Emotion recognition system and method for smart wearable devices

Abstract

The present invention relates to the field of smart wearable technology, and discloses an emotion recognition system and method for smart wearable devices. The emotion recognition method comprises: S1: wearing a wearable terminal at a designated position of a wearer; S2: a management terminal obtains real-time monitoring data of the wearer through a data acquisition module of the wearable terminal; S3: an administrator sets preset information data of the wearer through a preset information module of the management terminal; S4: the management terminal analyzes the preset information data and real-time monitoring data through an analysis module, and determines the wearer's current emotional state based on the analysis results; through this method, the administrator can obtain the wearer's real-time emotions, improve the reliability of the emotion recognition results of the smart wearable device, and enable the administrator to timely connect to the wearer's emotional changes and promptly process the wearer's emotions accordingly.

Description

Technical Field

[0001] The present invention relates to the field of smart wearable technology, and in particular to an emotion recognition system and method for a smart wearable device. Background Art

[0002] With the improvement of living standards, people are gradually paying more attention to health management, especially mental health management. Among them, emotional monitoring is an important part of mental health management. Adolescence is a stage with greater emotional fluctuations. Faced with multiple challenges such as academic pressure, peer relationships, and family expectations, they may not know how to cope and are prone to negative emotions such as anxiety, depression, and anger. In most families, due to the high pressure of life and busy work and study of parents and teenagers, there is less communication and companionship between parents and teenagers. It is difficult for parents to understand the emotional changes of teenagers in a timely manner, so that they cannot guide the emotions of teenagers in time, which affects their mental health.

[0003] Currently, emotion monitoring can be performed through smart wearable devices. Smart wearable devices usually use sensors to collect sound information or physiological information to perform emotion recognition. However, since each user's voice and physiological state are different, it is difficult to accurately recognize emotions by combining the user's voice and psychological state. Therefore, the reliability of emotion recognition results of existing smart wearable devices is low, making it impossible for parents to accurately understand their children's emotional state and unable to provide timely and correct guidance to their children's emotions. Summary of the Invention

[0004] The purpose of the present invention is to provide an emotion recognition system and method for smart wearable devices to solve the following technical problems:

[0005] How to improve the reliability of emotion recognition results of smart wearable devices.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] An emotion recognition method for a smart wearable device, the emotion recognition method comprising:

[0008] S1: Wear the wearable terminal at the designated position of the wearer;

[0009] S2: The management end obtains the wearer's real-time monitoring data through the wearable end's data collection module;

[0010] S3: The administrator sets the preset information data of the wearer through the preset information module of the management terminal;

[0011] S4: The management end analyzes the preset information data and real-time monitoring data through the analysis module, and determines the wearer's current emotional state based on the analysis results.

[0012] As a further solution of the present invention: the preset information data includes preset emotional states and a plurality of preset emotional state data corresponding to each preset emotional state; wherein, .

[0013] As a further solution of the present invention: the method for the administrator to set the preset information data of the wearer through the preset information module of the management terminal includes the following steps:

[0014] S10: Administrator observes the wearer;

[0015] S20: When the wearer shows a preset emotional state, the administrator stores the wearer's real-time monitoring data within a preset time period after the current moment as preset emotional state data corresponding to the preset emotional state through the preset information module of the management terminal;

[0016] S30: Repeat steps S10 and S20.

[0017] As a further solution of the present invention: the real-time monitoring data includes heart rate and sound; the sound includes real-time human voice and real-time environmental sound; the human voice includes timbre, voice content and intonation; the preset emotional state data includes preset heart rate, preset timbre, preset voice content and preset intonation.

