CN116196196A - Massage armchair control method, system and computer readable storage medium - Google Patents

Abstract

The invention discloses a massage armchair control method, a massage armchair control system and a computer readable storage medium, which belong to the field of intelligent massage chairs. Meanwhile, the massage feedback characteristics of the user are captured and analyzed by using the plurality of set deep learning algorithm models, and massage modes and parameters are adjusted for a plurality of times and the emotion recognition model is optimized by collecting massage feedback data of the user, so that the massage chair is organically combined with the emotion recognition algorithm and the deep learning algorithm, and the massage effect and the comfort level are improved; and meanwhile, the first human physiological data acquired by the sensor is processed by using a deep learning algorithm, so that more accurate second human physiological data output is realized, and the method has certain creativity and application value.

Description

Massage armchair control method, system and computer readable storage medium

Technical Field

The invention belongs to the field of intelligent massage chairs, and particularly relates to a massage chair control method, a massage chair control system and a computer readable storage medium.

Background

At present, the massage armchair is one of important leisure modes in the life of modern people, and has the effects of relieving muscle fatigue, relieving pressure, improving physical health and the like. However, the existing massage chairs still have a plurality of problems such as unsatisfactory massage effect, inconvenient use, inability to be personalized and the like.

The user-defined module of the massage chair meets the basic massage requirement and the personalized requirement of the massage chair, and a user can define a plurality of parameters such as massage force, massage sequence of a massage mechanism, massage duration and the like by operating the interactive display screen, so as to customize a set of massage scheme belonging to the user.

However, during the actual use process, we find that although the massage armchair can perform the massage according to the massage scheme defined by the customer, it cannot adjust the massage force of the massage armchair according to the emotion data fed back by the customer. Therefore, we propose a massage armchair control method, system and computer-readable storage medium to solve the above-mentioned problems.

Disclosure of Invention

In view of one or more of the above drawbacks or improvements in the prior art, the present invention provides a massage armchair control method, system and computer readable storage medium, which utilize a plurality of constructed deep learning models to denoise emotion recognition models before processing data, reconstruct a deep learning model between emotion states and massage parameters, and finally, collect emotion states fed back by a user again after the massage parameters are processed by a massage mechanism, and optimize and secondarily adjust the massage parameters.

In order to achieve the above object, the present invention provides a massage armchair control method, comprising the following steps,

s100: collecting first human physiological data by using a plurality of sensors arranged on the massage armchair;

s200: processing the first human physiological data acquired by the sensor by utilizing a convolutional neural network, and outputting second human physiological data by combining a deep learning algorithm;

s300: judging the second human physiological data by using an emotion recognition algorithm; inputting accurate second human physiological data into an emotion recognition algorithm, classifying the second human physiological data, outputting emotion classification indexes, and constructing an emotion recognition model by using a deep learning algorithm;

s400: the massage parameters which are correspondingly set by the user in the second human physiological data are obtained, emotion recognition and classification indexes are taken as input, the massage parameters are taken as output, a model for controlling the massage force of the massage chair is constructed, and the massage mechanism is controlled to massage the human body;

s500: collecting the first human physiological data again; collecting feedback characteristics of a user after the massage mechanism performs massage operation;

s600: and taking the feedback characteristics of the user as input, inputting the feedback characteristics into the emotion recognition model, constructing a deep learning model with optimized massage strength, and obtaining corresponding massage parameters as output.

As a further improvement of the present invention, in step S100, the following steps are further included: collecting first human physiological data of a user in a relaxed state by using a sensor, taking the first human physiological data as a standard database, and storing the first human physiological data into the standard database; the first human physiological data of the user when in use is collected by the sensor and used as real-time input data.

As a further improvement of the present invention, in step S300, the following steps are further included: the emotion recognition model compares the real-time input data with standard data stored in a standard database, and uses the comparison result for emotion classification.

As a further improvement of the present invention, in step S600, the following steps are further included: acquiring first human physiological data of a massage mechanism within a period of time after massage operation is executed, inputting the first human physiological data into the emotion recognition model, and acquiring emotion classification indexes of a user;

judging the current state of the user according to the first human physiological data; if the user is in a relaxed state, the massage mechanism is continuously controlled to work; and if the user is in a tension state, inputting the feedback characteristic into the emotion recognition model again, and adjusting the massage force of the massage mechanism.

