KR20020018541A - Apparatus and method for perceiving physical and emotional state - Google Patents

Abstract

PURPOSE: An apparatus and a method for recognizing the human body and an emotional status are provided to improve the technique for recognizing the human body and the emotional statues by detecting a biological signal in a convenient manner. CONSTITUTION: A biological signal detecting unit(100) is mounted on a predetermined part of the human body and converts at least one biological signal detected from the human body into an analog signal to output it. A biological signal recognizing unit(120) processes the biological signal from the biological signal detecting unit(100) and recognizes the human body and emotional status based on the processed result. The biological signal detecting unit(100) comprises a sensor(10) for detecting the biological signal, an analog signal processor(20) for amplifying and filtering the biological signal detected by the sensor(10), and an analog/digital converter(30) for converting the amplified and filtered biological signal from an analog signal into a digital signal. The biological signal recognizing unit(120) comprises a digital signal processor(40) for processing the biological signal in the form of digital, a controller(50) for recognizing the human body and emotional status and outputting the recognized result as a control signal, and a data output module(60) for outputting data in a predetermined form in response to the control signal.

Description

Apparatus and method for perceiving physical and emotional state

TECHNICAL FIELD The present invention relates to the use of a biosignal, and more particularly, to an apparatus and a method for recognizing a body and an emotional state using a biosignal to recognize a state of a body and an emotional state.

The detection or measurement of human biological signals has been studied and utilized primarily in the medical field in the early days. That is, a biosignal detected or measured is used to check a patient's health or to treat a patient by measuring blood pressure or pulse rate. In recent years, biosignals have been utilized not only in the medical field, but also in the field of human-machine interface to detect human emotional state. In particular, in order to provide a more convenient and friendly communication means for the interface between a human and a machine, it is emerging as essential to recognize human emotion using a biosignal. In response to this, various devices for detecting or measuring a biosignal (or a physiological signal) and using the same have been developed.

As a portable bio-signal measuring device of such conventional devices, an ear-receiver type optical-electric-pulse-plessimo-graph (PPG) disclosed in Korean Patent Laid-Open No. 1999-63100. PlethysmoGraph (PlethysmoGraph) sensor and "biological information measuring device" and the Republic of Korea Patent Publication No. 1997-14722 "real-time biological signal monitoring system using a wireless communication network" and the like. However, such conventional devices have a problem in that it is not easy to attach to the body depending on the gender or inconveniences the user's life when attached.

On the other hand, as an emotion recognition device of the biological signal measuring apparatus, "Method and an associated apparatus for remotely determining information as to person's emotional state" disclosed in US Patent No. US5,507,291 and "Interactive Man- disclosed in US Patent No. US5,367,454" Machine interface for simulating human emotion ". However, such conventional devices also have a problem in that it is inconvenient to attach to the body and inconvenience to the action when attaching to the body.

Also, when a conventional emotion recognizing apparatus recognizes human emotions, the human being must, for example, have to face his or her face toward the camera or to express his or her emotions so that the facial expression can be easily seen. Such artificial efforts should be made. As such, humans have a problem in that their emotions cannot be naturally transmitted to conventional emotion recognition devices. In addition, the conventional emotion recognition apparatus has a problem with many limitations, such as not only ambiguously recognize emotion but also assume only a response when a certain emotional stimulus is transmitted.

The first technical problem to be achieved by the present invention is to provide a body and emotional state recognition device that can recognize the body and emotional state by detecting a bio-signal, which is easy to wear and detach on the body and simply.

The second technical problem to be achieved by the present invention is to provide a body and emotional state recognition method that can recognize the body and emotional state by detecting a bio-signal, which is easy to wear and detach on the body and simply.

A third object of the present invention is to provide an emotional state recognition method capable of recognizing an emotional state using biometric parameter (s) extracted from a biological signal detected for a short or long time.

1 is a schematic block diagram of a body and emotional state recognition apparatus according to the present invention.

FIG. 2 is a flowchart for explaining a method of recognizing a physical and emotional state according to the present invention, which is performed in the apparatus shown in FIG. 1.

3 is a flowchart for explaining a preferred embodiment of the present invention with respect to step 212 shown in FIG.

4 is a flowchart for explaining an emotional state recognition method according to the present invention.

FIG. 5 is a flowchart for explaining a preferred embodiment of the present invention with respect to step 302 shown in FIG.

6 is a flowchart for explaining another preferred embodiment of the present invention for the step 302 shown in FIG.

7 is a flowchart for explaining a preferred embodiment of the present invention with respect to the embodiment shown in FIG.

8 is a flowchart for explaining a preferred embodiment of the present invention with respect to the embodiment shown in FIG.

The body and emotion state recognition apparatus according to the present invention for achieving the first object is a biological signal detection unit and the biological body mounted on a predetermined portion of the body for analog signal processing and outputting at least one biological signal detected from the body It is preferable that the biosignal processed by the signal detection unit is a digital signal processing, and the biosignal recognition unit for recognizing and expressing the physical and emotional state from the signal-processed result.

In accordance with another aspect of the present invention, there is provided a method of recognizing a body and an emotional state, the method including analog signal processing of at least one biosignal detected from a body and digital signal processing of the analog signal processed biosignal, Recognizing and expressing the physical and emotional state is preferably made using the result of the processing.

In accordance with another aspect of the present invention, there is provided a method for recognizing an emotional state, the method including: extracting at least one biometric parameter representing a characteristic of the body from at least one biosignal detected from a body for a predetermined time; Determining the degree of change of the parameter, and using the determined degree of change to determine the current emotional state, the value of each of the biometric parameters is preferably changed according to the change of the emotion.

Hereinafter, the configuration and operation of the apparatus for recognizing a physical and emotional state according to the present invention and the method for recognizing a physical and emotional state performed by the apparatus will be described as follows.

1 is a schematic block diagram of an apparatus for recognizing a body and an emotional state according to the present invention, and includes a biosignal detection unit 100 and a biosignal recognition unit 120.

The biosignal detection unit 100 illustrated in FIG. 1 includes a sensor unit 10, an analog signal processing unit 20, an analog-to-digital converter (ADC) 30, and a wireless signal transmission unit 70. The biosignal recognition unit 120 includes a wireless signal receiver 80, a digital signal processor 40, a controller 50, and a data output unit 60.

FIG. 2 is a flowchart illustrating a method of recognizing a body and an emotional state according to an embodiment of the present invention performed by the apparatus shown in FIG. 1, wherein analog signals are processed by biosignal (s) detected from a body (200 to 200). Step 206) and digital signal processing of the analog signal processed biosignal (s) to recognize and express a body and an emotional state (steps 208 to 214).

The biosignal detection unit 100 is mounted on a predetermined portion of the body and processes at least one biosignal detected from the body by analog signal processing and outputs the analog signal to the biosignal recognition unit 120 (steps 200 to 206).

First, the sensor unit 10 of the biosignal detection unit 100 detects the biosignal from the body (step 200). For example, the sensor unit 10 may detect analog biological signals regarding blood pressure, skin temperature, and skin resistance. To this end, the sensor unit 10 includes a blood pressure sensor (not shown) for measuring blood pressure, a skin temperature sensor (not shown) for measuring skin temperature, and a skin resistance sensor (not shown) for measuring skin resistance. At this time, the sensor unit 10 may further provide a breathing sensor (not shown) for measuring the respiratory rate.