[0018] As a further solution of the present invention: the analysis process of the analysis module includes the following steps:

[0019] S100: obtains real-time heart rate and sound through the data acquisition module;

[0020] S200: Separates real-time human voice and real-time environmental sound by analyzing the sound;

[0021] S300: Analyzing the real-time heart rate and the preset heart rate of each preset emotional state to obtain a heart rate matching index for each preset emotional state;

[0022] S400: Analyzing the timbre, voice content, and intonation of the real-time human voice with the preset timbre, voice content, and intonation of each preset emotional state to obtain a voice matching index for each preset emotional state;

[0023] S500: Analyzing the heart rate matching index and the voice matching index of each preset emotional state to obtain a comprehensive matching index for each preset emotional state;

[0024] S600: Determine the wearer's current emotional state by analyzing the comprehensive matching index of each preset emotional state.

[0025] As a further solution of the present invention: the heart rate matching index includes:

[0026]

[0027] Calculate the Heart rate matching index for preset emotional states ;

[0028] in, is the judgment function, when hour, ;when hour, ; For the Error adjustment coefficients for preset emotional states; is the first allowable error value; For the The number of preset emotional state data obtained for the preset emotional state, ; For the The first of the preset emotional states The start time of obtaining the preset emotional state data; The preset duration; For the The first of the preset emotional states a preset heart rate change curve of the preset emotional state data over time; The real-time heart rate change curve over time; is the current time; is the unit-removing coefficient; is the first preset constant.

[0029] As a further solution of the present invention: Error adjustment coefficient for a preset emotional state By formula: get;

[0030] in, is the preset number of times; is the second preset constant.

[0031] As a further solution of the present invention: the sound matching coefficient is obtained by the formula;

[0032]

[0033] Calculate the Voice matching coefficient of preset emotional states ;

[0034] in, Current time Past preset time The timbre of the real-time human voice in A preset emotional state a maximum timbre matching value of a preset timbre of a preset emotional state data; Current time Past preset time The real-time human voice content in A preset emotional state a maximum voice content matching value of a preset voice content of preset emotional state data; Current time Past preset time The intonation of the real-time human voice in A preset emotional state a maximum pitch matching value of a preset tone of preset emotional state data; is the first weight coefficient; is the second weight coefficient; is the third weight coefficient.

[0035] As a further solution of the present invention: the comprehensive matching index includes:

[0036]

[0037] Calculate the Comprehensive matching index of preset emotional states ;

[0038] in, is the first comprehensive weight coefficient; is the second comprehensive weight coefficient.

[0039] An emotion recognition system for a smart wearable device, the emotion recognition system comprising:

[0040] The emotion recognition system includes a wearable terminal and a management terminal;

[0041] The wearable terminal includes a data acquisition module for collecting real-time monitoring data of the wearer;

[0042] The management terminal includes a preset information module and an analysis module.

[0043] The preset information module is used to obtain the preset information data of the wearer;

[0044] The analysis module is used to analyze preset information data and real-time monitoring data.

[0045] And judge the wearer's current emotional state based on the analysis results.

[0046] Beneficial effects of the present invention:

[0047] The present invention first wears the wearable end at the designated position of the wearer, and then obtains the wearer's real-time monitoring data through the data acquisition module of the wearable end; then the administrator sets the wearer's preset information data through the preset information module of the management end; finally, the management end analyzes the preset information data and the real-time monitoring data through the analysis module, and judges the wearer's current emotional state based on the analysis results; through this method, the administrator can obtain the wearer's real-time emotions, improve the reliability of the emotion recognition results of the smart wearable device, and enable the administrator to timely connect to the wearer's emotional changes and promptly deal with the wearer's emotions accordingly. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The present invention will be further described below with reference to the accompanying drawings.