As a further improvement of the present invention, in step S600, the following steps are further included: and taking the massage parameters regulated by the user as input, and taking the feedback characteristics as training data together to construct an accurate massage force optimization model.

As a further improvement of the invention, the first human physiological data comprises sound, facial expression, skin conductivity, heart rate, body temperature or posture.

As a further improvement of the present invention, in step S300, constructing an emotion recognition model specifically includes the steps of:

feature extraction: extracting second human physiological data for judging emotion classification indexes from the second human physiological data;

feature selection: selecting the second physiological data with distinguishing property from the extracted second physiological data of the human body, and reducing the quantity of the second physiological data of the human body;

classification model: and constructing a classification model, and mapping the second human physiological data vector to different emotion categories.

On the basis, the invention also provides a massage armchair control system which comprises a controller, and an emotion recognition module, a sensor, a first deep learning module, a second deep learning module and a third deep learning module which are electrically connected with the controller;

the sensor is arranged on the massage armchair and used for collecting first human physiological data at different positions of a human body;

the first deep learning module is used for receiving the first human physiological data acquired by the sensor, denoising the first human physiological data and acquiring accurate second human physiological data;

the emotion recognition module is used for receiving the processed first human physiological data, constructing a classification model and mapping the second human physiological data to different emotion categories so as to recognize the emotion state of the user;

the second deep learning module is used for receiving the processed emotion states and constructing a massage strength control model taking emotion classification indexes as input and massage parameters as output;

and the third deep learning module is used for collecting feedback characteristics of a user and constructing a massage strength optimization model taking the feedback characteristics as input and the massage parameters as output.

As a further improvement of the invention, the sensor comprises a voice detection sensor, a skin electrical activity sensor, an acceleration sensor, a gyroscope, a blood pressure measurement sensor, a heart rate monitoring sensor and an infrared temperature measurement sensor.

On the basis, the invention also provides a computer readable storage medium, wherein the computer readable storage medium is stored with a computer program, and the computer program realizes the massage chair control method when being executed.

The above-mentioned improved technical features can be combined with each other as long as they do not collide with each other.

In general, the above technical solutions conceived by the present invention have the beneficial effects compared with the prior art including:

(1) The massage armchair control method, the massage armchair control system and the computer readable storage medium adopt first human physiological data of a user through the sensor arranged on the massage armchair, input the first human physiological data into the emotion recognition algorithm, acquire the emotion state of the user, and map the emotion state to corresponding massage parameters, so that intelligent adjustment of massage effect is realized. Meanwhile, the massage feedback characteristics of the user are captured and analyzed by using the plurality of set deep learning algorithm models, and the massage chair is organically combined with the emotion recognition algorithm and the deep learning algorithm by collecting the massage feedback data of the user, adjusting the massage mode and parameters for a plurality of times and optimizing the emotion recognition model.

(2) The massage armchair control method, the system and the computer readable storage medium respectively denoise the emotion recognition model before processing the data by utilizing the constructed multiple deep learning models, reconstruct the deep learning model between the emotion state and the massage parameters, and finally acquire the emotion state fed back by the user again after the massage parameters are massaged by utilizing the massage mechanism, so as to optimize and secondarily adjust the massage parameters.

(3) The massage armchair control method, the massage armchair control system and the computer readable storage medium integrate a plurality of technologies such as sensor acquisition, a deep learning algorithm, an emotion recognition algorithm, massage mechanism control and the like, and organically combine the technologies, so that the massage armchair is controlled based on the emotion recognition and the deep learning algorithm, and massage strength and area of the massage armchair can be automatically adjusted according to the emotion state of a human body, and the massage effect and comfort level are improved. The scheme also processes the human body posture data acquired by the sensor by utilizing a deep learning algorithm, so that more accurate second human body physiological data output is realized, and the method has important significance for the accurate control of the massage armchair. The scheme fully utilizes the current technological means, improves the intelligent degree and the massage effect of the massage armchair, and has certain creativity and application value.