Here, the blood pressure sensor (not shown) may be preferably implemented as a pressure sensor such as a piezo sensor or a strain gauge or a photo-electric pulse plethysmograph (PPG) sensor. Here, the blood pressure sensor starts from page 163 of a book published by REGENTS / Prentice Hall, 1993, entitled “Introduction to Biomedical equipment Technology (2nd Edition)” by Jeseph J. Carr and John M. Brown. Disclosed on page 177. Piezo Sensor is a page of a book published by REGENTS / Prentice Hall, published in 1993 by Jeseph J. Carr and John M. Brown, entitled "Introduction to Biomedical equipment Technology (2nd Edition)." Page 168, and in 1995, published by Robert Bentley, entitled "Bosch Automotive Electric-Electronic Systems Handbook" by Robert Bosch. Strain Guage is also published on pages 113 to 122 of the book, published by Maxwell Macmillan International Edition, 1991, entitled "Principles of Biomedical instrumentation and Measurement" by Richard Aston. Photo-electric pulse plethysmoGraph (PPG) sensors were published in 1993 by Jeseph J. Carr and John M. Brown, published by REGENTS / Prentice Hall, entitled "Introduction to Biomedical equipment Technology (2nd Edition)." See pages 207 through 209 of this page.

The skin temperature sensor (not shown) may be implemented as a sensor for measuring skin temperature using an electrode and a thermo couple. Here, the skin temperature sensor is disclosed on pages 100 to 107 of a book published by Maxwell Macmillan International Edition, 1991, entitled "Principles of Biomedical instrumentation and Measurement" by Richard Aston. At this time, Electrode was a page of a book published by REGENTS / Prentice Hall, published in 1993 by Jeseph J. Carr and John M. Brown, entitled "Introduction to Biomedical equipment Technology (2nd Edition)." See pages 25 through 37. The Thermo Couple is published on page 100 through page 107 of the book, published by Maxwell Macmillan International Edition, in 1991, entitled "Principles of Biomedical instrumentation and Measurement" by Richard Aston.

The skin resistance sensor (not shown) may be implemented as a sensor for measuring skin resistance using an electrode that directly or indirectly contacts the skin and a comparator connected to the electrode. Here, the skin resistance sensor was described in '1981 Don C. Fowles, Robert Edelberg and David T. The pages of pages 232 to 239 of the journal Psychophysiology, published by Th society for psychophysiological Research, Inc, entitled "Publication Recommendation for Electrodermal Measurements" by Lykken.

After step 200, the analog signal processor 20 amplifies and filters the analog signal detected by the sensor unit 10 and outputs the ADC signal to the ADC 30 (step 202). In this case, the filtering mainly uses a low pass filter to remove high frequency noise components, but may use a band pass filter to obtain a signal of a specific band.

After step 202, the ADC 30 converts the amplified and filtered analog biosignal output from the analog signal processor 20 into a digital biosignal and transmits the wireless signal transmitter 70 or the digital signal processor 40. (Step 204).

According to an exemplary embodiment of the present invention, the biosignal detection unit 100 and the biosignal recognition unit 120 may exchange digital signals of the biosignal with a wire. In this case, the process proceeds to step 210 after step 204. To this end, the biological signal of the digital form is output to the digital signal processor 40 by wire.

According to another exemplary embodiment of the present invention, the biosignal extractor 100 and the biosignal recognition unit 120 may exchange digital signals with each other wirelessly. In this case, the biosignal detection unit 100 may be formed in a belt-type to be detachable or attached to a predetermined part of the body. For example, the biosignal detection unit 100 may be worn on a wrist as a watch type or may be implemented to be worn on other parts of the body according to a purpose of use. In addition, the biosignal recognition unit 120 may also be implemented to be worn on the body, and may be installed outside at a predetermined distance from the biosignal detection unit 100. As such, when the wireless signal transmitter 70 and the wireless signal receiver 80 are provided at a predetermined distance to transmit and receive digital signals to each other wirelessly, after step 204, the wireless signal transmitter 70 The digital signal provided from the ADC 30 is converted into a wireless signal and transmitted to the biosignal recognition unit 120 (operation 206).

Meanwhile, the biosignal recognition unit 120 performs digital signal processing on the biosignal processed by the biosignal detection unit 100, and recognizes and expresses a body and an emotional state from the digital signal processed results (steps 208 to 214). ).

If the biosignal detection unit 100 transmits a signal wirelessly, after step 206, the radio signal receiver 80 of the biosignal recognition unit 120 receives the radio signal transmitted from the radio signal transmitter 70. In operation 208, the biosignal having a digital form is received and output to the digital signal processor 40. However, as described above, when the biosignal detection unit 100 transmits the digital signal biometric signal to the biosignal recognition unit 120 by wire, after step 204, the biosignal detection unit 100 does not perform steps 206 and 208. Proceed.

The digital signal processor 40 processes the digital signal of the digital type input from the ADC 30 or the wireless signal receiver 80 and outputs the biosignal to the controller 50 and the data output unit 60 (step 210). .

After operation 210, the controller 50 recognizes a body and an emotional state from the data processed by the digital signal processor 40, and outputs the recognized result to the data output unit 60 as a control signal (operation 212). . To this end, the controller 50 stores at least one preset reference value through the input terminal IN, compares the stored reference value with data processed by the digital signal processor 40, and compares the body and the body with the compared result. Recognize emotional states. Here, the reference value corresponds to an average value of the biosignal that ordinary people have.

Hereinafter, a preferred embodiment of the present invention for the second step 212 performed by the control unit 50 shown in FIG. 2 will be described with reference to the accompanying drawings.

FIG. 3 is a flowchart for describing a preferred embodiment of the present invention with respect to step 212 illustrated in FIG. 2, and determining whether to use an average value or a reference value to recognize a body and emotion (240). To step 246) and to recognizing the physical and emotional state using the determined average value or reference value (steps 248 and 250).

Referring to FIG. 3, the controller 50 stores at least one preset reference value (operation 240). As described above, the controller 50 may input and store the reference value from the outside through the input terminal IN or may store the input value in advance. After operation 240, at least one average value of the bio signals detected by the sensor unit 10 may be calculated in advance from data processed by the digital signal processor 40 (operation 242).

After step 242, the average value or the reference value is selected in response to the difference between the calculated average value and the reference value, and the selected average value or reference value is compared with the data currently processed by the digital signal processor 40 to recognize the body and the emotional state ( 244 to 250 steps). That is, after step 242, a difference between the average value and the reference value is obtained (step 244), where the average value corresponds to an average value of the bio signals previously measured by the sensor unit 10 from the user of the apparatus shown in FIG. do. After step 244, it is determined whether the difference is greater than a predetermined allowable error (step 246). If it is determined that the difference is greater than the predetermined tolerance, the average value is compared with the data processed in step 210 to recognize the physical and emotional state (step 248). However, if it is determined that the difference is not greater than the predetermined tolerance, the reference value is compared with the data processed in step 210 to recognize the physical and emotional state (step 250).

In order to help the understanding of the present invention, the operation of the controller 50 will be described as follows, assuming that the blood pressure, skin temperature, and skin resistance are measured by the sensor unit 10 as described above.