[0049] Figure 1 A method flow chart of an embodiment of the present invention;

[0050] Figure 2 This is a system module framework diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0052] See also Figure 1 As shown, in one embodiment, a method for emotion recognition of a smart wearable device is provided, the method comprising:

[0053] S1: Wear the wearable terminal at the designated position of the wearer;

[0054] S2: The management end obtains the wearer's real-time monitoring data through the wearable end's data collection module;

[0055] S3: The administrator sets the preset information data of the wearer through the preset information module of the management terminal;

[0056] S4: The management end analyzes the preset information data and real-time monitoring data through the analysis module, and determines the wearer's current emotional state based on the analysis results;

[0057] Through the above technical solution, this embodiment first wears the wearable end at the designated position of the wearer; the wearer can be a child, a teenager or an elderly person; then the wearer's real-time monitoring data is obtained through the data acquisition module of the wearable end; then the administrator sets the wearer's preset information data through the preset information module of the management end; the administrator can be the guardian of the wearer; finally, the management end analyzes the preset information data and real-time monitoring data through the analysis module, and judges the wearer's current emotional state based on the analysis results; through this method, the administrator can obtain the wearer's real-time emotions, improve the reliability of the emotion recognition results of the smart wearable device, and enable the administrator to timely connect to the wearer's emotional changes and promptly deal with the wearer's emotions accordingly.

[0058] As an embodiment of the present invention, the preset information data includes preset emotional states and a plurality of preset emotional state data corresponding to each preset emotional state; wherein, ;

[0059] Through the above technical solution, the preset emotional states in this embodiment may include calm, excited, angry, and upset, etc.

[0060] As an embodiment of the present invention, the method for the administrator to set the preset information data of the wearer through the preset information module of the management terminal includes the following steps:

[0061] S10: Administrator observes the wearer;

[0062] S20: When the wearer shows a preset emotional state, the administrator stores the wearer's real-time monitoring data within a preset time period after the current moment as preset emotional state data corresponding to the preset emotional state through the preset information module of the management terminal;

[0063] S30: Repeat steps S10 and S20;

[0064] Through the above technical solution, this embodiment observes the wearer through the administrator; when the wearer shows a certain preset emotional state, the administrator presses the preset information collection shortcut key on the management end; the real-time monitoring data of the wearer within a preset time after the current moment is stored as preset emotional state data through the preset information module; the administrator then matches the preset emotional state data with the preset emotional state; repeats steps S10 and S20; by collecting the preset emotional state data multiple times, more preset emotional state data is obtained, thereby improving the accuracy of judging the wearer's current emotional state.

[0065] As an embodiment of the present invention, the real-time monitoring data includes heart rate and sound; the sound includes real-time human voice and real-time environmental sound; the human voice includes timbre, voice content and intonation; the preset emotional state data includes preset heart rate, preset timbre, preset voice content and preset intonation;

[0066] The analysis process of the analysis module includes the following steps:

[0067] S100: obtains real-time heart rate and sound through the data acquisition module;

[0068] S200: Separates real-time human voice and real-time environmental sound by analyzing the sound;

[0069] S300: Analyzing the real-time heart rate and the preset heart rate of each preset emotional state to obtain a heart rate matching index for each preset emotional state;

[0070] S400: Analyzing the timbre, voice content, and intonation of the real-time human voice with the preset timbre, voice content, and intonation of each preset emotional state to obtain a voice matching index for each preset emotional state;

[0071] S500: Analyzing the heart rate matching index and the voice matching index of each preset emotional state to obtain a comprehensive matching index for each preset emotional state;

[0072] S600: Determine the wearer's current emotional state by analyzing the comprehensive matching index of each preset emotional state;

[0073] Through the above technical solution, the analysis module of this embodiment obtains real-time heart rate and sound through the data acquisition module; then, by analyzing the sound, real-time human voice and real-time environmental sound are separated; the method of separating human voice and environmental sound is an existing technology and will not be described in detail here; by analyzing the real-time heart rate and the preset heart rate of each preset emotional state, the heart rate matching index of each preset emotional state is obtained; then, by analyzing the timbre, voice content, and intonation of the real-time human voice with the preset timbre, preset voice content, and preset intonation of each preset emotional state, the sound matching index of each preset emotional state is obtained; then, by analyzing the heart rate matching index and sound matching index of each preset emotional state, a comprehensive matching index of each preset emotional state is obtained; finally, by analyzing the comprehensive matching index of each preset emotional state, the wearer's current emotional state is judged; the specific judgment process is to compare the comprehensive matching index of each preset emotional state; the preset emotional state corresponding to the maximum comprehensive matching index is the wearer's current emotional state.