Drawings

FIG. 1 is a schematic overall flow chart of a massage armchair control method according to an embodiment of the invention;

FIG. 2 is a block diagram of the overall logic of a massage chair control method in accordance with an embodiment of the present invention;

fig. 3 is a block diagram of a massage armchair control system according to an embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention. In addition, the technical features of the embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.

In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the device or element being referred to must have a specific orientation, be configured and operated in a specific orientation, and therefore should not be construed as limiting the present invention.

Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present invention, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise.

In the present invention, unless explicitly specified and limited otherwise, the terms "mounted," "connected," "secured," and the like are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally formed; can be mechanically or electrically connected; either directly or indirectly, through intermediaries, or both, may be in communication with each other or in interaction with each other, unless expressly defined otherwise. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.

In the present invention, unless expressly stated or limited otherwise, a first feature "up" or "down" a second feature may be the first and second features in direct contact, or the first and second features in indirect contact via an intervening medium. Moreover, a first feature being "above," "over" and "on" a second feature may be a first feature being directly above or obliquely above the second feature, or simply indicating that the first feature is level higher than the second feature. The first feature being "under", "below" and "beneath" the second feature may be the first feature being directly under or obliquely below the second feature, or simply indicating that the first feature is less level than the second feature.

Examples:

referring to fig. 1 and 2, the massage armchair control method in the preferred embodiment of the present invention is based on a deep learning algorithm and an emotion recognition algorithm, and further optimizes the massage armchair control method to make it more intelligent.

Specifically, the control method includes the steps of:

s100: collecting first human physiological data by using a sensor; the first human physiological data is collected by a plurality of sensors arranged on the massage armchair.

Further specifically, the first human physiological data is acquired with a sensor in step S100, which further includes the steps of: the first human physiological data of the user in a relaxed state is collected by the sensor and is used as a standard database and stored in the standard database. The first human physiological data of the user when in use is collected by the sensor and used as real-time input data.

In a specific embodiment, the standard database parameters can be obtained by voice guidance, so that the user can sit on the massage armchair naturally and loosely without being massaged, and at the moment, the first human physiological data of the user is collected, and the standard database can be used.

Still further, the first human physiological data includes parameters such as sound, facial expression, skin conductivity, heart rate, body temperature or posture; and if the parameters are obtained, the relevant sensors can be set in the massage armchair directly corresponding to the target to be tested, and the creative labor is not required to be paid by the person skilled in the art, so that the method is a conventional technical means which can be mastered by the person skilled in the art.

S200: outputting more accurate second human physiological data based on a deep learning algorithm; and processing the data acquired by the sensor by using algorithms such as convolutional neural network and the like so as to output more accurate second human physiological data.

The method aims to provide more accurate input data for emotion recognition model establishment, reduce the preprocessing process aiming at data in the subsequent emotion recognition model establishment process and reduce the data processing amount.

S300: constructing an emotion recognition model, and judging the second human physiological data by using the emotion recognition model; inputting accurate second human physiological data into the emotion recognition model, classifying the data, constructing the emotion recognition model by using a deep learning algorithm, and outputting emotion classification indexes.

Specifically, in step 300, the following steps are included:

s301: feature extraction: second human physiological data is acquired and data preprocessing, such as noise and outlier removal, is performed to ensure accuracy and consistency of the input data, and data including emotion data, such as sound, images, text, etc., is collected. Data preprocessing:

s302: the collected data is processed and cleaned, noise, abnormal values and the like are removed, and the data is converted into a format which can be processed by a computer.

S303: feature extraction: useful features are extracted from the data, such as frequency, energy, harmonics, etc. of the sound signal, texture, color, etc. of the image, word frequency, grammatical structure, etc. of the text.

S304: feature selection: the most distinguishable characteristic is selected from the extracted characteristics, so that the number of the characteristics is reduced, and the classification accuracy is improved.

And S305, selecting a proper algorithm model, such as a support vector machine, a decision tree, a neural network and the like, according to a specific emotion recognition task, training by using the labeled data set, adjusting model parameters, and improving model classification accuracy.

S306: model evaluation and optimization: the model is evaluated by using an independent test data set, various evaluation indexes such as accuracy, precision, recall, F1 value and the like are calculated, and model optimization is performed according to the evaluation result.