The controller 50 stores the predetermined first, second and third reference values of the bio signals for blood pressure, skin temperature and skin resistance, respectively, and pre-detects the bio signals for blood pressure, skin temperature and skin resistance. Each of the first, second, and third average values is calculated and stored. At this time, if it is determined that the difference between the corresponding X (1 ≦ X ≦ 3) reference value and the X average value is not greater than the predetermined X tolerance, the controller 50 processes the X reference value and the digital signal processor 40. Compare the data and if it is determined that the difference between the X reference value and the X average value is greater than the predetermined X tolerance, the X average value is compared with the data processed by the digital signal processor 40 to recognize a physical or emotional state. . As such, the reason why the average value is compared with the processed data instead of the reference value is to accurately recognize the physical and emotional state in consideration of the user's body characteristics. For example, if the blood pressure or body temperature currently measured by the sensor unit 10 is higher than the reference value or the calculated average value by more than a predetermined level, the controller 50 recognizes as "excitement or anger," and the current measurement by the sensor unit 10. If the given skin resistance is lower by a predetermined level than the predetermined reference value or average value, it can be recognized as "tension" or the like. This is because sweat is secreted more than usual during tension, and when sweat is secreted, skin resistance drops. These recognized physical and emotional states can be added or changed in various ways through experiments. In addition, recognition data with higher reliability may be obtained by the combination of the recognized states. That is, the controller 50 may recognize each state of the body and the emotion according to each of the first, second and third comparison values comparing the average value or the reference value with the processed data. It is also possible to recognize the state of the body and emotion using the result of logical OR or AND of the three comparison values. The emotional state recognition method according to the present invention for recognizing the emotional state using the result of comparing the average value with the processed data will be described later in detail.

On the other hand, after step 212, the data output unit 60 voices the processed digital data, i.e., bio signals and recognition signals, input from the digital signal processing unit 40 in response to the control signal generated from the controller 50. And / or present the image to the user (step 214). To this end, the data output unit 60 may provide a speaker and / or a monitor, and simultaneously voice and / or image of body and emotional state recognition data and detected biosignal data in response to a control signal of the controller 50. To the user.

Hereinafter, a method of recognizing an emotional state according to the present invention and embodiments thereof will be described with reference to the accompanying drawings.

FIG. 4 is a flowchart illustrating a method for recognizing an emotional state according to the present invention, comprising: determining an emotional state using biometric parameter (s) extracted from at least one biosignal detected during a predetermined time (300; Step 302).

In order to perform the emotional state recognition method according to the present invention, first, at least one biometric parameter representing a characteristic of the body is extracted from at least one or more biosignals detected from the body for a predetermined time (step 300). For example, in step 300, biometric parameters representing the frequency of the electrocardiogram, the number of beats of the heart, and the electrocardiogram according to respiration are extracted from the biosignal having information on blood pressure, and the biosignal having information on skin temperature and skin resistance is extracted. Biometric parameters indicative of skin temperature and skin resistance can be extracted from the signals.

The value of each of the biometric parameters extracted in operation 300 changes according to a change in emotion, and a predetermined time is a minimum time required for the changed emotion to be expressed as a biosignal when the emotion is changed (hereinafter, referred to as a short predetermined time). Or at least longer than the minimum time (hereinafter, referred to as a 'long predetermined time'). For example, the short predetermined time may be any value within 10 seconds to 60 seconds, and the long predetermined time may be any time larger than 60 seconds.

In the emotional state recognition method according to the present invention, the human emotional state is recognized using biometric parameters detected at a short predetermined time and a long predetermined time, respectively. That is, the emotion state recognition method according to the present invention uses biometric parameters detected at a short predetermined time to quickly recognize a rapidly changing emotion state, and detects at a long predetermined time to recognize an emotional state remaining after the change. The generated biometric parameters.

On the other hand, after step 300, the degree of change of the currently extracted biometric parameters is determined, and the current emotional state is determined using the determined degree of change (step 302). To this end, the degree of change in the currently extracted biometric parameters is determined by comparing the basic biometric parameters with the currently extracted biometric parameters. Here, the basic bioparameters mean bioparameters measured during a baseline for which no emotion occurs, for example, for about 10-20 minutes with a comfortable mind.

The emotional state recognition method according to the present invention illustrated in FIG. 4 corresponds to step 248 illustrated in FIG. 3 and may be performed by the controller 50. In this case, the basic biometric parameter corresponds to the mean value.

Hereinafter, embodiments of the present invention for the step 302 of recognizing an emotional state using biometric parameters detected at short time and long time, respectively, will be described with reference to the accompanying drawings.

FIG. 5 is a flowchart for describing a preferred embodiment 302A according to the present invention with respect to step 302 shown in FIG. 4, and recognizing a human emotional state according to a degree of change of biometric parameters (310). To step 318).

According to the embodiment shown in FIG. 5, an emotional state is recognized using biometric parameters detected for a short predetermined time. First, it is recognized that the emotional state is joy using biometric parameters for the number of beats of the heart, the skin temperature, and the frequency of the electrocardiogram (step 310). In addition, the biometric parameters for the number of beats of the heart and the frequency of the electrocardiogram are used to recognize that the emotional state is stress (step 312). In addition, using the biometric parameters for the heart rate and skin temperature of the heart to recognize that the emotional state is sad (step 314). In addition, biometric parameters for skin temperature and skin resistance are used to recognize that the emotional state is free (step 316). In operation 318, the emotional state may be recognized using biometric parameters indicating skin resistance and frequency of ECG. Here, steps 310 to 318 illustrated in FIG. 5 may be performed in any order.

FIG. 6 is a flowchart for explaining another preferred embodiment 302B according to the present invention with respect to step 302 shown in FIG. 4, and recognizing a human emotional state according to a degree of change of biometric parameters (330). To step 338).

6 recognizes an emotional state using biometric parameters detected for a long predetermined time. First, it is recognized that the emotional state is free using the biometric parameters for the skin temperature, the skin resistance, the electrocardiogram frequency, and the electrocardiogram according to the breathing (step 330). Recognizing that the emotional state is anger using biometric parameters for skin resistance and ECG according to breathing (step 332). In operation 334, the emotional state is recognized using biometric parameters such as the number of beats of the heart, the frequency of the electrocardiogram, and the respiratory rate. Recognizing that the emotional state is sad using the biometric parameters for the number of heart beats and skin resistance (step 336). The biometric parameters for skin resistance are used to recognize that the emotional state is stress (step 338). Here, steps 330 to 338 illustrated in FIG. 6 may be performed in any order.

For example, the biometric parameters mentioned in the embodiments 302A and 302B shown in FIG. 5 or 6 may be defined as follows.

First, the biometric parameter with respect to the frequency of the electrocardiogram is a biometric parameter (LF) representing a value obtained by integrating a power spectrum belonging to a low frequency range from a fast Fourier transform (FFT) of a biosignal detected for blood pressure. Low Frequency (FFT), a biometric parameter (HF: High Frequency) or LF that represents the integrated value of the power spectrum in the high frequency range from the result of the Fast Fourier Transform (FFT) It may be a biometric parameter (LF / HF) representing a value divided by. Here, the low frequency range may be 0.04 to 0.15 Hz, and the high frequency range may be 0.15 to 0.5 Hz.

In addition, the biometric parameter for the number of heart beats is a biometric parameter RR representing an average value of intervals between positive peaks of the body's blood pressure, or a biometric parameter HR indicating a number of heartbeats during a unit time. (Heart Rate). Here, the positive peak usually means the R peak and is disclosed in the physiological texts related to the electrocardiogram. Positive Peak, for example, was published in 1996 in the title of "Medical Instrumentation Design" by Joseph J Carr & John M Brown, translated by Lee Myung-ho, and published by a publisher called Kyung Moonsa. See pages 171 through 216. Here, the biometric parameter HR can be obtained by calculating, for example, how much the heart beats for 1 minute by inverting the biometric parameter RR.

In addition, the biometric parameter for ECG according to breathing is a value obtained by subtracting the result of subtracting the minimum interval from the maximum interval among the positive peaks for blood pressure by the minimum interval [(maximum interval-minimum interval) / minimum Interval] may be a biological parameter (RSA: Respiratory Sinus Arrhythmia). Here, various methods of extracting biometric parameters (RSA) are generally well known. For example, in 1994, A.W. Co-published in the journal Computer in Cardiology, published by Frey, C. Hagenmiller, J Baumert, F Grueneis, M Dambacher, K Theisen, and M Adt 'entitled "The Respiratory Sinus Arrhythmia as a Function of Breathing Frequency Revisited" Pages 41 to 44 of the published IEEE paper.