[0074] As an embodiment of the present invention, the heart rate matching index includes:

[0075]

[0076] Calculate the Heart rate matching index for preset emotional states ;

[0077] in, is the judgment function, when hour, ;when hour, ; For the Error adjustment coefficients for preset emotional states; is the first allowable error value; For the The number of preset emotional state data obtained for the preset emotional state, ; For the The first of the preset emotional states The start time of obtaining the preset emotional state data; The preset duration; For the The first of the preset emotional states a preset heart rate change curve of the preset emotional state data over time; The real-time heart rate change curve over time; is the current time; is the unit-removing coefficient; is the first preset constant;

[0078] Through the above technical solution, this embodiment For the The first of the preset emotional states The preset time after the start time of obtaining the preset emotional state data Accumulate preset heart rate value; For the The first of the preset emotional states obtaining a first average preset heart rate value of the preset emotional state data;

[0079] For the Preset emotional state a second average preset heart rate value of the secondary preset emotional state data; Current time Past preset time Accumulated real-time heart rate value; Current time Past preset time Average real-time heart rate value; For the Preset emotional state The second average preset heart rate value of the preset emotional state data and the current time Past preset time The absolute value of the difference between the average real-time heart rate value; For the A permissible error value for a preset emotional state; For the The tolerance value of the preset emotional state and the Preset emotional state The second average preset heart rate value of the preset emotional state data and the current time Past preset time The absolute value of the difference between the average real-time heart rate values ​​within the time interval is the difference; in this formula, the judgment function in Reference , Reference ;when When Preset emotional state The second average preset heart rate value of the preset emotional state data and the current time Past preset time The absolute value of the difference between the average real-time heart rate values ​​within the The allowable error value of the preset emotional state; heart rate mismatch;

[0080] therefore ;when When Preset emotional state The second average preset heart rate value of the preset emotional state data and the current time Past preset time The absolute value of the difference between the average real-time heart rate values ​​in the first The heart rate matches the allowable error value of the preset emotional state; therefore ; and the first Preset emotional state The second average preset heart rate value of the preset emotional state data and the current time Past preset time The absolute value of the difference between the average real-time heart rate value The smaller; Heart rate matching index for preset emotional states The bigger;

[0081] at this time ;

[0082] It should be noted that the first allowable error value , preset duration , remove the unit coefficient and the first preset constant It is a preset value obtained based on experience and will not be described in detail here.

[0083] As an embodiment of the present invention, the Error adjustment coefficient for a preset emotional state By formula: get;

[0084] in, is the preset number of times; is the second preset constant;

[0085] Through the above technical solution, this embodiment For the The difference between the preset number of times of the preset emotional state and the number of preset emotional state data acquisitions; when When The number of preset emotional state data acquisitions for a preset emotional state exceeds the preset number of times ; Therefore There are enough preset emotional state data for the preset emotional state, which can be reduced by The allowable error value of the preset emotional state makes the calculated Heart rate matching index for preset emotional states More accurate; therefore Error adjustment coefficient for a preset emotional state ;when When The number of preset emotional state data acquisitions for a preset emotional state is equal to the preset number of times ;No. Error adjustment coefficient for a preset emotional state ;when When The number of preset emotional state data acquisitions for a preset emotional state is less than the preset number of times ; Therefore If the preset emotional state data for the first preset emotional state is insufficient, you can increase the The allowable error value of the preset emotional state makes the calculated Heart rate matching index for preset emotional states More accurate; Error adjustment coefficient for a preset emotional state ;

[0086] It should be noted that the preset number of times and the second preset constant It is a preset value obtained based on experience and will not be described in detail here.

