Intelligent emotion monitoring method and device

Abstract

The invention relates to the technical field of emotion monitoring, and discloses an intelligent emotion monitoring method and device. The current electrocardio potential of the subject is obtained through the flexible electrode; according to the current electrocardio potential, the corrected electrocardio potential is obtained through the correction relation; the correction relation is used for representing the corresponding relation between the electrocardio-potential measured by the flexible electrode and the electrocardio-potential measured by the standard electrode; the corrected electrocardio potential is input into an emotion monitoring model for prediction, and the emotion category of the subject is obtained. Through the correction relation, the current electrocardio potential of the subject is corrected, the corrected electrocardio potential is obtained, the problem that the electrocardio potential is reduced due to impedance is solved, and the emotion monitoring accuracy is improved.

Description

Technical Field

[0001] The present application relates to the technical field of emotion monitoring, and in particular to an intelligent emotion monitoring method and device. Background Art

[0002] In the field of modern mental health, emotion monitoring is becoming increasingly important. Emotion monitoring is a technology that evaluates and identifies an individual's emotional state by analyzing their physiological signals and behavioral manifestations. These physiological signals include heart rate, skin conductance, blood pressure, etc., while behavioral manifestations may include facial expressions, voice intonation, etc. The purpose of emotion monitoring is to provide real-time feedback on the individual's emotional changes, thereby helping them better manage their emotions, especially when facing stress and challenges.

[0003] For patients with mental health problems, high-stress professionals, chronic disease patients, the elderly, highly sensitive people, people experiencing major life events, pregnant and postpartum women, cancer patients and other critically ill patients. These people may have their quality of life affected by mood swings, and need long-term mood monitoring to help them adjust their mood in time and prevent the occurrence of mood disorders.

[0004] In order to enhance the subject's physical sense during long-term emotion monitoring, flexible electrodes are usually used to monitor the ECG potential. However, during data collection, the flexible electrodes do not have close enough contact with the skin, which results in impedance and reduces the ECG potential obtained by monitoring, resulting in poor accuracy of the final monitoring results.

[0005] Therefore, how to meet the comfort level while more accurately collecting the data required for emotion monitoring and improving the accuracy of emotion prediction results has become an urgent problem to be solved. Summary of the invention

[0006] In order to solve the above technical problems, the present application provides an intelligent emotion monitoring method, which comprises the following steps: The subject's current ECG potential is obtained through flexible electrodes; According to the current electrocardiographic potential, a corrected electrocardiographic potential is obtained through a correction relationship; the correction relationship is used to represent the corresponding relationship between the electrocardiographic potential measured by the flexible electrode and the electrocardiographic potential measured by the standard electrode; The corrected ECG potential is input into the emotion monitoring model for prediction to obtain the subject's emotion category.

[0007] Furthermore, a fitted linear function is used as the correction relationship.

[0008] Furthermore, the calculation process of the fitted linear function includes: Obtaining electrocardiographic potentials measured by multiple groups of flexible electrodes and electrocardiographic potentials measured by standard electrodes, and calculating the average electrocardiographic potentials measured by the flexible electrodes and the average electrocardiographic potentials measured by the standard electrodes respectively; According to the ECG potentials measured by each group of flexible electrodes, the ECG potentials measured by each group of standard electrodes, the average ECG potentials measured by the flexible electrodes and the average ECG potentials measured by the standard electrodes, the slope of the fitted linear function is obtained by the least square method; The intercept of the fitted linear function is calculated by fitting the slope of the linear function, the mean value of the ECG potential measured by the flexible electrode and the mean value of the ECG potential measured by the standard electrode; The fitted linear function is determined according to the slope of the fitted linear function and the intercept of the fitted linear function.

[0009] Furthermore, before inputting the corrected ECG potential into the emotion monitoring model, it also includes filtering the corrected ECG potential; The parameter settings of the filtering operation include a passband cutoff frequency wp of 2.8 / fs, a stopband cutoff frequency ws of 1.2 / fs, a passband ripple devel of 0.005, and a stopband ripple Rs of 20.

[0010] Further, the emotion monitoring model is a neural network model; The neural network model includes an input layer, a CNN module, a Transformer module, an average pooling layer and a fully connected layer; the CNN module includes a convolutional layer, a normalization layer, a maximum pooling layer and a residual block.