In addition, the biometric parameter with respect to the skin temperature may be a biometric parameter (SKT: SKT) representing an average skin temperature. The biometric parameter for skin resistance refers to a skin parameter (SCL: Skin Conductive Level) representing the average of the skin resistance reciprocal value, and positive zero crossing (measured in one graph in which the skin resistance value is converted to the reciprocal). It may be a biometric parameter (N-SCR: Number of Skin Conductive Response) or a biometric parameter (SCRM: Skin Conductive Response Magnitude) indicating a height from positive zero crossing to negative zero crossing. Here, positive zero crossing means that the rate of change is changed from + to-, and negative zero crossing means that the rate of change of the curve represented in the graph is changed from-to +. At this time, since positive or negative zero crossing may occur due to noise, it is also possible to low pass filter the curve displayed on the graph in advance.

In addition, biometric parameters [(FPV: Finger Pulse Volumn) and PPG] may be extracted in operation 300. Here, FPV and PPG both represent the amplitude of blood flow.

Each of the aforementioned biometric parameters RR, HR, SKT, SCL, N-SCR, SCRM, LF, HF, LF / HF, RSA, RESP, FPV and PPG are obtained at different times. At this time, the biometric parameters are all detected for a short predetermined time when applied to the embodiment 302A shown in FIG. 5, and all are detected for a long predetermined time when applied to the embodiment 302B shown in FIG. 6. Here, the biometric parameters may be generated in the controller 50 shown in FIG. 1 as follows.

The biosignal for the blood pressure sensed by the blood pressure sensor (not shown) provided in the sensor unit 10 shown in FIG. 1 is processed as described above and input to the digital signal processor 40. In this case, the controller 50 may extract the biometric parameters RR, HR, LF, HF, LF / HF, RSA, and RESP using data on blood pressure processed and input by the digital signal processor 40. have. For example, the controller 50 obtains the biometric parameters RR and RSA from the data on the squeezing pressure, calculates the biometric parameter HR by inverting the biometric parameter RR, and analyzes the processed data by FFT to LF and HF. And the biometric parameters (LF / HF) using LF and HF. Here, when obtaining the biometric parameter RESP, the biosignal for the breath sensed by the breathing sensor (not shown) may be used instead of the biosignal for the blood pressure sensed by the blood pressure sensor (not shown). In this case, the biosignal for the breath sensed by the breathing sensor (not shown) provided in the sensor unit 10 shown in FIG. 1 is input to the digital signal processor 40 through the above-described process. At this time, the controller 50 extracts the biometric parameter RESP using data on respiration processed by the digital signal processor 40. Many common ways of extracting RESPs are known to be null. For example, a common method of extracting RESPs was published in April 1997 by Pei Z Zhang, Walter N Tapp, Stanley S Reisman, and Benjamin H Natelson, entitled "Respiration Response Curve Analysis of Heart Rate Variability." Pages 321 to 325 of vol 44 no 4 of the journal IEEE Transaction on Biomedical Engineering, published and published in the journal, are published.

In addition, the biosignals for the skin temperature and skin resistance sensed by the skin temperature sensor (not shown) and the skin resistance sensor (not shown) provided in the sensor unit 10 shown in FIG. 1 are processed as described above. And input to the digital signal processing unit 40. At this time, the controller 50 may extract the biometric parameters (SKT, SCL, N-SCR and SCRM) using data on the skin temperature and skin resistance processed by the digital signal processor 40.

In addition, the blood pressure sensor (not shown) included in the sensor unit 10 illustrated in FIG. 1 to measure the biometric parameters FPV and PPG may be implemented as a PPG sensor (not shown). A PPG sensor (not shown), for example, is provided at the fingertips to sense information about blood flow. At this time, the biosignal for the blood flow having the sensed information is processed as described above and input to the digital signal processor 40. The controller 50 extracts biometric parameters FPV and PPG using data on the blood flow processed by the digital signal processor 40.

As described above, when the biometric parameters are extracted from the controller 50, the controller 50 determines the emotion state by determining the degree of change of the biometric parameters currently extracted based on the previously extracted basic bioparameters. That is, the steps 302 shown in FIG. 4, the embodiments 302A and 302B shown in FIGS. 5 and 6, respectively, may be performed by the controller 50.

Hereinafter, preferred embodiments of the present invention for the embodiments 302A and 302B shown in FIGS. 5 and 6 will be described with reference to the accompanying drawings. At this time, the basic biometric parameters are RR (base), HR (base), SKT (base), SCL (base), N-SCR (base), SCRM (base), LF (base), HF (base), LF / HF (base), RSA (base) and RESP (base).

FIG. 7 is a flowchart for explaining a preferred embodiment according to the present invention with respect to the embodiment 302A shown in FIG. 5, in which pleasure is determined (steps 350 to 360), and stress is determined. (Steps 362 and 364), determining sadness (steps 366 and 368), determining freedom (steps 370 to 374), determining anger (steps 376 to 376). Steps 382) and determining that there is no special emotional state (step 384).

In the embodiment shown in FIG. 7, the emotional state is determined using the biometric parameters (RR, HR, SKT, SCL, N-SCR, SCRM, LF and LF / HF) detected in a short time.

First, in the initial state, it is determined whether the emotional state is joy as follows (350 to 360 steps). Here, the initial state is a state in which no emotion predominantly exists, for example, a state in which emotions (or mental states) of joy, sadness, stress, anger, and boredom are uniform or mixed.

After operation 300, it is determined whether the HR value in the initial state is reduced by more than the HR threshold value HR_th (operation 350). To this end, it is determined whether the value obtained by subtracting HR_th from HR (base) is greater than the HR value. If it is determined that the HR value is reduced by more than the HR threshold value HR_th, it is determined whether the SKT value has been increased by more than the SKT threshold value SKT_th (step 352). Here, the HR value reduced by more than the HR threshold HR_th means that the probability that the emotional state is joy, sadness or stress is 50%, for example. However, since it is still difficult to determine which emotional state among joy, sadness or stress, it is determined whether the SKT value has increased by more than the SKT threshold value SKT_th. To this end, it is determined whether the value obtained by adding SKT (base) and SKT_th is smaller than the SKT value.

At this time, if it is determined that the SKT value is increased by more than the SKT threshold value (SKT_th), the LF value is reduced by more than the LF threshold value (LF_th) or the LF / HF value is reduced by more than the LF / HF threshold value (LF / HF_th) or more. It is determined whether or not (step 354). Here, the increase in the SKT value by more than the SKT threshold value SKT_th means that the probability that the emotional state is joy or sadness is 60%, for example. However, since it is still difficult to determine which emotional state of joy or sadness, whether the LF value is reduced by more than the LF threshold (LF_th) or the LF / HF value is reduced by more than the LF / HF threshold (LF / HF_th). To judge. To this end, it is determined whether the value of subtracting LF_th from LF (base) is greater than LF or the value of subtracting LF / HF_th from LF / HF (base) is greater than LF / HF.

If it is determined that the LF value is reduced by more than the LF threshold LF_th or the LF / HF value is reduced by more than the LF / HF threshold LF / HF_th, the probability of the emotional state as joy is determined as the first predetermined percentage. (Step 356). Here, the first predetermined percentage may be set to 70%. After step 356, it is determined whether the RR value has increased by more than the RR threshold value RR_th (step 358). To this end, it is determined whether the value obtained by adding RR (base) and RR_th is smaller than the RR value. If it is determined that the RR value is not increased by more than the RR threshold value RR_th, the flow proceeds to step 356. However, if it is determined that the RR value has increased by more than the RR threshold RR_th, the probability that the emotional state is joy is determined as the second predetermined percentage increased from the first predetermined percentage. Here, the second predetermined percentage may be set to 80%.

