A method of mood monitoring and apparatus therefor

Abstract

The application discloses an emotion monitoring method and device, and relates to the technical field of embedded development, and comprises the following steps: collecting original physiological parameters of a person to be monitored; performing signal amplification and signal noise reduction processing on the original physiological parameters to obtain processed physiological parameters; inputting the processed physiological parameters into a preset first model to output a first emotion state grade of the person to be monitored; judging whether the first emotion state grade of the person to be monitored is located in a first condition; if not, taking the first emotion state grade as the emotion state grade of the person to be monitored; if yes, performing difference matching on the processed physiological parameters and sample data in a preset database to select a preset database with the minimum absolute value of the difference as the emotion state grade of the person to be monitored; and performing corresponding warning according to the emotion state grade of the person to be monitored; wherein the warning degree comprises no warning, warning degree one and warning degree two. The application can improve the accuracy of emotion judgment.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of embedded development, and particularly relates to an emotion monitoring method and device. BACKGROUND

[0002] Humans have a rich emotional system, especially the fluctuation of emotions, which not only represents the psychological activities of the person, but also can be used to judge some hidden danger signals. Different degrees of emotional fluctuations reflect different degrees of potential signals. If the potential danger signals can be predicted as early as possible through emotional fluctuations, people can take better measures to cope with and prevent the danger before it really comes, so as to avoid events that may cause loss of life and property.

[0003] With the development of natural language processing technology, emotion recognition technology has received more and more attention. In order to monitor the emotions of the monitored people in real time and effectively, at present, the audio or picture or electroencephalogram signal of the monitored person is usually collected to determine the emotion of the monitored person. However, this method cannot determine deeper information, which easily leads to low accuracy and low reliability of emotion monitoring.

[0004] Therefore, it is urgent to use a new monitoring method to improve the accuracy and reliability of monitoring. SUMMARY

[0005] In order to solve the above problems in the prior art, the application provides an emotion monitoring method and device. The technical problems to be solved by the application are solved by the following technical solutions.

[0006] In a first aspect, the application provides an emotion monitoring method, comprising:

[0007] Collecting original physiological parameters of a monitored person; the original physiological parameters include sleep quality signals, heart rate signals, blood pressure signals, skin temperature signals, respiration signals, skin electrical signals and body movement signals;

[0008] Signal amplification and signal noise reduction are performed on the original physiological parameters to obtain processed physiological parameters;

[0009] The processed physiological parameters are input into a preset first model to output a first emotional state level of the monitored person;

[0010] It is judged whether the first emotional state level of the monitored person is located in a first condition. If not, the first emotional state level is taken as the emotional state level of the monitored person. If yes, the processed physiological parameters are matched with sample data in a preset database to select a preset database with the smallest absolute difference value as the emotional state level of the monitored person;

[0011] According to the emotional state level of the person to be monitored, a corresponding warning is given; wherein the warning level includes no warning, warning level one and warning level two.

[0012] Optionally, the processed physiological parameters are input into a preset first model, and the process of outputting the first emotional state level of the person to be monitored includes:

[0013] The range of each physiological parameter is divided into a normal range, a first deviation range and a second deviation range; wherein the first deviation range includes greater than the normal range and less than the normal range, and the second deviation range includes greater than the normal range and less than the normal range.

[0014] According to the range of the physiological parameters, the emotional state level is set to be normal emotion, emotional fluctuation level one and emotional fluctuation level two, respectively.

[0015] Different weight values are assigned to the ranges of different physiological parameters.

[0016] The processed physiological parameters are corresponded to the ranges of the physiological parameters, and the weight values of the processed physiological parameters are assigned based on the weight values of the ranges of the physiological parameters.

[0017] Based on the weight values of the processed physiological parameters, a comprehensive weight value is obtained.

[0018] According to the comprehensive weight value, the first emotional state level is obtained.