[0087] As an embodiment of the present invention, the sound matching coefficient is obtained by the formula;

[0088]

[0089] Calculate the Voice matching coefficient of preset emotional states ;

[0090] in, Current time Past preset time The timbre of the real-time human voice in A preset emotional state a maximum timbre matching value of a preset timbre of a preset emotional state data; Current time Past preset time The real-time human voice content in A preset emotional state a maximum voice content matching value of a preset voice content of preset emotional state data; Current time Past preset time The intonation of the real-time human voice in A preset emotional state a maximum pitch matching value of a preset tone of preset emotional state data; is the first weight coefficient; is the second weight coefficient; is the third weight coefficient;

[0091] Through the above technical solution, this embodiment The maximum timbre matching value of a preset emotional state The bigger, Voice matching coefficient of preset emotional states The bigger; The maximum voice content matching value of a preset emotional state The bigger, Voice matching coefficient of preset emotional states The bigger; The maximum pitch matching value of a preset emotional state The bigger, Voice matching coefficient of preset emotional states The bigger;

[0092] It should be noted that the first weight coefficient , the second weight coefficient , the third weight coefficient It is a preset value obtained based on experience. Obtaining timbre matching, voice content matching and pitch matching through two sound segments is a prior art and will not be described in detail here.

[0093] As an embodiment of the present invention, the comprehensive matching index includes:

[0094]

[0095] Calculate the Comprehensive matching index of preset emotional states ;

[0096] in, is the first comprehensive weight coefficient; is the second comprehensive weight coefficient;

[0097] Through the above technical solution, this embodiment Heart rate matching index for preset emotional states The bigger, Comprehensive matching index of preset emotional states The bigger; A sound matching index for a preset emotional state The bigger, Comprehensive matching index of preset emotional states The bigger;

[0098] It should be noted that the first comprehensive weight coefficient and the second comprehensive weight coefficient It is a preset value obtained based on experience and will not be described in detail here.

[0099] See also Figure 2 As shown, an emotion recognition system for a smart wearable device includes a wearable end and a management end;

[0100] The wearable terminal includes a data acquisition module for collecting real-time monitoring data of the wearer;

[0101] The management terminal includes a preset information module and an analysis module.

[0102] The preset information module is used to obtain the preset information data of the wearer;

[0103] The analysis module is used to analyze the preset information data and real-time monitoring data, and determine the wearer's current emotional state based on the analysis results.