[0011] On the other hand, the present application also provides an intelligent emotion monitoring device, which performs any of the above-mentioned intelligent emotion monitoring methods, wherein the monitoring device includes a clothing body, an electrode fixing module, a villus electrode and an emotion monitoring module; The clothing body, as a carrier, carries the electrode fixing module, the chorionic villus electrode and the emotion monitoring module; The electrode fixing module is arranged on the side of the clothing body close to the skin of the person being monitored and at the chest position of the clothing body; it is used to fix the chorionic villus electrode to stably contact with the person being monitored; The velvet electrode is fixed to the chest of the clothing body through the electrode fixing module, and is in direct contact with the skin of the person being monitored to collect electrocardiographic potential; the velvet electrode is made by electrostatic self-assembly of a conductive material and a fabric substrate; The emotion monitoring module is connected to the chorionic villus electrodes and is used for performing emotion monitoring according to the electrocardiographic potential.

[0012] Furthermore, the electrode fixing module is provided with an elastic band and two loops, the two loops are symmetrically arranged on both sides of the center of the chest of the clothing body, and the elastic band passes through the loops and is fixed to the clothing body; The villus electrodes are symmetrically arranged on both sides of the center point of the elastic band.

[0013] Furthermore, a metal lead buckle is provided at the center of each of the villus electrodes; The electrocardiogram potential sensor, Bluetooth, LED light, battery and central processing unit are packaged in a box to form the emotion monitoring module; The emotion monitoring module and the villus electrode are connected via a metal lead slot and a metal lead buckle to transmit electrocardiographic potential.

[0014] Furthermore, two of the villus electrodes are fixed on the left and right sides of the center point of the electrode fixing module, and the distance between the center points of the two metal lead buckles is 6 cm.

[0015] Furthermore, the conductive material constituting the velvet electrode is silver nanowires, and the fabric base is a blended 21-strand corduroy fabric.

[0016] The embodiments of the present application have the following technical effects: The intelligent emotion monitoring method in the present application corrects the subject's current ECG potential through a correction relationship before inputting the collected ECG potential into the emotion monitoring model to obtain the corrected ECG potential, thereby overcoming the problem of ECG potential reduction caused by impedance and improving the accuracy of emotion monitoring.

[0017] In the intelligent emotion monitoring device, chorionic villus electrodes are used to collect ECG potential, increase the contact between the electrode and the skin, improve the accuracy of the collected data, and further improve the accuracy of the emotion monitoring results. At the same time, the clothing body is used as a carrier, which not only meets the portability requirements for long-term monitoring, but also provides sufficient carriers for expanding the electrode area, increasing the coverage area of ​​the electrode and the human body, providing more data for emotion monitoring, and assisting in improving the accuracy of the monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 is a schematic diagram of the structure of the intelligent emotion monitoring device provided in an embodiment of the present application; Figure 2is a schematic diagram of the structure of the elastic band provided in an embodiment of the present application; Figure 3 is a schematic diagram of a villus electrode provided in an embodiment of the present application; Figure 4 Schematic diagram showing that the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode provided in the embodiment of the present application have a linear relationship Figure 5 is a structural diagram of an emotion monitoring module provided in an embodiment of the present application; Figure 6 is a graph of the accuracy of the verified emotion monitoring model provided in the embodiment of the present application; Figure 7 is a schematic diagram of the architecture of the emotion monitoring model provided in the embodiment of the present application; Figure 8 is a structural diagram of an emotion monitoring module provided in an embodiment of the present application; Fig. 9 This is a comparison diagram of impedances of villus electrodes of different areas provided in the embodiments of the present application; Fig.10 This is a display diagram of the electrocardiographic potential without filtering operation provided in an embodiment of the present application; Fig.11 It is a display diagram of the ECG potential after filtering operation provided by an embodiment of the present application; Fig.12 This is a flow chart of the intelligent emotion monitoring method provided in the embodiment of the present application.

[0020] 1. Garment body; 2. Elastic band; 3. Velcro; 4. Villus electrode; 5. Metal lead buckle; 6. Emotion monitoring module; 7. Metal lead slot; 8. LED light; 9. Strap; 10. Switch. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work belong to the scope of protection of the present application.