[0019] Optionally, before the processed physiological parameters are input into the preset first model, it further includes:

[0020] If the processed physiological parameters change and the change amplitude is greater than 10% of the processed physiological parameters, it is determined that the processed physiological parameters are invalid, and the weight value of the processed physiological parameters is set to a normal value.

[0021] Optionally, before the processed physiological parameters are input into the preset first model, it further includes:

[0022] Based on the motion signal, the heart rate signal and the respiration signal, the physiological parameter range of the human body in different motion states is constructed, which is a light exercise range and an intensive exercise range, respectively.

[0023] According to the physiological parameter range of the human body in different motion states, the weight values corresponding to the motion signal, the heart rate signal and the respiration signal are adjusted and set to a normal value.

[0024] Optionally, the first condition is a fuzzy range, and the boundary of the emotional state level of -5% to 5% is set as the fuzzy range.

[0025] Optionally, it further includes:

[0026] The original physiological parameter verification sample is obtained and processed, the processed physiological parameter verification sample is input into a preset first model, a first emotional state level corresponding to the processed physiological parameter sample is output, it is judged whether the output result is accurate, if not, the weight value corresponding to the processed physiological parameter training sample of the trained neural network model is adjusted.

[0027] Optionally, the preset database includes a normal emotion library, an emotional fluctuation level one library and an emotional fluctuation level two library.

[0028] Optionally, before the weight value corresponding to the processed physiological parameter is differentially matched with the sample data in the preset database, it further includes:

[0029] A special scene database is established; wherein the special scene database includes a slight motion scene, a vigorous motion scene and a washing scene;

[0030] The special scene database is divided into a normal emotion library.

[0031] In a second aspect, the application also provides an emotion monitoring device, comprising:

[0032] A signal acquisition module is configured to acquire original physiological parameters of a person to be monitored; the original physiological parameters include sleep quality signals, heart rate signals, blood pressure signals, skin temperature signals, breathing signals, skin electric signals and body motion signals;

[0033] A signal preprocessing module is configured to perform signal amplification and signal noise reduction processing on the original physiological parameters to obtain processed physiological parameters;

[0034] A signal and emotional state analysis module is configured to input the processed physiological parameters into a preset first model to output a first emotional state level of the person to be monitored;

[0035] A judgment module is configured to judge whether the first emotional state level of the person to be monitored is located in a first condition, if not, the first emotional state level is taken as the emotional state level of the person to be monitored; if yes, the processed physiological parameters are differentially matched with sample data in a preset database, and the preset database with the smallest absolute value of the difference is selected as the emotional state level of the person to be monitored;

[0036] An emotional fluctuation warning module is configured to perform corresponding warning according to the emotional state level of the person to be monitored; wherein the warning degree includes no warning, warning degree one and warning degree two.

[0037] Optionally, the signal acquisition module includes a plurality of sensors, which are heart rate sensors, photoplethysmography sensors, skin temperature sensors, motion sensors, skin electric inductance sensors and angular velocity sensors.

[0038] The present application has the following advantages:

[0039] The mood monitoring method and device provided by the present application can flexibly adjust the selected parameter part in the required input signal, the weight in the algorithm and the algorithm type, can better match the characteristics of the person to be monitored, and improve the accuracy of mood judgment.

[0040] The present application will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 is a flow chart of the mood monitoring method provided by the embodiment of the present application;

[0042] Figure 2 is another flow chart of the mood monitoring method provided by the embodiment of the present application;

[0043] Figure 3 is another flow chart of the mood monitoring method provided by the embodiment of the present application;

[0044] Figure 4 is a structural schematic diagram of the mood monitoring device provided by the embodiment of the present application;

[0045] Figure 5 is a logic schematic diagram of the embedded system provided by the embodiment of the present application;

[0046] Figure 6 is a structural schematic diagram of the hardware framework of the embedded system provided by the embodiment of the present application;

[0047] Figure 7 is a structural schematic diagram of the mood monitoring device provided by the embodiment of the present application;

[0048] Figure 8 is another structural schematic diagram of the mood monitoring device provided by the embodiment of the present application;

[0049] Figure 9 is another structural schematic diagram of the mood monitoring device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0050] The present application will be further described in detail below with reference to the accompanying drawings and embodiments.