[0104] The above is a detailed description of an embodiment of the present invention. However, the content described is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for emotion recognition of a smart wearable device, characterized in that: The emotion recognition method comprises: S1: Wear the wearable terminal at the designated position of the wearer; S2: The management end obtains the wearer's real-time monitoring data through the wearable end's data collection module; S3: The administrator sets the preset information data of the wearer through the preset information module of the management terminal; S4: The management end analyzes the preset information data and real-time monitoring data through the analysis module, and determines the wearer's current emotional state based on the analysis results; The analysis process of the analysis module includes the following steps: S100: obtains real-time heart rate and sound through the data acquisition module; S200: Separates real-time human voice and real-time environmental sound by analyzing the sound; S300: Analyzing the real-time heart rate and the preset heart rate of each preset emotional state to obtain a heart rate matching index for each preset emotional state; S400: Analyzing the timbre, voice content, and intonation of the real-time human voice with the preset timbre, voice content, and intonation of each preset emotional state to obtain a voice matching index for each preset emotional state; S500: Analyzing the heart rate matching index and the voice matching index of each preset emotional state to obtain a comprehensive matching index for each preset emotional state; S600: Determine the wearer's current emotional state by analyzing the comprehensive matching index of each preset emotional state; The heart rate matching index includes: P_n=f(δ_n W_1-|(∑_m▒(∫_(t_mns)^(t_mns+Δt)▒〖C_mn (t)dt〗) / εΔt) / M_n -(∫_(t_x-Δt)^(t_x)▒C(t)dt) / εΔt|) Calculate the heart rate matching index P_n of the nth preset emotional state; Wherein, f(YX) is the judgment function, when YX>0, f(YX)=Z_1 / X; when X≤0, f(X)=0; δ_n is the error adjustment coefficient of the nth preset emotional state; W_1 is the first allowable error value; M_n is the number of preset emotional state data obtained for the nth preset emotional state, 〖m∈M〗_n; t_mns is the start time of the mth acquisition of preset emotional state data for the nth preset emotional state; ∆t is the preset duration; C_m (t) is the preset heart rate change curve of the mth preset emotional state data for the nth preset emotional state over time; C(t) is the real-time heart rate change curve of the real-time heart rate over time; t_x is the current time; ε is the de-unit coefficient; Z_1 is the first preset constant; The error adjustment coefficient δ_n of the nth preset emotional state is obtained by the formula: δ_n=e^((M_0-M_n) / Z_2); Wherein, M_0 is the preset number of times; Z_2 is the second preset constant; The sound matching index is obtained by formula; V_n=γ_1 S_max+γ_2 T_max+γ_3 D_max Calculate the voice matching coefficient V_n of the nth preset emotional state; Wherein, S_max is the maximum timbre matching value between the timbre of the real-time human voice within the preset time length ∆t past the current time t_x and the preset timbre of the M_n preset emotional state data of the nth preset emotional state; T_max is the maximum voice content matching value between the voice content of the real-time human voice within the preset time length ∆t past the current time t_x and the preset voice content of the M_n preset emotional state data of the nth preset emotional state; D_m is the maximum pitch matching value between the intonation of the real-time human voice within the preset time length ∆t past the current time t_x and the preset intonation of the M_n preset emotional state data of the nth preset emotional state; γ_1 is the first weight coefficient; γ_2 is the second weight coefficient; γ_3 is the third weight coefficient; The comprehensive matching index includes: R_n=μ_1 P_n+μ_2 V_n Calculate the comprehensive matching index R_n of the nth preset emotional state; Among them, μ_1 is the first comprehensive weight coefficient; μ_2 is the second comprehensive weight coefficient.

2. The emotion recognition method for a smart wearable device according to claim 1, characterized in that: The preset information data includes N preset emotional states and a plurality of preset emotional state data corresponding to each preset emotional state; wherein n∈N.

3. The emotion recognition method for a smart wearable device according to claim 2, characterized in that: The method for the administrator to set the preset information data of the wearer through the preset information module of the management terminal includes the following steps: S10: Administrator observes the wearer; S20: When the wearer shows a certain preset emotional state, the administrator stores the wearer's real-time monitoring data within a preset time period after the current moment as preset emotional state data corresponding to the preset emotional state through the preset information module of the management terminal; S30: Repeat steps S10 and S20.

4. The emotion recognition method for a smart wearable device according to claim 3, characterized in that: The real-time monitoring data includes heart rate and sound; the sound includes real-time human voice and real-time environmental sound; the human voice includes timbre, voice content and tone; the preset emotional state data includes preset heart rate, preset timbre, preset voice content and preset tone.

5. An emotion recognition system for a smart wearable device, applicable to the emotion recognition method for a smart wearable device according to any one of claims 1 to 4, characterized in that: The emotion recognition system comprises: The emotion recognition system includes a wearable terminal and a management terminal; The wearable terminal includes a data acquisition module for collecting real-time monitoring data of the wearer; The management terminal includes a preset information module and an analysis module. The preset information module is used to obtain the preset information data of the wearer; The analysis module is used to analyze the preset information data and real-time monitoring data, and determine the wearer's current emotional state based on the analysis results.

Images