Next, it is determined whether the emotional state is stressed (steps 362 and 364).

If it is determined in step 352 that the SKT value has not increased by more than the SKT threshold value SKT_th, it is determined whether the LF value has decreased by more than the LF threshold value LF_th (step 362). Here, the fact that the SKT value has not been increased by more than the SKT threshold value SKT_th means that the probability that the emotional state is stress is still 50%, for example. Therefore, since it is difficult to determine that the emotional state is stressed yet, it is determined whether the LF value is reduced by more than the LF threshold LF_th. To this end, it is determined whether the value obtained by subtracting LF_th from LF (base) is greater than the LF value. If it is determined that the LF value is reduced by more than the LF threshold LF_th, the probability that the emotional state is stress is determined as the second predetermined percentage (step 364).

Next, it is determined whether the emotional state is sad (steps 366 and 368).

If it is determined in step 354 that the LF value is not reduced by more than the LF threshold LF_th and the LF / HF value is not reduced by more than the LF / HF threshold LF / HF_th, the RR value is determined by the RR threshold value ( RR_th) or more (Step 366). Here, the fact that the LF value has not been reduced by more than the LF threshold LF_th and the LF / HF value has not been reduced by more than the LF / HF threshold LF / HF_th has a probability that the emotional state is sad, for example. It means 60%. Therefore, since it is difficult to determine yet that the probability that the emotional state is sad is 60%, it is determined whether the RR value has increased by the RR threshold value RR_th or more. To this end, it is determined whether the value obtained by adding RR (base) and RR_th is smaller than the RR value. If it is determined that the RR value is increased by more than the RR threshold value RR_th, the probability that the emotional state is sad is determined as the second predetermined percentage (step 368).

Next, it is determined whether the emotional state is free (steps 370 to 374).

If it is determined in step 350 that the HR value is not reduced by more than the HR threshold value HR_th, the N-SCR value is increased by more than the N-SCR threshold value N-SCR_th and the SCRM value is greater than or equal to the SCRM threshold value SCRM_th. It is determined whether the increase by (step 370). Here, that the HR value is not reduced by more than the HR threshold value HR_th means that the emotional state is still an initial state. Therefore, it is determined whether the RR value has increased by more than the RR threshold value RR_th. To this end, it is determined whether the value obtained by adding N-SCR (base) and N-SCR_th is smaller than the value of N-SCR and the value obtained by adding SCRM (base) and SCRM_th is smaller than the SCRM value.

If the N-SCR value is increased by more than the N-SCR threshold value (N-SCR_th) and the SCRM value is increased by more than the SCRM threshold value (SCRM_th), is the SKT value increased by more than the SKT threshold value (SKT_th)? In operation 372, the process is determined. Here, the fact that the N-SCR value is increased by the N-SCR threshold value (N-SCR_th) or more and the SCRM value is increased by the SCRM threshold value (SCRM_th) or more means that the probability that the emotional state is angry or free is, for example, 50%. It means. Therefore, it is difficult to determine which emotional state in anger or indulgence, so it is determined whether the SKT value has increased by more than the SKT threshold value SKT_th. To this end, it is determined whether the value obtained by adding SKT (base) and SKT_th is smaller than the SKT value. If it is determined that the SKT value has been increased by more than the SKT threshold value SKT_th, the probability that the emotional state is free is determined as the second predetermined percentage (step 374).

Finally, determine if the emotional state is anger as follows (steps 376-382).

If it is determined in step 372 that the SKT value has not increased by more than the SKT threshold value SKT_th, it is determined whether the LF / HF value has decreased by more than the LF / HF threshold value LF / HF_th (step 376). Here, the fact that the SKT value is not increased by more than the SKT threshold value SKT_th means that the probability that the emotional state is anger is still 50%, for example. Therefore, since it is difficult to determine yet that the probability that the emotional state is anger is 60%, it is determined whether the LF / HF value is reduced by more than the LF / HF threshold value (LF / HF_th). To this end, it is determined whether the value obtained by subtracting the LF / HF_th from the LF / HF (base) is greater than the LF / HF value.

If it is determined that the LF / HF value is not reduced by more than the LF / HF threshold value (LF / HF_th), the flow proceeds to step 372. However, if it is determined that the LF / HF value is reduced by more than the LF / HF threshold value (LF / HF_th), the probability that the emotional state is anger is determined as the first predetermined percentage (step 378). After step 378, it is determined whether the SCL value has increased by more than the SCL threshold value SCL_th (step 380). If it is determined that the SCL value is not increased by more than the SCL threshold value SCL_th, the flow proceeds to step 378. However, if it is determined that the SCL value is increased by the SCL threshold value SCL_th or more, the probability that the emotional state is anger is determined as the second predetermined percentage (step 382).

In the embodiment illustrated in FIG. 7, it is determined that the LF value is not reduced by more than the LF threshold value LF_th in operation 362, or the N-SCR value is the N-SCR threshold value in operation 370. If it is determined that it has not been increased by the abnormality or the SCRM value has not been increased by the SCRM threshold value (N-SCR_th) or more, or it is determined in step 366 that the RR value has not been increased by the RR threshold value (RR_th), the dominant emotional state It can be determined that is not present (step 384). In other words, it is difficult to see that a particular emotional state prevails.

FIG. 8 is a flowchart for explaining a preferred embodiment of the present invention with respect to the embodiment 302B shown in FIG. 5, which includes determining the freeness (steps 400 to 414) and determining anger. (Steps 416 and 418), determining joy (steps 420 to 432), determining sadness (steps 434 and 436), and determining stress (step 438 And step 440) and determining that there is no special emotional state (step 442).

In the embodiment shown in FIG. 8, the emotional state is determined using the biometric parameters (HR, SKT, SCL, N-SCR, SCRM, LF, HF, LF / HF, RSA and RESP) detected at a long time. .

First, it is determined in the initial state whether the emotional state is free (steps 400 to 414).

It is determined whether the RSA value is as small as the RSA threshold RSA_th or more, the SCL value is as large as the SCL threshold SCL_th or more, and the SCRM value is as large as the SCRM threshold SCRM_th or more (step 400). To this end, is the value obtained by subtracting RSA_th from RSA (base) greater than the RSA value, the value of adding SCL (base) and SCL_th less than the SCL value, and the value of adding SCRM (base) and SCRM_th less than the SCRM value? Judge. If the RSA value is smaller than the RSA threshold RSA_th or more, the SCL value is larger than the SCL threshold SCL_th, and the SCRM value is larger than the SCRM threshold SCRM_th, the N-SCR value is N-. It is determined whether the SCR is greater than the N-SCR_th (step 402). To this end, it is determined whether the value obtained by adding N-SCR (base) and N-SCR_th is smaller than the N-SCR value. If it is determined that the N-SCR value is greater than or equal to the N-SCR threshold value (N-SCR_th), it is determined whether the SKT value is greater than or equal to the SKT threshold value (SKT_th) (step 404). To this end, it is determined whether the value obtained by adding SKT (base) and SKT_th is smaller than the SKT value. If it is determined that the SKT value is greater than or equal to the SKT threshold value SKT_th, the probability that the emotion state is free is determined as the third predetermined percentage (step 406). Here, the third predetermined percentage may be set to 60%.