[0051] Please refer to Figure 1 , Figure 1 is a flow chart of the mood monitoring method provided by the embodiment of the present application, the mood monitoring method provided by the present application comprises:

[0052] S101, collect the original physiological parameters of the person to be monitored; the original physiological parameters include sleep quality signals, heart rate signals, blood pressure signals, skin temperature signals, respiratory signals, skin electrical signals and body movement signals;

[0053] S102, signal amplification and signal noise reduction processing are performed on the original physiological parameters to obtain processed physiological parameters;

[0054] S103, input the processed physiological parameters into a preset first model, and output the first emotional state level of the person to be monitored;

[0055] S104, judge whether the emotional state level of the person to be monitored is located in the first condition, if not, take the first emotional state level as the emotional state level of the person to be monitored; if yes, difference match the processed physiological parameters with the sample data in the preset database, and select the preset database with the smallest absolute value difference as the emotional state level of the person to be monitored;

[0056] S105, according to the emotional state level of the person to be monitored, corresponding warning is carried out; wherein, the warning degree includes no warning, warning degree one and warning degree two.

[0057] In the embodiment, an emotional monitoring method is provided, which can flexibly adjust the selected parameter part in the required input signal, the weight in the algorithm and the algorithm type, can better fit the characteristics of the person to be monitored, and improve the accuracy of emotional judgment.

[0058] It should be noted that in the process of difference matching the processed physiological parameters with the sample data in the preset database, each group of sample data contains multiple physiological parameters, and each processed physiological parameter has a closest database, that is, the sample parameters in the database are closest to the value (with the smallest difference) of the processed physiological parameter, and it is judged that the parameter falls within the database. However, each parameter in a group of data may fall into different databases, at this time, the database in which each parameter in a group of data falls should be considered, and the database in which the processed physiological parameters are obtained the most is selected as the output result. If there is an equal case, the database with relatively poor emotional state is selected as the output result.

[0059] In an optional embodiment of the present application, please refer to Figure 2 , Figure 2 is another flowchart of the emotional monitoring method provided by the embodiment of the present application, and the process of inputting the processed physiological parameters into a preset first model to output the first emotional state level of the person to be monitored includes:

[0060] The range of each physiological parameter is divided into a normal range, a first deviation range and a second deviation range, wherein the first deviation range includes greater than the normal range and less than the normal range, and the second deviation range includes greater than the normal range and less than the normal range;

[0061] According to the range of the physiological parameter, an emotional state level is set, which is a normal emotion, an emotional fluctuation level one and an emotional fluctuation level two;

[0062] Different weight values are given to the ranges of different physiological parameters;

[0063] The processed physiological parameter is corresponded to the range of the physiological parameter, and the processed physiological parameter is given a weight value based on the weight value of the range of the physiological parameter;

[0064] A comprehensive weight value is obtained based on the weight values of the processed physiological parameters;

[0065] A first emotional state level is obtained corresponding to the comprehensive weight value.

[0066] It should be noted that the emotional state of the person to be monitored is determined by collecting the original physiological parameters of the person to be monitored. Research shows that the normal value of heart rate changes in the range of 60-100 times / min, and the heart rate increases when excited. The normal value of respiratory rate changes in the range of 16-18 times / min, and respiratory rate exceeding 24 times / min is tachypnea, which is common in emotional excitement or excessive tension. The skin conductance level is usually closely related to the degree of emotional fluctuation, and the larger the fluctuation is, the stronger the emotion is, and the smaller the fluctuation is, the weaker the emotion is, which is common in excessive tension and anxiety. Each emotion corresponds to different physiological parameters. The range of the physiological parameter is obtained according to different physiological parameters of different emotions.