[0022] Based on the numerical problem of impedance reduction monitoring data caused by flexible electrode data collection, this application provides an intelligent emotion monitoring method, such as Fig.12 As shown, the monitoring method comprises the following steps: S1: Obtain the subject’s current ECG potential through flexible electrodes; S2: according to the current ECG potential, a corrected ECG potential is obtained through a correction relationship; the correction relationship is used to indicate the corresponding relationship between the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode; Due to the impedance brought by the flexible electrodes, the value of the monitoring data is reduced. Therefore, the intelligent emotion monitoring method in the present application corrects the current ECG potential of the subject through the correction relationship before inputting the collected ECG potential into the emotion monitoring model to obtain the corrected ECG potential, thereby overcoming the problem of ECG potential reduction caused by impedance and improving the accuracy of emotion monitoring.

[0023] For example, in calculating the correction relationship, the flexible electrode uses a villus electrode and the standard electrode uses a gel electrode. The gel electrode has good contact performance with the skin and its impedance with the skin is negligible, so it is selected as the standard electrode to determine the corresponding relationship between the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode.

[0024] In some embodiments, a fitted linear function is used as the correction relationship, and the calculation process of the fitted linear function includes: S21: obtaining the electrocardiographic potentials measured by n groups of flexible electrodes and the electrocardiographic potentials measured by standard electrodes, and calculating the average values ​​of the electrocardiographic potentials measured by the flexible electrodes and the average values ​​of the electrocardiographic potentials measured by the standard electrodes respectively; ; in, represents the mean value of the ECG potential measured by the flexible electrode, represents the mean value of the ECG potential measured by the standard electrode, x i represents the ECG potential measured by the flexible electrode in the i-th group of data, y i Represents the ECG potential measured by the standard electrodes in the i-th group of data.

[0025] The present application determines whether the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode have a linear relationship by calculating the Pearson correlation coefficient; and determines the difference between the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode by performing an independent sample T test.

[0026] The calculation method of Pearson correlation coefficient is: A1: Calculate the covariance Cov(X,Y) based on the ECG potential measured by the flexible electrode, the ECG potential measured by the standard electrode, the mean of the ECG potential measured by the flexible electrode and the mean of the ECG potential measured by the standard electrode; ; A2: Calculate the standard deviation based on the ECG potential measured by the flexible electrode, the ECG potential measured by the standard electrode, the mean of the ECG potential measured by the flexible electrode and the mean of the ECG potential measured by the standard electrode; ; Among them, S X represents the standard deviation of the ECG potential measured by the flexible electrode, S Y Represents the standard deviation of the ECG potential measured by standard electrodes.

[0027] A3: Calculate the Pearson correlation coefficient r based on the covariance and standard deviation.

[0028] ; The Pearson Correlation Coefficient is a statistic that measures the strength and direction of the linear relationship between two variables. It reflects the degree of covariation between two variables, that is, whether they tend to increase or decrease together. The value of the Pearson correlation coefficient is between -1 and +1. When r=+1, it indicates a complete positive linear correlation, which means that when the value of one variable increases, the value of the other variable also increases at a fixed ratio; when r=−1, it indicates a complete negative linear correlation, which means that when the value of one variable increases, the value of the other variable decreases at a fixed ratio; when r=0, it indicates that there is no linear relationship, but this does not necessarily mean that there is no relationship at all between the two variables, and there may be a nonlinear relationship. The Pearson correlation coefficient r calculated in this application is 0.9954, from which it can be seen whether the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode have a linear relationship.

[0029] The calculation method of the independent sample T test is: B1: Calculate the sample variance based on the ECG potential measured by the flexible electrode, the ECG potential measured by the standard electrode, the mean of the ECG potential measured by the flexible electrode and the mean of the ECG potential measured by the standard electrode; ; Among them, S 2 X represents the sample variance of the ECG potential measured by the flexible electrode, S 2 Y It represents the sample variance of the ECG potential measured by the standard electrode, n and m are the number of data groups of the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode, respectively. In this application, when obtaining the data of the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode in the early stage, n groups are obtained, so here m=n.