After the step 406, it is determined whether the HF value is as small as the HF threshold value HF_th or more (step 408). To this end, it is determined whether the value obtained by subtracting HF_th from HF (base) is greater than the HF value. If it is determined that the HF value is not smaller than the HF threshold value HF_th, the flow proceeds to step 406. However, if it is determined that the HF value is smaller than or equal to the HF threshold value HF_th, the probability that the emotional state is free is determined as the fourth predetermined percentage increased from the third predetermined percentage (step 410). Here, the fourth predetermined percentage may be set to 70%.

After operation 410, it is determined whether the LF value is larger than the LF threshold LF_th (operation 412). To this end, it is determined whether the value obtained by subtracting LF_th from LF (base) is greater than the LF value. If it is determined that the LF value is not larger than the LF threshold LF_th, the flow proceeds to step 410. However, if it is determined that the LF value is greater than or equal to the LF threshold LF_th, the probability that the emotional state is free is determined as the fifth predetermined percentage increased from the fourth predetermined percentage (step 414). Here, the fifth predetermined percentage may be set to 80%.

Next, determine whether the emotional state is anger as follows (steps 416 and 418).

If it is determined in step 402 that the N-SCR value is not greater than the N-SCR threshold value (N-SCR_th), the probability that the emotional state is anger is determined as the fifth predetermined percentage (step 416). In addition, if it is determined in step 404 that the SKT value is not greater than the SKT threshold value SKT_th, the probability that the emotional state is anger is determined as a third predetermined percentage (step 418).

Next, it is determined whether the emotional state is joy (steps 420 to 432).

In step 400, if it is determined that the RSA value is not smaller than the RSA threshold RSA_th or more, the SCL value is not larger than the SCL threshold SCL_th, or the SCRM value is not larger than the SCRM threshold SCRM_th, the HR value. It is determined whether the amount is smaller than the HR threshold HR_th or more (step 420). To this end, it is determined whether the value obtained by subtracting HR_th from HR (base) is greater than the HR value. If it is determined that the HR value is smaller than or equal to the HR threshold value HR_th, it is determined whether the RESP value is larger than or equal to the RESP threshold value RESP_th (step 422). To this end, it is determined whether the value obtained by adding RESP (base) and RESP_th is smaller than the RESP value. If it is determined that the RESP value is greater than or equal to the RESP threshold value RESP_th, the probability that the emotional state is joy is determined as the third predetermined percentage (step 424).

After the step 424, it is determined whether the LF / HF value is smaller than the LF / HF threshold value (LF / HF_th) or more (step 426). To this end, it is determined whether the value obtained by subtracting the LF / HF_th from the LF / HF (base) is greater than the LF / HF value. If it is determined that the LF / HF value is not smaller than the LF / HF threshold value (LF / HF_th) or more, the flow proceeds to step 424. However, if it is determined that the LF / HF value is less than or equal to the LF / HF threshold value (LF / HF_th), the probability that the emotional state is joy is determined as the fourth predetermined percentage (step 428).

After operation 428, it is determined whether the LF value is smaller than the LF threshold LF_th or more (operation 430). To this end, it is determined whether the value obtained by subtracting LF_th from LF (base) is greater than the LF value. If it is determined that the LF value is not smaller than the LF threshold LF_th, the flow proceeds to step 428. However, if it is determined that the LF value is smaller than or equal to the LF threshold LF_th, the probability that the emotional state is joy is determined as the fifth predetermined percentage (step 432).

Next, it is determined whether the emotional state is sad (steps 434 and 436).

If it is determined in step 422 that the RESP value is not greater than the RESP threshold value (RESP_th), it is determined whether the SCL value is smaller than the SCL threshold value (SCL_th) or more (step 434). To this end, it is determined whether the value obtained by subtracting SCL_th from the SCL (base) is greater than the SCL value. If it is determined that the SCL value is smaller than the SCL threshold value SCL_th, the probability that the emotional state is sad is determined as the fifth predetermined percentage (step 436).

Finally, it is determined whether the emotional state is stressed (steps 438 and 440).

If it is determined in step 420 that the HR value is not smaller than the HR threshold value HR_th, it is determined whether the N-SCR value is larger than the N-SCR threshold value N-SCR_th (step 438). To this end, it is determined whether the value obtained by adding N-SCR (base) and N-SCR_th is smaller than the N-SCR value. If it is determined that the N-SCR value is greater than or equal to the N-SCR threshold value (N-SCR_th), the probability that the emotional state is stress is determined as a fourth predetermined percentage (step 440).

8, if it is determined in step 434 that the SCL value is not smaller than the SCL threshold SCL_th, or in step 438, it is determined that the N-SCR value is not larger than the N-SCR threshold, It may be determined that there is no dominant emotional state (step 442).

In the embodiments for the step 302 illustrated in FIG. 7 or FIG. 8, when the probability that there is one or more emotional states is a predetermined percentage of X (1 ≦ X ≦ 5), the probability that the remaining emotional states exist is 100. It is the result of subtracting the X-th predetermined percentage from the percentage divided by the number of remaining emotional states. For example, if there is a 70% chance that an emotional state is anger, then there is a 7.5% chance that the current emotional state is joy, sadness, boredom or stress.

According to an embodiment of the present invention, the biometric parameters (HR, RESP, FPV, SKT, SCL, N-SCR, SCRM, HF, LF, LF / HF and RR) detected during a short predetermined time period are used for sex. Regardless of 20 children, autonomous nervous system (ANS) responses in five emotional states such as anger, boredom, joy, sadness and stress can be seen in Table 1 below. .

division HR RESP FPV SKT SCL N-SCR SCRM HF LF LF / HF RR anger ↓? ↑ ↑ ↑ ↓ Free ↑ ↑ ↑ ↓? pleasure ↓ ↑? ↑ ↓ ↓ ↑ sadness ↓ ↑? ↑ ↑ stress ↓ ↑? ↓

Where ↑ indicates that the currently extracted biometric parameter is larger than the default biometric parameter value, ↓ indicates that it is decreasing than the default biometric parameter value, and? Indicates that the value of the currently extracted biometric parameter is greater than the value of the default biometric parameter. It does not increase or decrease to the extent, respectively. At this time, a blank means that a slight change has occurred. Thus, biometric parameters that did not change significantly or changed slightly were not used in determining the emotional state in the embodiment shown in FIG. 7.

According to another embodiment of the present invention, the biometric parameters (HR, RESP, RSA, PPG, SKT, SCL, N-SCR, SCRM, HF, LF and LF / HF) detected during a long predetermined time period are used for sex. Regardless of autonomous nervous system responses in five emotional states, such as anger, boredom, joy, sadness, and stress, about 20 children can be seen in Table 2 below.

HR RESP RSA PPG SKT SCL N-SCR SCRM HF LF LF / HF anger L L? H H H Free L H H H L L pleasure L H H? L L sadness L L? L stress H

Here, if the value of the currently extracted bio-parameter is greater than the value of the default bio-parameter, it is marked as H, and if it is small, it is indicated as L, and? Is so large or serious that the value of the currently extracted bio-parameter is more serious than the value of the basic bio-parameter. Each indicates that it is not small enough. At this time, a blank means that a slight change has occurred. Thus, biometric parameters that did not change severely or weakly were not used in determining the emotional state in the embodiment shown in FIG. 8.

As can be seen from Table 1 or Table 2, there are biometric parameters that show a special change in each emotional state, and some biometric parameters that do not show a special change. Therefore, the emotional state recognition method according to the present invention can determine the emotional state as described above using a special change in the biometric parameter.