Further specifically, a common way of computation is by encoding emotion tags and converting them into a digital representation, e.g. encoding happiness as 1, sadness as 2, etc. And then processing the collected physiological data, extracting the characteristics, and inputting the characteristics into a machine learning or deep learning model for training so as to predict or classify emotion.

For example, in emotion classification using a support vector machine, the classification result can be calculated using the following formula:

wherein,,

feature vector representing training data, ++>

Label representing training data->

Lagrangian multiplier representing support vector machine,>

representing a kernel function->

Is a bias term.

In deep learning, neural networks may be used for emotion classification. For example, when classifying image emotion using a Convolutional Neural Network (CNN) with a convolution and pooling layer, the convolution operation may be calculated using the following formula:

wherein,,

representing pixel values in the input data, +.>

Is the weight of the convolution kernel, +.>

Is a bias item->

Is an activation function->

Is the output feature image pixel value.

It can be understood that in the prior art, recognition of emotion of a human body is realized based on a plurality of parameters such as voice, image, heart rate and the like, the detailed recognition steps are not described herein, and the acquisition of original data is not a key point to be protected by the invention.

Further, in step S300, the method further includes the steps of: the emotion recognition model compares the real-time input data with the standard data and uses the result for emotion classification; by comparing the difference between the real-time input data and the standard data, whether the current state of the user is relaxed or stressed can be judged, so that the emotion of the user is recognized, namely, after the input data is processed and analyzed, the input data is mapped to different emotion categories, and corresponding labels are output.

S400: constructing a model for controlling the massage force of the massage armchair; and acquiring massage parameters set by a user in the data state, and controlling the massage mechanism to massage the human body by taking emotion recognition and classification indexes as input and massage parameters as output.

Further specifically, in step S400, the following steps are further included, where the collected emotion expression data is processed, for example, converted into a digital signal or the like, so as to facilitate the processing by the neural network. And selecting a proper neural network structure, such as a convolutional neural network, a cyclic neural network and the like according to the requirements, and performing model training. The input is emotion expression data, and the output is massage parameters. Training is carried out by using the marked data set, and the parameters of the model are adjusted so that the massage parameters output by the model are as close as possible to the marked massage parameters.

S500: collecting first human physiological data by using the sensor again; the feedback characteristics of the user after the massage mechanism performs the massage operation are collected.

S600: constructing a deep learning model for optimizing the massage force, and optimizing the massage force; and taking the feedback characteristic of the user as input, inputting the feedback characteristic into the emotion recognition model, and obtaining corresponding massage parameters as output.

In step S600, the method further includes the steps of: acquiring first human physiological data of a massage mechanism within a period of time after massage operation is executed, inputting the first human physiological data into an emotion recognition model, and acquiring emotion classification indexes of a user; judging the current state of the user according to the emotion classification index of the user; if the user is in a relaxed state, the massage mechanism is continuously controlled to work; if the user is in a tension state, the first human physiological data index is input into the emotion recognition model again, and the massage force of the massage mechanism is adjusted.

It can be understood that after the massage parameters are output, each person cannot feel comfortable after receiving the massage operation executed by the massage mechanism due to different physique of the person, so that the emotion of the user needs to be collected for the second time after the user receives the massage operation, so as to realize accurate control operation of the massage chair.

In a specific embodiment, when the user is in a massage state, the first human physiological data is collected through the sensor and is input into the emotion recognition model for emotion classification, so that the current emotion state of the user is obtained. If the user is in a relaxed state, the massage mechanism can continue to work according to the current massage parameters; if the user is in tension, the massage parameters need to be adjusted to achieve better massage effect. The tension of the current state of the user is judged by analyzing the first human physiological data index such as heart rate, respiratory rate, skin electric activity and the like. If the user is in a tension state, the massage force is required to be increased so as to strengthen the massage effect; if the user is in a more relaxed state, the massage force can be properly reduced so as to avoid the uncomfortable feeling of excessive massage. In this process, if the user state is not improved, the first human physiological data is input into the emotion recognition model again to readjust the massage parameters until the user reaches a relaxed state.

Of course, in this process, the user can manually set the massage parameters by himself or herself, and control the massage chair to perform the massage operation.