[0030] B2: Calculate the combined variance S based on the sample variance of the ECG potential measured by the flexible electrode, the sample variance of the ECG potential measured by the standard electrode, and the number of data sets of the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode. 2 p .

[0031] .

[0032] B3: Calculate the statistic t by combining the variance, the mean of the ECG potential measured by the flexible electrode, the mean of the ECG potential measured by the standard electrode, the number of data sets of the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode: .

[0033] B4: The number of data sets of the ECG potential measured by the flexible electrodes and the ECG potential measured by the standard electrodes, and the degree of freedom df is calculated.

[0034] df=n+m-2.

[0035] B5: The calculated df=1548, based on the statistic t and the degrees of freedom df, for a two-tailed t-test, the significance level α=0.05, and P=0.9437. By looking up the t-distribution table to get the critical value, the critical value of t is approximately ±1.96. This experiment calculated t=-0.0707, combined with P=0.9437>0.05, indicating that the difference is not significant.

[0036] B6: Determine the significance level of the data difference between the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode by comparing the critical value and the critical threshold.

[0037] Independent Samples T-test, also known as two-sample T-test, is a statistical method used to compare whether the mean difference between two independent groups is significant.

[0038] The present application obtains that the data difference between the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode is not significant, and the correction relationship can be expressed by a fitting function. It is previously known whether the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode have a linear relationship, so a first-order fitting function can meet the requirements.

[0039] S22: According to the ECG potentials measured by each group of flexible electrodes, the ECG potentials measured by each group of standard electrodes, the average ECG potentials measured by the flexible electrodes and the average ECG potentials measured by the standard electrodes, the slope k of the fitted linear function is obtained by the least squares method: .

[0040] S23: The intercept b of the fitted linear function is calculated by fitting the slope of the linear function, the mean value of the ECG potential measured by the flexible electrode, and the mean value of the ECG potential measured by the standard electrode: .

[0041] S24: Determine the fitted linear function according to the slope of the fitted linear function and the intercept of the fitted linear function.

[0042] In the present application, the slope k=1.0776, the intercept b=0.0010, and the fitted linear function is expressed as y=1.0776x+0.0010.

[0043] In the same application scenario, flexible electrodes and standard electrodes are used to test ECG electrodes. Figure 4 As shown, the horizontal axis represents the ECG potential measured by the flexible electrode. For example, a chorionic villus electrode is used as the flexible electrode, and the total axis represents the ECG potential measured by the standard electrode. Figure 4 It can be seen that the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode have a linear relationship, and the difference is not large. The relationship between the ECG potential measured by the flexible electrode and the ECG potential measured by the standard electrode can be expressed by a first-order fitting linear function. S3: The corrected ECG potential is input into the emotion monitoring model for prediction to obtain the subject's emotion category.

[0044] In some embodiments, before inputting the corrected ECG potential into the emotion monitoring model, the method further includes filtering the corrected ECG potential; The parameter settings of the filtering operation include passband cutoff frequency wp of 2.8 / fs, stopband cutoff frequency ws of 1.2 / fs, passband ripple devel of 0.005, and stopband ripple Rs of 20. Call the numpy data package in python, and output the passband ripple coefficient Rp according to the passband ripple devel, that is, Rp= 20× np.log10((1 + devel) / (1 - devel)), where np. indicates calling the numpy data package. By using ellipord and ellip in the signal database in python, the order of the elliptic filter and the coefficient of the elliptic filter are calculated respectively, and then the filtering process is performed. Fig.10 This is an ECG potential diagram without filtering provided by the embodiment of the present application, wherein the ECG potential collected has obvious fluctuations in the time period of 40000s-60000s, and Fig.11 This is the ECG potential diagram after filtering operation provided in the embodiment of the present application. The ECG potential fluctuation after filtering operation is not obvious.

[0045] The villus electrode has excellent comfort and conductivity. The villus electrode itself is stable, but when integrated with clothing, it is easy to slip and introduce motion artifacts as the clothing moves. The elliptical filter is used to filter the corrected ECG potential to remove the noise and baseline drift caused by motion artifacts, avoiding interference with emotion recognition in the later stage.

[0046] In some embodiments, the emotion monitoring model is a neural network model; The electrocardiographic potential data were collected under four emotions (happy, calm, sad, and angry) by using the video-induced emotion method.