On the other hand, in the emotional state recognition method according to the present invention described above, the emotional state that can be recognized is limited to five kinds of anger, boredom, joy, sadness and stress. However, the emotional state recognition method according to the present invention is not limited thereto, and biometric parameters (HR, RESP, RSA, PPG, SKT, SCL, N-SCR, SCRM, HF, LF, LF / HF, FPV and RR) You can also recognize various emotional states.

As described above, the apparatus and method for recognizing the physical and emotional state according to the present invention can be mounted on a predetermined part of the body without causing inconvenience to the user's movement, and can easily detect the biological signal to be transmitted by wireless or wired. The user may provide a recognized body and emotional state to the user by using the detected biosignal, and have an effect of real-time recognition of a rapidly changing emotional state or a long-lasting emotional state.

Claims (32)

A biosignal detection unit mounted on a predetermined portion of the body to output an analog signal by processing at least one biosignal detected from the body; And And a biosignal recognition unit configured to digitally process the biosignal signal processed by the biosignal detection unit and to recognize and express the body and emotional state from the signal processed result. According to claim 1, The biosignal detection unit A sensor unit which detects the biological signal from the body in an analog form; An analog signal processor for amplifying and filtering the biosignal in analog form detected by the sensor unit; And An analog / digital converter configured to convert the amplified and filtered analogue biosignal into a digital form; The biosignal recognition unit, A digital signal processor configured to perform data processing on the digital signal; A controller for recognizing the body and the emotional state from the data processed by the digital signal processor, and outputting the recognized result as a control signal; And And a data output unit configured to output the data processed by the digital signal processor in a predetermined form in response to the control signal. The method of claim 2, wherein the sensor unit A blood pressure sensor for measuring blood pressure; A skin temperature sensor for measuring skin temperature; And Body and emotional state recognition device comprising a skin resistance sensor for measuring the skin resistance. The apparatus of claim 3, wherein the blood pressure sensor is a pressure sensor such as a piezo sensor or a strain gauge. The apparatus of claim 3, wherein the skin temperature sensor measures skin temperature using an electrode and a thermo couple. The apparatus of claim 3, wherein the skin resistance sensor measures skin resistance using an electrode and a comparator. The biosignal detection unit of claim 2, wherein the biosignal detection unit further comprises a wireless signal transmitter for converting the digital signal output from the analog / digital converter into a wireless signal and transmitting the converted biosignal. And a wireless signal receiver configured to receive a digital signal and output the digital signal to the digital signal processor. The apparatus of claim 1, wherein the biosignal detection unit may be attached to or detached from a predetermined portion of the body in a belt-type. 9. The method of claim 2, wherein the control unit And storing at least one reference value, comparing the stored reference value with data processed by the digital signal processor, and recognizing the body and the emotional state from the compared result. The method of claim 9, wherein the control unit At least one average value of the bio signals detected by the sensor unit is calculated from data preprocessed by the digital signal processor, and the average value or the reference value is selected in response to the difference between the average value and the stored at least one reference value. And recognize the body and the emotional state by comparing the selected average value or the reference value with data currently processed by the digital signal processor. (a) analog signal processing at least one or more biological signals detected from the body; And and (b) digital signal processing the analog signal processed biosignal, and recognizing and expressing the physical and emotional states by using the digital signal processed result. The method of claim 11, wherein In step (a), (a1) detecting the biosignal from the body; (a2) amplifying and filtering the biological signal in the detected analog form; And (a3) converting the amplified and filtered analogue biosignal into digital form; In step (b), (b1) data processing the digital signal; (b2) recognizing the physical and emotional state from the processed data; And (b3) providing the processed data to a user corresponding to the recognition result. The method of claim 12, wherein the step (a) further comprises the step of converting the digital signal into a radio signal after the step (a3) and transmitting the radio signal. Body and emotional state recognition method further comprising the step of proceeding to (b1). The method of claim 12, wherein step (b2) Storing at least one reference value; And And comparing the stored reference value with the data processed in the step (b1) and recognizing the body and the emotional state using the compared result. The method of claim 12, wherein step (b2) Storing at least one reference value; Calculating at least one average value of the bio signals detected in the step (a1) from data previously processed in the step (b1); Obtaining a difference between the average value and the reference value; Determining whether the difference is greater than a predetermined tolerance; If it is determined that the difference is greater than the predetermined tolerance, recognizing the body and the emotional state by comparing the average value with the data processed in step (b1); And If it is determined that the difference is not greater than the predetermined tolerance, comparing the reference value with the data processed in the step (b1) and recognizing the body and the emotional state. . (c) extracting at least one biological parameter indicative of the characteristic of the body from at least one or more biological signals detected from the body for a predetermined time; And (d) determining a degree of change of the extracted biological parameter, and determining a current emotional state using the determined degree of change, And a value of each of the biometric parameters changes according to the change of the emotion. The method of claim 16, wherein step (c) Extracting the biometric parameters for the electrocardiogram according to the frequency of the electrocardiogram, the number of beats of the heart and the breathing from the biosignal having the information on the blood pressure, And extracting the biometric parameters for skin temperature and skin resistance from the biosignals having information on skin temperature and skin resistance. The method of claim 17, wherein step (d) (d11) recognizing that the emotional state is joy using the biometric parameters for the number of beats of the heart, the skin temperature and the frequency of the electrocardiogram; (d12) recognizing that the emotional state is stress using the biometric parameters for the number of beats of the heart and the frequency of the electrocardiogram; (d13) recognizing that the emotional state is sad using the biometric parameters for the number of beats of the heart and the skin temperature; (d14) recognizing that the emotional state is free using the biometric parameters for the skin temperature and the skin resistance; And (d15) recognizing that the emotional state is anger using the biometric parameters for the skin resistance and the frequency of the electrocardiogram, And the predetermined time corresponds to a minimum time required for the changed emotion to be expressed by the bio signal when the emotion is changed. The method of claim 17, wherein step (d) (d21) recognizing that the emotional state is free using the biometric parameters for the skin temperature, the skin resistance, the frequency of the electrocardiogram and the electrocardiogram according to the breathing; (d22) recognizing that the emotional state is anger using the biometric parameters for the skin resistance and the electrocardiogram according to the breathing; (d23) recognizing that the emotional state is joy using the biometric parameters for the number of beats of the heart, the frequency of the electrocardiogram and the respiratory rate; (d24) recognizing that the emotional state is sad using the biometric parameters for the number of beats of the heart and the skin resistance; And (d25) using the biometric parameters for the skin resistance to recognize that the emotional state is stress; And wherein the predetermined time is at least greater than a minimum time required for the changed emotion to be represented by the bio signal when the emotion is changed. The method of claim 18 or 19, The biometric parameters for the frequency of the electrocardiogram are biometric parameters (LF and HF) representing a value obtained by integrating a power spectrum belonging to a low and high frequency range from a fast Fourier transform of the biosignal detected for the blood pressure. Corresponds to the biometric parameter (LF / HF) representing the value obtained by dividing the LF by the HF, The biometric parameters for the number of beats of the heart are determined by a biometric parameter (RR) representing an average value of intervals between positive peaks of the body's blood pressure and a biometric parameter (HR) representing a number of times the heart beats during the unit time. Corresponding, The biometric parameter for the ECG according to the breath corresponds to a biometric parameter (RSA) representing a value obtained by subtracting the minimum interval from the maximum interval among the intervals between the positive peaks for the blood pressure by the minimum interval. and, The biometric parameter for the skin temperature corresponds to a biometric