[0067] In an optional embodiment of the present application, before the processed physiological parameter is input into the preset first model, it further includes:

[0068] If the processed physiological parameter changes and the change amplitude is greater than 10% of the processed physiological parameter, it is determined that the processed physiological parameter is invalid, and the weight value of the processed physiological parameter is set to a normal value.

[0069] In an optional embodiment of the present application, before the processed physiological parameter is input into the preset first model, it further includes:

[0070] Based on the motion signal, the heart rate signal and the respiratory signal, the physiological parameter range of the human body in different motion states is constructed, which is a light exercise range and an intensive exercise range;

[0071] According to the physiological parameter range of the human body in different motion states, the weight values corresponding to the motion signal, the heart rate signal and the respiratory signal are adjusted and set to a normal value.

[0072] In an optional embodiment of the present application, the first condition is a fuzzy range, and -5% to 5% of the boundary of the emotional state level is set as the fuzzy range.

[0073] In an optional embodiment of the present application, the method further includes:

[0074] Obtain the original physiological parameter verification sample and process it, input the processed physiological parameter verification sample into the preset first model, and output the first emotional state level of the processed physiological parameter sample accordingly, and judge whether the output result is accurate. If it is inaccurate, adjust the weight value corresponding to the processed physiological parameter training sample of the training neural network model.

[0075] In an optional embodiment of the present application, the preset database includes a normal emotion library, an emotion fluctuation level 1 library, and an emotion fluctuation level 2 library.

[0076] In an optional embodiment of the present application, before differentially matching the weight values ​​corresponding to the processed physiological parameters with the sample data in the preset database, the method further includes:

[0077] Establish a special scene database; the special scene database includes light exercise scenes, intense exercise scenes, and washing scenes;

[0078] Divide the special scene database into the normal emotion database.

[0079] In an optional embodiment of the present application, see FIG. Figure 3 As shown, combined with Figure 2 As shown, Figure 3 This is another flow chart of the emotion monitoring method provided by an embodiment of the present invention, which obtains the emotional state of the person to be monitored through the following three methods, specifically:

[0080] (1) Basic algorithm

[0081] Collect the original physiological parameters of the person to be monitored, including sleep quality signals, heart rate signals, blood pressure signals, skin temperature signals, breathing signals, skin electrical signals and body movement signals;

[0082] Perform signal amplification and signal noise reduction on the collected original physiological parameters to obtain purified physiological parameters;

[0083] Divide the range of physiological parameters into normal range, first-level deviation range and second-level deviation range;

[0084] According to the range of physiological parameters, the emotional state level is set, namely normal emotion, emotional fluctuation level 1 and emotional fluctuation level 2; and each physiological parameter range is assigned a weight value;

[0085] According to the correspondence between the processed physiological parameters and the physiological parameter range, a weight value is assigned to the processed physiological parameters, the weight values of the processed physiological parameters are synthesized into a comprehensive weight value, and a first emotional state level is obtained based on the comprehensive weight value;

[0086] In addition, the preset first model also needs to be judged, and the weight values of the corresponding original physiological parameters in the original physiological parameter training sample used for constructing the preset first model are adjusted accordingly, to further adjust the accuracy of the preset first model.