[0047] Analyze the collected data. First, preprocessing is performed. The preprocessing part includes three steps. The first is to use the fitting linear function y=1.0776x+0.0010 to correct the ECG data measured by the flexible electrode; the second is to remove the noise outside the ECG signal. The specific steps are to obtain the spectrum diagram and analyze it through Fourier transform to eliminate the data outside the target frequency, including baseline drift, power frequency interference, myoelectric interference, etc. Finally, label the data. For the ECG potential data collected under four emotions, corresponding labels are attached: happy = 1, calm = 2, sad = 3, angry = 4. After preprocessing, the data is randomly divided into 80% of the data for model training and 20% of the data for verification. The data is input into the neural network model for model training. The verification accuracy rate reaches 96.25%, completing the construction of the emotion monitoring model. The results of the emotion monitoring model verification are shown in the figure below. Figure 6 shown.

[0048] like Figure 7 As shown in the figure, it is a model of a neural network model, which specifically includes an input layer, a CNN module, a Transformer module, an average pooling layer, and a fully connected layer; the CNN module includes a convolutional layer, a normalization layer, a maximum pooling layer, and a residual block.

[0049] Emotions are judged by ECG potential, which is a one-dimensional time series signal. For this time series signal, the residual module ResidualBlock is introduced on the basis of CNN. Through a sufficiently deep neural network, the spatial characteristics of physiological data between different individuals are learned. At the same time, the Transformer module is introduced on this basis. The Transformer model is a deep learning model based on the self-attention mechanism. It abandons the cyclic structure of RNN and completely relies on the self-attention mechanism to process sequence data. Because the residual module is good at capturing spatial features, and the attention module in the Transformer is good at capturing temporal features, the combination of the two can analyze the emotions of the subjects with high accuracy.

[0050] Transformer captures long-distance dependencies between data through the self-attention mechanism. When this feature is combined with the residual module, the model can not only learn local features, but also understand global information, thereby improving the accuracy of data analysis. And adding Transformer can enhance the robustness of the model, because the Transformer's attention mechanism can help the model focus on the most important features, thereby reducing the impact of noise.

[0051] Transplant the emotion monitoring model to the emotion monitoring module to complete the model construction.

[0052] In order to meet the needs of long-term emotion monitoring anytime and anywhere, and to ensure the accuracy of monitoring, the present application provides an emotion monitoring intelligent device, such as Figure 1 As shown in , the intelligent device includes a clothing body 1, an electrode fixing module, a villus electrode 4 and an emotion monitoring module 6; The clothing body 1, as a carrier, carries the electrode fixing module, the villus electrode 4 and the emotion monitoring module 6; Using the clothing itself as a carrier, while meeting the portability requirements for long-term monitoring, it also provides sufficient carriers for expanding the electrode area, increasing the coverage area between the electrodes and the human body, expanding the area for obtaining physiological information from the body, providing more data for emotion monitoring, and improving monitoring accuracy.

[0053] In some embodiments, a small hole is opened at the center point of the chest of the clothing body 1 for the metal lead buckle 5 and the metal lead slot 7 to be connected and for the wire to pass through.

[0054] In some embodiments, the fabric of the clothing body 1 is composed of nylon and spandex, and the mass ratio of nylon to spandex is 4:1. It has good elasticity, mechanical properties, durability, moisture absorption and breathability, and is comfortable for human body to wear. The good elasticity of the clothing body can also meet the matching requirements of the elastic band, and the cost is low.

[0055] In some embodiments, the clothing body 1 can also be designed to be a tight-fitting style, consisting of two pieces, and the sleeves are designed with raglan sleeves to make the chest of the clothing fit the human body better.

[0056] In some embodiments, the color of the clothing body 1 is black to enhance the color contrast with the LED light 8 and facilitate the identification of emotion types.

[0057] The electrode fixing module is arranged on the side of the clothing body 1 close to the skin of the person being monitored and at the chest position of the clothing body 1; it is used to fix the chorionic villus electrode 4 to stably contact with the person being monitored; In some embodiments, the electrode fixing module is provided with an elastic band 2 and two straps 9, the two straps 9 are symmetrically arranged on both sides of the center of the chest of the clothing body 1, and the elastic band 2 passes through the straps 9 and is fixed to the clothing body 1, such as Figure 2 As shown; The loop belt 9 is used to fix the position of the elastic belt 2 to prevent the elastic belt from slipping and thus separating the villus electrode 4 from the human body, thereby introducing artifacts and interfering with the signal.