parameter (SKT) representing an average skin temperature value, The biometric parameter for the skin resistance is a biometric parameter (SCL) representing an average of skin reciprocal resistance values, and a biometric parameter (N-SCR) representing a number of positive zero crossings in a graph representing the skin resistance reciprocal value and the Emotional state recognition method, characterized in that it corresponds to a biometric parameter (SCRM) indicating the height from positive zero crossing to negative zero crossing. The method of claim 20, wherein the step (d11) when the predetermined time is the minimum time, (d111) determining whether the HR value is decreased by an HR threshold value or more; (d112) if it is determined that the HR value is reduced by more than the HR threshold value, determining whether the SKT value has been increased by more than the SKT threshold value; (d113) if it is determined that the SKT value is increased by more than the SKT threshold value, determining whether the LF value is reduced by more than the LF threshold value or the LF / HF value is reduced by more than the LF / HF threshold value; (d114) if it is determined that the LF value has been reduced by more than the LF threshold or the LF / HF value has been reduced by more than the LF / HF threshold, determining a probability of the emotional state being joy by a first predetermined percentage; (d115) determining whether the RR value is increased by more than the RR threshold value, and if it is determined that the RR value is not increased by more than the RR threshold value, proceeding to step (d114); And (d116) if it is determined that the RR value is increased by more than the RR threshold, determining a probability that the emotional state is joy as a second predetermined percentage increased from the first predetermined percentage; State awareness method. The method of claim 21, wherein step (d12) (d121) if it is determined in step (d112) that the SKT value has not increased by more than the SKT threshold value, determining whether the LF value has decreased by more than the LF threshold value; And and (d122) if it is determined that the LF value is reduced by more than the LF threshold, determining the probability of the emotional state as the second predetermined percentage. The method of claim 22, wherein step (d13) (d131) If it is determined in step (d113) that the LF value is not reduced by more than the LF threshold value and the LF / HF value is not reduced by more than the LF / HF threshold value, the RR value is the RR Determining whether it has increased by a threshold value or more; And (d132) if it is determined that the RR value is increased by the RR threshold or more, determining the probability of the emotional state as sadness as the second predetermined percentage. The method of claim 23, wherein step (d14) (d141) If it is determined in step (d111) that the HR value is not reduced by more than the HR threshold value, the N-SCR value is increased by an N-SCR threshold value or more and the SCRM value is increased by an SCRM threshold value or more. Determining whether it has been done; (d142) if it is determined that the N-SCR value is increased by the N-SCR threshold value and the SCRM value is increased by the SCRM threshold value or more, determining whether the SKT value is increased by the SKT threshold value or more; And (d143) if it is determined that the SKT value has been increased by more than the SKT threshold value, determining the probability of the emotional state being free as the second predetermined percentage. The method of claim 24, wherein step (d15) (d151) If it is determined in step (d142) that the SKT value is not increased by more than the SKT threshold value, it is determined whether the LF / HF value is reduced by more than the LF / HF threshold value, and the LF / HF value Proceeding to step (d142) if it is determined that the number has not been reduced by more than the LF / HF threshold value; (d152) if it is determined that the LF / HF value is reduced by more than the LF / HF threshold value, determining a probability that the emotional state is anger as the first predetermined percentage; (d153) determining whether the SCL value is increased by more than the SCL threshold value, and if it is determined that the SCL value is not increased by more than the SCL threshold value, proceeding to step (d152); And (d154) if it is determined that the SCL value is increased by more than the SCL threshold value, determining the probability that the emotional state is anger as the second predetermined percentage. The method of claim 25, wherein step (d) In step d121, it is determined that the LF value is not reduced by more than the LF threshold value, or in step d141, the N-SCR value is not increased by more than the N-SCR threshold value or the SCRM value. If it is determined that the increase has not been increased by the SCRM threshold or more, or in step (d131) it is determined that the RR value has not increased by the RR threshold, determining that the dominant emotional state does not exist. Emotional state recognition method characterized in that it comprises. The method of claim 20, wherein the step (d21) when the predetermined time is at least greater than the minimum time (d211) determining whether the RSA value is smaller than the RSA threshold value, the SCL value is greater than the SCL threshold value, and the SCRM value is greater than the SCRM threshold value; (d212) If it is determined that the RSA value is smaller than the RSA threshold value, the SCL value is larger than the SCL threshold value, and the SCRM value is larger than the SCRM threshold value, the N-SCR value is N-SCR. Determining whether it is as large as a threshold value or more; (d213) if it is determined that the N-SCR value is greater than or equal to the N-SCR threshold, determining whether the SKT value is greater than or equal to an SKT threshold; (d214) determining, as a third predetermined percentage, a probability that the emotional state is free when it is determined that the SKT value is greater than or equal to the SKT threshold value; (d215) determining whether the HF value is smaller than the HF threshold value, and if it is determined that the HF value is not smaller than the HF threshold value, proceeding to step (d214); (d216) if it is determined that the HF value is less than or equal to an HF threshold, determining a probability that the emotional state is free as a fourth predetermined percentage increased from the third predetermined percentage; (d217) determining whether the LF value is greater than or equal to the LF threshold value, and if it is determined that the LF value is not greater than or equal to the LF threshold value, proceeding to step (d216); And (d218) if it is determined that the LF value is greater than or equal to the LF threshold, determining a probability that the emotional state is free as a fifth predetermined percentage increased from the fourth predetermined percentage; Recognition method. The method of claim 27, wherein (d22) is (d221) if it is determined in step (d212) that the N-SCR value is not greater than the N-SCR threshold, determining the probability that the emotional state is anger as the fifth predetermined percentage; And (d222) if it is determined in step (d213) that the SKT value is greater than or equal to the SKT threshold value, determining a probability of the emotional state as anger as the third predetermined percentage; Way. The method of claim 28, wherein step (d23) (d231) If it is determined in step (d211) that the RSA value is not smaller than the RSA threshold value, the SCL value is not larger than the SCL threshold value, or the SCRM value is not greater than the SCRM threshold value, Determining whether the HR value is as small as an HR threshold value or more; (d232) if it is determined that the HR value is smaller than or equal to the HR threshold value, determining whether the RESP value is greater than or equal to an RESP threshold value; (d233) determining, by the third predetermined percentage, a probability that the emotional state is joy if it is determined that the RESP value is greater than or equal to the RESP threshold value; (d234) determining whether the LF / HF value is smaller than the LF / HF threshold value, and if it is determined that the LF / HF value is not smaller than the LF / HF threshold value, proceeding to step (d233); (d235) determining, by the fourth predetermined percentage, a probability that the emotional state is joy when it is determined that the LF / HF value is smaller than the LF / HF threshold value; (d236) determining whether the LF value is smaller than or equal to the LF threshold value, and if it is determined that the LF value is not smaller than or equal to the LF threshold value, proceeding to step (d235); And (d237) if it is determined that the LF value is smaller than or equal to the LF threshold value, determining the probability that the emotional state is joy as the fifth predetermined percentage. The method of claim 29, wherein step (d24) (d241) if it is determined in step (d232) that the RESP value is not greater than the RESP threshold value, determining whether the SCL value is smaller than the SCL threshold value; And and (d242) determining, by the fifth predetermined percentage, a probability that the emotional state is sad when it is determined that the SCL value is smaller than or equal to the SCL threshold. The method of claim 30, wherein step (d25) (d251) if it is determined in step (d231) that the HR value is not smaller than the HR threshold value, determining whether the N-SCR value is larger than the N-SCR threshold value; And and (d252) determining, by the fourth predetermined percentage, a probability that the emotional state is stress if it is determined that the N-SCR value is greater than or equal to the N-SCR threshold. 32. The method of claim 31, wherein step (d) If it is determined in step (d241) that the SCL value is not less than the SCL threshold value, or in step (d251) it is determined that the N-SCR value is not greater than the N-SCR threshold value, the dominant And determining that the emotional state does not exist.