[0087] (2) Matching library algorithm

[0088] The original physiological parameters of the person to be monitored are collected, including sleep quality signals, heart rate signals, blood pressure signals, skin temperature signals, breathing signals, skin electrical signals, and body movement signals;

[0089] The collected original physiological parameters are subjected to signal amplification and signal noise reduction processing to obtain purified physiological parameters;

[0090] A preset database is constructed, including a normal emotion library, an emotional fluctuation level one library, and an emotional fluctuation level two library; each database contains a large number of physiological data groups corresponding to the corresponding emotions of the human body under normal circumstances, and the emotional fluctuation level one library and the emotional fluctuation level two library are distinguished according to the size of the emotional fluctuation degree of the human body; optionally, the sample data in the normal emotion library, the emotional fluctuation level one library, and the emotional fluctuation level two library is not less than 30 groups;

[0091] After constructing the preset database, the accuracy of the preset data also needs to be adjusted, which can be achieved by adjusting the weight values of the physiological parameters corresponding to the physiological parameters used for training the preset database;

[0092] The processed physiological parameters are matched with the sample data in the database, and the preset database with the smallest absolute difference value is selected as the output emotional state level of the person to be monitored.

[0093] (3) Hybrid algorithm

[0094] First, the emotional state level of the person to be monitored is obtained by using the above basic algorithm, and the weight range of the emotional state level in the basic algorithm is re-allocated, and the intersection line of the two different emotional state level ranges is divided to re-judge the 5% interval range corresponding to the two intervals, which is the fuzzy interval;

[0095] The fuzzy interval is re-judged by using the library matching algorithm to obtain the emotional state level of the person to be monitored.

[0096] The method can be adjusted according to the original physiological parameters of the to-be-detected person, the physiological data that is not necessary to collect can be selected or not, the emotion state grade range can be divided, and the weight value of the involved parameter can be adjusted, so that the matching degree of the algorithm with the to-be-monitored person is improved, the accuracy of the judgment is improved, and the individual differences are met.

[0097] Based on the same inventive concept, please refer to Figure 4 , Figure 4 is a structural schematic diagram of an emotion monitoring device provided by an embodiment of the present application. The present application also provides an emotion monitoring device for implementing the emotion monitoring method provided by the above-mentioned embodiments of the present application. The present application will not be described here again. The emotion monitoring device provided by the present application comprises:

[0098] The signal acquisition module 201 is configured to acquire the original physiological parameters of the to-be-monitored person. The original physiological parameters comprise a sleep quality signal, a heart rate signal, a blood pressure signal, a skin temperature signal, a breathing signal, a skin electricity signal and a body movement signal.

[0099] The signal preprocessing module 202 is configured to perform signal amplification and signal noise reduction processing on the original physiological parameters to obtain processed physiological parameters.

[0100] The signal and emotion state analysis module 203 is configured to input the weight value corresponding to the processed physiological parameters into a preset first model to output a first emotion state grade of the to-be-monitored person.

[0101] The judgment module 204 is configured to judge whether the first emotion state grade of the to-be-monitored person is located in a first condition. If not, the first emotion state grade is taken as the emotion state grade of the to-be-monitored person. If yes, the weight value corresponding to the processed physiological parameters is matched with sample data in a preset database to select a preset database with the smallest absolute value of the difference as the emotion state grade of the to-be-monitored person.

[0102] The emotion fluctuation warning module 205 is configured to perform corresponding warning according to the emotion state grade of the to-be-monitored person. The warning degree comprises no warning, warning degree one and warning degree two.

[0103] Specifically, the emotion monitoring device provided in the embodiment can accurately identify the emotion fluctuation degree and the emotion state of the to-be-monitored person, can send warning information to the to-be-monitored person or a person who cares about the to-be-monitored person, can timely infer potential danger signals through emotion fluctuation, can timely and effectively take measures to cope with and prevent, and thus the loss of life and property can be avoided.

[0104] In an optional embodiment of the present application, the signal acquisition module includes a plurality of sensors, namely, a heart rate sensor, a photoplethysmography sensor, a skin temperature sensor, a motion sensor, a skin inductance sensor, and an angular velocity sensor.