[0058] The villus electrodes 4 are symmetrically arranged on both sides of the center point of the elastic band.

[0059] In some embodiments, the elastic band 2 has high elasticity and can adjust the tightness of the user's wear at will. The elastic band 2 is 100 cm long and 4 cm wide, and Velcro 3 is sewn on both ends. Velcro 3 is 8 cm long and 3.5 cm wide, which is used to adjust the length of the elastic band. This allows the villus electrode 4 to fit closely with the human skin, thereby reducing the skin contact impedance and improving the stability and accuracy of emotion monitoring.

[0060] The chorionic villus electrode 4 is fixed to the chest of the clothing body 1 through an electrode fixing module, and is in direct contact with the skin of the person being monitored to collect electrocardiographic potential. The chorionic villus electrode 4 is made by electrostatic self-assembly of a conductive material and a fabric substrate; The electrocardiogram potential evaluates the activity of the autonomic nervous system by analyzing the changes in the intervals between consecutive heartbeats, and then infers the emotional state. In this application, the chorionic villus electrode is fixed to the clothing body through the electrode fixing module, close to the chest, and the electrocardiogram potential is directly collected for emotional monitoring.

[0061] Furthermore, the villus electrode is made by electrostatically self-assembling a conductive material and a fabric substrate. The fabric substrate forms a microscopic structure similar to villus. These structures can penetrate into the tiny depressions of the skin, form a closer contact with the skin, and pass through the sweat hairs on the skin surface, increasing the contact area between the fabric electrode and the skin, thereby reducing the skin contact impedance of the electrode and improving the quality of the ECG potential. The villus electrode is soft and comfortable, with a good touch, which can avoid allergic reactions of the skin when worn for a long time, and meet the needs of long-term wearing and monitoring.

[0062] In some embodiments, a metal lead buckle 5 is provided at the center of each chorionic villus electrode 4 .

[0063] The metal lead buckle 5 is a silver chloride electrode buckle, which is composed of a male buckle and a female buckle. The metal lead buckle 5 is a male buckle, and the metal lead slot 7 is a female buckle, which is used to connect the textile villus electrode 4 and the emotion monitoring module 6. The metal lead buckle 5 is in the shape of a circle with a radius of 0.5 cm, and is fixed at the center point of each villus electrode 4.

[0064] In some embodiments, two villus electrodes 4 are fixed on the left and right sides of the center point of the electrode fixing module, and the distance between the center points of the two metal lead buckles 5 is 6 cm.

[0065] The choroid electrodes 4 are symmetrically fixed on both sides of the center point of the elastic band 2 by sewing, and the center point of the choroid electrodes 4 is 3 cm away from the center point of the elastic band 2 .

[0066] In some embodiments, the area of ​​the villus electrode 4 is 9 cm 2 The shape of the villus electrode 4 is not limited and can be square, circular, etc.

[0067] Fig. 9 is a comparison diagram of the impedance of the villus electrodes of different areas provided in the embodiment of the present application, Fig. 9 It can be seen that the area of ​​the villus electrode 4 is 9 cm 2 The impedance is the smallest when the area is small, and the monitoring effect is better when using a villus electrode of this size.

[0068] In this embodiment, two villous electrodes 4 are provided. However, in practical applications, the number of villous electrodes 4 is not limited to two. One neutral line electrode can be added on the basis of two positive and negative electrodes. Alternatively, 4, 6 or other electrodes can be provided according to the data demand requirement.

[0069] In some embodiments, the conductive material constituting the velvet electrode 4 is silver nanowires, and the fabric substrate is a blended 21-strand corduroy fabric.

[0070] Among them, silver nanowires (AgNWS) are prepared by glycerol, polyvinyl pyrrolidone, sodium chloride, and silver nitrate at a certain concentration ratio. The conductive material and the fabric substrate are electrostatically self-assembled to evenly deposit the silver nanowires (AgNWS) inside and on the surface of the fluff fabric. The fabric has long soft hairs, which increases the contact area with the human body, has better conductivity, and has a suitable tightness, which is conducive to reducing the contact impedance with the skin.