Further, the user sets the massage region and time, and the deep learning algorithm model can adjust the strength and frequency of the massage mechanism according to the emotion recognition result and the first human physiological data acquired by the sensor, so that the user obtains more personalized massage experience. Meanwhile, a default massage scheme can be provided for the user, the user can adjust on the basis of the default massage scheme to meet the personalized requirements of the user, parameters set by the user can be used as references, and the final massage parameters are comprehensively generated by combining the massage parameters output by the emotion recognition and control algorithm.

Specifically, a priority, such as high, medium, and low levels, is set for the parameters set by the user, and the degree of influence on the massage parameters is determined. And acquiring the emotion state of the user through the emotion recognition model and the first human physiological data acquired by the sensor, and determining the influence degree of the emotion recognition model on the massage parameters. And according to the influence degree of the two factors, combining the massage parameters output by the emotion recognition model to calculate the comprehensive massage parameters. The final massage parameters can be dynamically adjusted according to the feedback of the user and the real-time first human physiological data so as to achieve the optimal massage effect.

Notably, the deep learning model can have a plurality of input values including emotion expressions, user set parameters, and first human physiological data. These input values are converted into corresponding values or feature vectors through specific data processing and preprocessing steps, and are sent to a deep learning model for training or reasoning. During training or reasoning of the model, the deep learning algorithm automatically adjusts the weights and deviations inside the model to map the input data to the correct massage parameter output as much as possible.

On the basis, the invention also provides a massage armchair control system, referring to FIG. 3, which comprises a controller, and an emotion recognition module, a sensor, a first deep learning module, a second deep learning module and a third deep learning module which are electrically connected with the controller;

the sensor is arranged on the massage armchair and used for collecting first human physiological data at different positions of a human body; the first deep learning module is used for receiving the first human physiological data acquired by the sensor, denoising the first human physiological data and acquiring accurate second human physiological data; the emotion recognition module is used for receiving the processed second human physiological data, constructing a classification model and mapping the feature vectors to different emotion categories so as to recognize the emotion states of the user; the second deep learning module is used for receiving the processed emotion states and constructing a massage strength control model taking emotion classification indexes as input and massage parameters as output; and the third deep learning module is used for collecting feedback characteristics of a user and constructing a massage strength optimization model taking the feedback characteristics as input and the massage parameters as output.

Further specifically, the sensors include a voice detection sensor, a skin electrical activity sensor, an acceleration sensor, a gyroscope, a blood pressure measurement sensor, a heart rate monitoring sensor, and an infrared temperature measurement sensor.

On the basis, the invention also provides a computer readable storage medium, wherein the computer readable storage medium is stored with a computer program, and the computer program realizes each step in the massage chair control method when being executed.

(1) The massage armchair control method, the massage armchair control system and the computer readable storage medium collect first human physiological data of a user through the sensor arranged on the massage armchair, input the first human physiological data into the emotion recognition algorithm, acquire the emotion state of the user, and map the emotion state to corresponding massage parameters, so that intelligent adjustment of massage effect is realized. Meanwhile, the massage feedback characteristics of the user are captured and analyzed by using the plurality of set deep learning algorithm models, and the massage chair is organically combined with the emotion recognition algorithm and the deep learning algorithm by collecting the massage feedback data of the user, adjusting the massage mode and parameters for a plurality of times and optimizing the emotion recognition model.

(2) The massage armchair control method, the system and the computer readable storage medium respectively denoise first human physiological data before processing the data by utilizing the constructed multiple deep learning models, then constructs the deep learning model between the emotion state and the massage parameters, finally acquires the emotion state fed back by the user again after the massage parameters are processed by utilizing the massage mechanism, and optimizes and secondarily adjusts the massage parameters.

(3) The massage armchair control method, the massage armchair control system and the computer readable storage medium integrate a plurality of technologies such as sensor acquisition, a deep learning algorithm, an emotion recognition algorithm, massage mechanism control and the like, and organically combine the technologies, so that the massage armchair is controlled based on the emotion recognition and the deep learning algorithm, and massage strength and area of the massage armchair can be automatically adjusted according to the emotion state of a human body, and the massage effect and comfort level are improved. The scheme also processes the first human physiological data acquired by the sensor by utilizing a deep learning algorithm, so that more accurate second human physiological data output is realized, and the method has important significance for the accurate control of the massage armchair. The scheme fully utilizes the current technological means, improves the intelligent degree and the massage effect of the massage armchair, and has certain creativity and application value.