[0105] Please refer to Figures 5-6 , as shown in the figure, Figure 5 is a logical diagram of an embedded system provided by an embodiment of the present application, Figure 6 is a structural diagram of a hardware framework of an embedded system provided by an embodiment of the present application, as shown in the figure, Figure 5 , wherein the central processing unit (CPU) drives various peripherals through a bus, and the peripherals can be connected to each other through a UART communication protocol, a GPIO communication protocol, a USB, and the like. The functional modules include an emotion recognition module, a warning signal module, a power management module, and a database module. Each module is connected to a peripheral device related thereto through a communication interface. The peripheral devices include sensors for signal acquisition, an information sending port for communication with the outside world, a display, and the like. In this way, signal acquisition of the sensors, sending of the warning signal, and data display on the display can be achieved. As shown in the figure, Figure 6 , different peripheral devices need to be controlled and allocated by the processor. The processor and the peripheral devices with different functions work together to form an embedded system as a whole. The processor can be a central processing unit (CPU), and the peripheral devices can be a sensor 1 connection port, a sensor 2 connection port, …, a sensor n connection port, a display screen, a USB interface, a system memory (SDRAM), a hard disk (non-linear Flash), a power management system, a general packet radio service (GPRS), and a global positioning system (GPS).

[0106] In an optional embodiment of the present application, the device, module, or unit mentioned in the above embodiments can be implemented as a computer chip or an entity, or by a product with certain functions. A typical implementation device is a computer, and the specific form of the computer can be a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email transceiver device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0107] The embodiments in the present application can be in the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Moreover, the embodiments in the present application can be in the form of a computer program product including one or more computer usable storage media (including but not limited to disk storage, CD-ROMs, optical storage media, and the like) having computer usable program code embodied therein.

[0108] Please refer to Figures 7-9 ,Figure 7 is a structural schematic diagram of an emotion monitoring device provided by an embodiment of the present application, Figure 8 is another structural schematic diagram of an emotion monitoring device provided by an embodiment of the present application, Figure 9 is another structural schematic diagram of an emotion monitoring device provided by an embodiment of the present application, in an optional embodiment of the present application, the emotion monitoring device comprises a hand strap 2, a hand ring body and a buckle, the hand ring body is installed in the middle of the hand strap, and the buckle assembly is installed at both ends of the hand strap.

[0109] The hand ring body comprises a signal acquisition module 9, a communication module 7, a GPS module 10, an embedded development module and a battery. The signal acquisition module comprises a plurality of sensors for acquiring physiological parameters of a person to be monitored. The communication module is connected with a SIM card and acquires information of the SIM card, accesses a mobile network through the module to realize information interaction. The GPS module can acquire position information of the hand ring. The embedded development module processes the acquired physiological parameters of the person to be monitored and acquires an emotional state of the person to be monitored, and makes a corresponding warning according to the emotional state level. The battery is electrically connected with the signal acquisition module, the communication module, the GPS module and the embedded development module to supply power for the above modules. An opening 6 is provided in cooperation with the battery 5 to charge the battery.

[0110] In an optional embodiment of the present application, the hand ring body further comprises a socket and a charging prompt lamp, which are electrically connected with the battery. The socket can charge the battery, and the charging prompt lamp is used to reflect the charging state when charging.

[0111] In an optional embodiment of the present application, the hand ring body further comprises a display screen 1 electrically connected with the embedded development module for displaying the acquired original physiological parameters. The embedded development module comprises an embedded development chip 3 and an embedded development board 4, and the embedded development chip is electrically connected with the embedded development board.

[0112] In an optional embodiment of the present application, the hand ring body further comprises a SIM card seat 8 electrically connected with the communication module for placing a SIM card to enable the communication module to realize information interaction with the SIM card.

[0113] In an optional embodiment of the present application, the hand ring body and the hand ring body mounting part in the hand strap are matched oval shapes. The material of the watchband is selected to be elastic silica gel, the size of the mounting part is the same as that of the hand ring body, and the base part of the mounting part has a hollow part. The hollow part is used for the back of the hand ring body to contact the skin of the person to be monitored.