[0071] In some embodiments, the villus electrode 4 is made by electrostatic self-assembly, and the raw materials include 4 mg / ml silver nanowire solution (AgNWs)-conductive, 1 mg / ml polyetherimide solution (PEI), and an area of ​​9 cm 2 21 corduroy fabrics are blended in a square. PEI, as a positively charged polyelectrolyte, can combine with negatively charged surfaces or nanoparticles through electrostatic interaction to form a multilayer composite film, thus realizing the electrostatic self-assembly of the villi electrode 4.

[0072] The corduroy fabric was immersed in a 4 mg / ml silver nanowire solution and placed in a centrifuge for 10 minutes to evenly load the silver nanowire solution on the fabric. After washing with deionized water, it was dried in a vacuum oven. Then, it was immersed in a 1 mg / ml polyetherimide solution and placed in a centrifuge for 10 minutes. After washing with deionized water, it was dried in a vacuum oven to ensure that the generated velvet electrode was dry. At this time, an electrostatic self-assembly was completed. Repeat this operation five times to obtain a velvet electrode. Figure 3 shown.

[0073] The chorionic villus electrode is flexible and, when combined with a clothing carrier made of the same flexible material, it fits better and has a larger contact area with the human body, providing a basis for improving the accuracy of emotion monitoring.

[0074] The emotion monitoring module 6 is connected to the villus electrodes 4 and is used for performing emotion monitoring based on the collected electrical potential.

[0075] In some embodiments, the ECG potential sensor, Bluetooth, LED light 8, battery and CPU are packaged in a box to form the emotion monitoring module 6, such as Figure 7 As shown; the structure of the emotion monitoring module 6 is shown in Figure 5 , Figure 5 The upper middle part of the figure is a schematic diagram of the front side (the side facing away from the clothing body) of the emotion monitoring module 6, and the upper half of the figure is a schematic diagram of the back side (the side in contact with the clothing body) of the emotion monitoring module 6.

[0076] The LED light 8 is used as a feedback device after emotion recognition. For example, it emits yellow light when a happy emotion is recognized, green light when calm, blue light when sad, and red light when angry. The four emotions are visualized by four different colors of light.

[0077] The monitoring results obtained by the emotion monitoring module can be transmitted to computers, mobile phones and other terminal devices via Bluetooth for further visualization and feedback, such as WeChat applet display. Wireless transmission and synchronization of data is achieved, facilitating long-term, real-time and convenient monitoring.

[0078] In some embodiments, the emotion monitoring module 6 may further include a switch 10 for controlling the on and off of the emotion monitoring module 6 .

[0079] A metal lead slot 7 is left outside the emotion monitoring module 6 , and the emotion monitoring module 6 is connected to the chorionic villus electrode 4 through the metal lead slot 7 and the metal lead buckle 5 to transmit the electrocardiographic potential.

[0080] Since the emotion monitoring module 6 is connected to the metal lead buckle 5 via the metal lead slot 7, by undoing the connection between the metal lead slot 7 and the metal lead buckle 5, the emotion monitoring module 6 can be separated from the entire emotion monitoring intelligent device, and can be connected and monitored when there is a monitoring need, and can be detached when there is no monitoring need, so that the clothing body 1 can be cleaned, etc., thereby improving the flexibility of the monitoring device and making it convenient and quick.

[0081] It should be noted that the terms used in this application are only for describing specific embodiments, rather than limiting the scope of this application. As shown in the specification of this application, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of more restrictions, the elements defined by the sentence "include one..." do not exclude the presence of other identical elements in the process, method or device including the elements.

[0082] It should also be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", etc. should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present application.

Claims

1. An intelligent emotion monitoring method, characterized in that: The monitoring method comprises the following steps: The subject's current ECG potential is obtained through flexible electrodes; According to the current electrocardiographic potential, a corrected electrocardiographic potential is obtained through a correction relationship; the correction relationship is used to represent the corresponding relationship between the electrocardiographic potential measured by the flexible electrode and the electrocardiographic potential measured by the standard electrode; The corrected ECG potential is input into the emotion monitoring model for prediction to obtain the subject's emotion category.