The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (10)

1. A massage chair control method is characterized by comprising the following steps,

s100: collecting first human physiological data by using a plurality of sensors arranged on the massage armchair;

s200: processing the first human physiological data acquired by the sensor by utilizing a convolutional neural network, and outputting second human physiological data by combining a deep learning algorithm;

s300: judging the second human physiological data by using an emotion recognition algorithm; inputting accurate second human physiological data into an emotion recognition algorithm, classifying the second human physiological data, outputting emotion classification indexes, and constructing an emotion recognition model by using a deep learning algorithm;

s400: the massage parameters which are correspondingly set by the user in the second human physiological data are obtained, emotion recognition and classification indexes are taken as input, the massage parameters are taken as output, a model for controlling the massage force of the massage chair is constructed, and the massage mechanism is controlled to massage the human body;

s500: collecting the first human physiological data again; collecting feedback characteristics of a user after the massage mechanism performs massage operation;

s600: and taking the feedback characteristics of the user as input, inputting the feedback characteristics into the emotion recognition model, constructing a deep learning model with optimized massage strength, and obtaining corresponding massage parameters as output.

2. The massage armchair control method according to claim 1, characterized in that it further comprises the following steps in step S100:

collecting first human physiological data of a user in a relaxed state by using a sensor as standard data to form a standard database;

the first human physiological data of the user when in use is collected by the sensor and used as real-time input data.

3. The massage armchair control method according to claim 2, characterized in that it further comprises the following steps in step S300:

the emotion recognition model compares the real-time input data with standard data stored in a standard database, and the comparison result is used for emotion classification.

4. The massage armchair control method according to any of claims 1-3, characterized in that in step S600, it further comprises the steps of:

acquiring first human physiological data of a massage mechanism within a period of time after massage operation is executed, inputting the first human physiological data into the emotion recognition model, and acquiring emotion classification indexes of a user;

judging the current state of the user according to the first human physiological data;

if the user is in a relaxed state, the massage mechanism is continuously controlled to work;

and if the user is in a tension state, inputting the feedback characteristic into the emotion recognition model again, and adjusting the massage force of the massage mechanism.

5. The massage armchair control method of claim 4, further comprising the step of, in step S600:

and taking the massage parameters regulated by the user as input, and taking the feedback characteristics as training data together to construct an accurate massage force optimization model.

6. The massage armchair control method according to claim 1, characterized in that said first human physiological data comprise sound, facial expression, skin conductivity, heart rate, body temperature or posture.

7. The massage armchair control method according to claim 1, characterized in that in step S300, the construction of the emotion recognition model comprises the following steps:

feature extraction: extracting second human physiological data for judging emotion classification indexes from the second human physiological data;

feature selection: selecting the second human physiological data with distinguishing property from the extracted second human physiological data, and reducing the quantity of the second human physiological data;

classification model: and constructing a classification model, and mapping the second human physiological data vector to different emotion categories.

8. The massage chair control system is characterized by comprising a controller, and an emotion recognition module, a sensor, a first deep learning module, a second deep learning module and a third deep learning module which are electrically connected with the controller;

the sensor is arranged on the massage armchair and is used for collecting first human physiological data at different positions of a human body;

the first deep learning module is used for receiving the first human physiological data acquired by the sensor, denoising the first human physiological data and acquiring accurate second human physiological data;

the emotion recognition module is used for receiving the processed second human physiological data, constructing a classification model and mapping the second human physiological data to different emotion categories so as to recognize the emotion state of the user;

the second deep learning module is used for receiving the processed emotion states and constructing a massage strength control model taking emotion classification indexes as input and massage parameters as output;

and the third deep learning module is used for collecting feedback characteristics of a user and constructing a massage strength optimization model taking the feedback characteristics as input and the massage parameters as output.

9. The massage armchair control system of claim 8, wherein the sensor comprises a voice detection sensor, a skin electrical activity sensor, an acceleration sensor, a gyroscope, a blood pressure measurement sensor, a heart rate monitoring sensor, and an infrared temperature measurement sensor.

10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when executed, implements the massage chair control method according to any one of claims 1 to 7.

Images