[0114] In an optional embodiment of the present application, the signal acquisition module is integrated on the back of the hand ring body, and a part of the signal acquisition module is located on the surface of the hand ring body to realize the function of signal acquisition.

[0115] In an optional embodiment of the present application, the embedded development module can undertake the function of data processing, can make a judgment according to the input physiological parameters, and is located inside the bracelet main body and directly connected with other modules. The embedded module is located below the display screen and above the signal acquisition module.

[0116] In an optional embodiment of the present application, the battery is located inside the bracelet main body, below the embedded development module, and close to the edge of the bottom layer of the bracelet main body.

[0117] In an optional embodiment of the present application, the buckle for adjusting the wearing size and the groove matched with the buckle, and the part of the hand strap connected with the bracelet main body can satisfy the functional use of the bracelet main body.

[0118] In the above embodiments, after the bracelet is configured with a SIM card, information interaction between the communication module built-in the bracelet and the mobile base station is matched, and when the emotional fluctuation of the person to be detected is abnormal, warning information is sent to the pre-designated object. Each module contained in the bracelet can constitute an emotion detection system. The system is equipped with a physiological parameter acquisition module of a human body in the device, analyzes and classifies the real-time emotion of the person to be monitored according to the collected physiological parameters, and activates the GPS module and the communication module if the warning program is triggered.

[0119] The emotion monitoring method and device provided by the present application can flexibly adjust the selected parameter part in the required input signal, the weight in the algorithm, and the algorithm type, can better fit the characteristics of the person to be monitored, and improves the accuracy of emotion judgment.

[0120] The above is a further detailed description of the present application in combination with specific preferred embodiments, and the specific implementation of the present application cannot be limited to these descriptions. For ordinary skilled persons in the technical field to which the present application belongs, some simple deductions or replacements can be made without departing from the concept of the present application, and all of them should be regarded as falling within the protection scope of the present application.

Claims

1. A method of mood monitoring, characterized by, The method comprises the following steps: Collecting original physiological parameters of a person to be monitored; The original physiological parameters include sleep quality signals, heart rate signals, blood pressure signals, skin temperature signals, respiration signals, skin electrical signals, and body movement signals; Signal amplification and signal noise reduction processing are performed on the original physiological parameters to obtain processed physiological parameters; The processed physiological parameters are input into a preset first model to output a first emotional state level of the person to be monitored, comprising: Dividing the range of each physiological parameter into a normal range, a first deviation range, and a second deviation range; wherein the first deviation range includes a range greater than the normal range and a range less than the normal range, and the second deviation range includes a range greater than the upper limit of the first deviation range and a range less than the lower limit of the first deviation range; According to the range of the physiological parameters, set the emotional state level to be normal emotion, emotional fluctuation level one, and emotional fluctuation level two, respectively; Different weight values are assigned to different ranges of the physiological parameters; The processed physiological parameters are corresponded to the ranges of the physiological parameters, and the weight values of the ranges of the physiological parameters are assigned to the processed physiological parameters based on the weight values of the ranges of the physiological parameters; Based on the weight values of each processed physiological parameter, a comprehensive weight value is obtained; According to the comprehensive weight value, a first emotional state level is obtained; Before the processed physiological parameters are input into the preset first model, the method further comprises the following steps: Based on the movement signal, the heart rate signal, and the respiration signal, the range of the physiological parameters of the human body under different movement states is constructed, which is a light movement range and a strenuous movement range, respectively; According to the range of the physiological parameters of the human body under different movement states, the weight values corresponding to the movement signal, the heart rate signal, and the respiration signal are adjusted and set to normal values; Determine whether the first emotional state level of the person to be monitored is located in a first condition, if not, the first emotional state level is taken as the emotional state level of the person to be monitored; if so, the processed physiological parameters are matched with the sample data in the preset database to select the preset database with the smallest difference absolute value as the emotional state level of the person to be monitored; the first condition is a fuzzy range, and the boundary of the emotional state level is set to a fuzzy range of -5% to 5%; According to the emotional state level of the person to be monitored, a corresponding warning is given; wherein the warning degree includes no warning, warning degree one, and warning degree two.