2. The intelligent emotion monitoring method according to claim 1, characterized in that: The correction relationship is taken as a fitted linear function.

3. The intelligent emotion monitoring method according to claim 2, characterized in that: The calculation process of the fitted linear function includes: Obtaining electrocardiographic potentials measured by multiple groups of flexible electrodes and electrocardiographic potentials measured by standard electrodes, and calculating the average electrocardiographic potentials measured by the flexible electrodes and the average electrocardiographic potentials measured by the standard electrodes respectively; According to the ECG potentials measured by each group of flexible electrodes, the ECG potentials measured by each group of standard electrodes, the average ECG potentials measured by the flexible electrodes and the average ECG potentials measured by the standard electrodes, the slope of the fitted linear function is obtained by the least square method; The intercept of the fitted linear function is calculated by fitting the slope of the linear function, the mean value of the ECG potential measured by the flexible electrode and the mean value of the ECG potential measured by the standard electrode; The fitted linear function is determined according to the slope of the fitted linear function and the intercept of the fitted linear function.

4. The intelligent emotion monitoring method according to claim 1, characterized in that: Before inputting the corrected ECG potential into the emotion monitoring model, it also includes filtering the corrected ECG potential; The parameter settings of the filtering operation include a passband cutoff frequency wp of 2.8 / fs, a stopband cutoff frequency ws of 1.2 / fs, a passband ripple devel of 0.005, and a stopband ripple Rs of 20.

5. The intelligent emotion monitoring method according to claim 1, characterized in that: The emotion monitoring model is a neural network model; The neural network model includes an input layer, a CNN module, a Transformer module, an average pooling layer and a fully connected layer; the CNN module includes a convolutional layer, a normalization layer, a maximum pooling layer and a residual block.

6. An intelligent emotion monitoring device, executing an intelligent emotion monitoring method according to any one of claims 1 to 5, characterized in that: The monitoring device comprises a clothing body (1), an electrode fixing module, chorionic villus electrodes (4) and an emotion monitoring module (6); The clothing body (1) serves as a carrier, carrying the electrode fixing module, the chorionic villus electrodes (4) and the emotion monitoring module (6); The electrode fixing module is arranged on the side of the clothing body (1) close to the skin of the person to be monitored, and is located at the chest of the clothing body (1); and is used to fix the chorionic villus electrode (4) to ensure stable contact with the person to be monitored. The chorionic villus electrode (4) is fixed to the chest of the clothing body (1) through the electrode fixing module, and is in direct contact with the skin of the person whose emotions are being monitored, so as to collect electrocardiographic potential; the chorionic villus electrode (4) is made by electrostatically self-assembling a conductive material and a fabric substrate; The emotion monitoring module (6) is connected to the chorionic villus electrode (4) and is used to perform emotion monitoring based on the electrocardiographic potential.

7. The intelligent emotion monitoring device according to claim 6, characterized in that: The electrode fixing module is provided with an elastic band (2) and two loop bands (9), the two loop bands (9) being symmetrically arranged on both sides of the center of the chest of the clothing body (1), and the elastic band (2) passing through the loop bands (9) to be fixed to the clothing body (1); The villus electrodes (4) are symmetrically arranged on both sides of the center point of the elastic band.

8. The intelligent device for monitoring emotions according to claim 6, characterized in that: A metal lead buckle (5) is provided at the center of each of the chorionic electrodes (4); The electrocardiographic potential sensor, Bluetooth, LED light (8), battery and central processing unit are packaged in a box to form the emotion monitoring module (6); The emotion monitoring module (6) and the chorionic villus electrode (4) are connected via a metal lead slot (7) and a metal lead buckle (5) to transmit electrocardiographic potential.

9. The intelligent device for emotion monitoring according to claim 8, characterized in that: Two of the villus electrodes (4) are fixed on the left and right sides of the center point of the electrode fixing module, and the distance between the center points of the two metal lead buckles (5) is 6 cm.

10. The intelligent device for emotion monitoring according to claim 6, characterized in that: The conductive material constituting the velvet electrode (4) is silver nanowires, and the fabric base is a blended 21-strand corduroy fabric.

Images