a reference position in the center of the screen to an edge of the screen, and collecting the extreme movement position when the eyeball of the target object dynamically captures the sight mark, wherein the extreme movement position includes a horizontal extreme position and a vertical extreme position; An actual viewing angle of the target object is calculated based on the limit movement position.
5. The method according to claim 4, characterized in that Calculating the actual field of view angle of the target object based on the extreme movement position includes: Determining a screen distance from an eyeball of the target object to a screen of the head-mounted display; The actual field of view angle α of the target object is calculated using the following formula: α=arctan(Δx / d); Wherein, d is the screen distance, Δx=x1-x2, x1 is the extreme movement position, and x2 is the reference position.
6. The method according to claim 3, characterized in that Calculating the eye movement information of the target object according to the third electromyographic signal includes: Extracting time domain features and frequency domain features of the third electromyographic signal; Combining the time domain features and the frequency domain features into a feature vector; Inputting the feature vector into a pre-trained eye movement prediction model, and outputting the gaze coordinates of the target object; The eye movement direction of the target object is calculated based on the gaze coordinates, wherein the eye movement information includes the eye movement direction.
7. The method according to claim 6, characterized in that Calculating the eye movement direction of the target object based on the gaze coordinates includes: Determine the horizontal radius R of the model field of view of the eye movement prediction model H and vertical radius R V , and determining the horizontal viewing angle θ of the gaze coordinate in the model viewing angle H and vertical field angle θ V ; The horizontal rotation angle α is calculated using the following formulas: H and the vertical rotation angle α V : Wherein, the gaze coordinate is (X i , Y i ), the eye movement direction includes the horizontal rotation angle and the vertical rotation angle.
8. The method according to claim 3, characterized in that Calculating the degree of completion of the visual guidance training according to the actual field of view angle and the eye movement information includes: Converting the eye movement information into standard eye coordinates in a world coordinate system based on the actual field of view angle; Generate eye movement trajectories based on fitting of multiple sets of eye standard coordinates during the training cycle; Calculating the deviation of the eye movement trajectory relative to the visual guidance trajectory; The completion degree of the visually guided training is calculated based on the deviation degree, wherein the completion degree is negatively correlated with the deviation degree.
9. An intelligent control device for an eye massager, characterized in that: include: A first acquisition module, used for acquiring a first electromyographic signal of an eye muscle through an electromyographic sensor of the eye massager, wherein the electromyographic sensor is arranged in a contact area between the eye massager and the eye of the target object; A first judging module, used for judging whether the eye muscles are in a non-tense state according to the first electromyographic signal; The first control module is used to control the eye massager to perform visual guidance training on the target object if the eye muscles are in a non-tense state.
10. A storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program is configured to execute the method according to any one of claims 1 to 8 when executed.