2. The method of claim 1, wherein, Before the processed physiological parameters are input into the preset first model, the method further comprises the following steps: If the processed physiological parameters change and the change amplitude is greater than 10% of the processed physiological parameters, it is determined that the processed physiological parameters are invalid, and the weight value of the processed physiological parameters is set to a normal value.

3. The method of claim 1, wherein, The method further comprises the following steps: Obtain the original physiological parameter verification sample and process it, input the processed physiological parameter verification sample into the preset first model, and correspondingly output the first emotional state level of the processed physiological parameter sample, and determine whether the output result is accurate, if not, adjust the weight value corresponding to the processed physiological parameter training sample of the training neural network model.

4. The method of claim 1, wherein, The preset database includes a normal emotion database, an emotional fluctuation level one database, and an emotional fluctuation level two database.

5. The method of claim 1, wherein, Before the processed physiological parameters corresponding to the weight values are differentially matched with the sample data in the preset database, it further includes: A special scene database is established; wherein, the special scene database includes a slight motion scene, an intense motion scene and a washing-up scene; The special scene database is divided into a normal emotion library.

6. An emotion monitoring apparatus characterized by, It includes: A signal acquisition module for acquiring the original physiological parameters of the person to be monitored; The original physiological parameters include sleep quality signals, heart rate signals, blood pressure signals, skin temperature signals, respiration signals, skin electrical signals and body motion signals; A signal preprocessing module for signal amplification and signal noise reduction processing of the original physiological parameters to obtain processed physiological parameters; A signal and emotion state analysis module for inputting the processed physiological parameters into a preset first model to output a first emotion state level of the person to be monitored, including: Dividing the range of each physiological parameter into a normal range, a first deviation range and a second deviation range; wherein, the first deviation range includes greater than the normal range and less than the normal range, and the second deviation range includes greater than the upper limit of the first deviation range and less than the lower limit of the first deviation range; According to the range of the physiological parameters, set the emotion state level, respectively, as normal emotion, emotion fluctuation level one and emotion fluctuation level two; Different weight values are given to the ranges of different physiological parameters; The processed physiological parameters are corresponded to the ranges of the physiological parameters, and the weight values of the ranges of the physiological parameters are based on the weight values of the processed physiological parameters; Based on the weight values of each processed physiological parameter, a comprehensive weight value is obtained; According to the comprehensive weight value, a first emotion state level is obtained; Before the processed physiological parameters are input into the preset first model, it further includes: Based on the motion signal, the heart rate signal and the respiration signal, the range of the physiological parameters of the human body in different motion states is constructed, respectively, as a slight motion range and an intense motion range; According to the range of the physiological parameters of the human body in different motion states, the weight values corresponding to the motion signal, the heart rate signal and the respiration signal are adjusted and set to normal values; A judgment module for judging whether the first emotion state level of the person to be monitored is located in a first condition, if not, the first emotion state level is taken as the emotion state level of the person to be monitored; if yes, the processed physiological parameters are differentially matched with the sample data in the preset database, and the preset database with the smallest absolute difference value is selected as the emotion state level of the person to be monitored; the first condition is a fuzzy range, and the boundary of the emotion state level is set to a fuzzy range of-5% to 5%; An emotion fluctuation warning module for making corresponding warnings according to the emotion state level of the person to be monitored; wherein, the warning degree includes no warning, warning degree one and warning degree two.

7. A mood monitoring device according to claim 6, wherein, The signal acquisition module includes a plurality of sensors, respectively, a heart rate sensor, a photoplethysmography sensor, a skin temperature sensor, a motion sensor, a skin inductance sensor and an angular velocity sensor.

Images