JP5921973B2 - Synchro action detection method, detection apparatus, and detection program - Google Patents

Description

  The present invention relates to a synchronization behavior detection method, a detection apparatus, and a detection program. More specifically, the present invention relates to a technique for detecting a suitable synchronized behavior by using, for example, time-sequential behavior detection between people.

  People perform unique actions (hereinafter referred to as “synchronous actions”), such as high touch, banzai, thirty-three seven beats, and one-stop tightening, to facilitate communication. In addition, raising the sense of unity through synchronized rhythm-synchronized actions such as Bon Odori, children's hand play (for example), and mass games is also an expression of natural affinity that has been practiced since ancient times. Such synchronized behavior is one of the indicators for measuring human affinity, and if it can be quantified in some way, it will be useful for understanding various human relationships.

  In recent years, a system that promotes communication and extracts issues related to human relationships by analyzing and visualizing the connection between people using information and communication technology (hereinafter referred to as ICT). Many services have appeared. For example, the technology adopted in some systems called social networks is based on a technology that captures human relationships with an abstract model called a social graph and recommends information based on connections. In this case, a deep connection is automatically extracted based on the click history, comment history, link information, etc. of the participant (Non-patent Document 1).

  Traditional research in the field of interaction analysis and multi-person interaction analysis has shown how information and communication technology is used to understand how people interact with non-verbal communication such as position, posture, nodding and hand movements. Development of an assistant system according to the situation by utilizing and analyzing, and development of an information communication technology system for promoting communication have been performed (Non-Patent Document 2).

Miyawaki et al. “Discovery of Social Relations Based on Understanding Context of Interaction”, DEIM Forum 2011 A2-6, http://db-event.jpn.org/deim2011/proceedings/pdf/a2-6.pdf Bono et al. “Method of multi-person interaction research-analytical units and concepts for language / non-verbal communication research”, Journal of Artificial Intelligence, Vol. 22, No. 6 (2007/11), pp. 838-845. , 2007.

  However, in conventional research that analyzes human-human interaction using information communication technology, such as Non-Patent Document 1 and Non-Patent Document 2, it is not possible to generically detect moments of breath. . For this reason, it is hard to say that versatility is high.

  Therefore, an object of the present invention is to provide a detection method, a detection device, and a detection program for synchro behavior that can detect a moment when a breath comes in general.

In order to achieve this object, the synchronization behavior detection method according to claim 1 performs measurement using a sensor attached to the same part of each body for each synchronization behavior detection target, and for each synchronization behavior detection target. Acquiring acceleration vector measurement data in time series, converting the acceleration vector for each measurement time t into an acceleration scalar a (t) according to Equation 1, and the acceleration scalar for each synchronization action detection target a step of calculating _a cross-correlation ra (t ₀ ) related to acceleration at the same part of the body at each reference time t ₀ of the evaluation in the pair of synchronized behavior detection subjects using a (t), and the acceleration And performing a synchronization behavior evaluation at each reference time t ₀ of the evaluation based on the cross-correlation r _a (t ₀ ).

According to a fifth aspect of the present invention, there is provided the synchronization behavior detection device according to the synchronization behavior detection target for each synchronization behavior detection target, which is acquired in time series by measurement using a sensor attached to the same part of each body . Means for reading measured value data of an acceleration vector from a storage device; means for converting the acceleration vector at each measurement time t into an acceleration scalar a (t) according to Equation 1; and the acceleration scalar a for each synchronization action detection target person means for calculating _a cross-correlation ra (t ₀ ) related to acceleration at the same part of the body at each evaluation reference time t ₀ in the pair of synchronized behavior detection subjects using Equation (2) using (t); based on the cross-correlation r _a (t ₀₎ so that a means for evaluating the synchronous behavior for each reference time t ₀ of the evaluation.

According to a ninth aspect of the present invention, there is provided a synchronization behavior detection program for each synchronization behavior detection target for each synchronization behavior detection target, which is acquired in time series by measurement using a sensor attached to the same part of each body . Means for reading measured value data of an acceleration vector from a storage device; means for converting the acceleration vector for each measurement time t into an acceleration scalar a (t) according to Formula 1; and the acceleration scalar a (t for each synchronization action detection target ), A means for calculating _a cross-correlation ra (t ₀ ) related to acceleration at the same part of the body at each evaluation reference time t ₀ in the pair of subjects to be detected by the synchronization behavior using Equation 2, a cross-correlation r related to the acceleration _a (t ₀₎ the evaluation of the synchro act on not to cause a computer to function as a means for performing for each reference time t ₀ of the evaluation .

According to a second aspect of the present invention, there is provided a method for detecting a synchronized behavior, wherein each synchronized behavior detection target is measured using a sensor attached to the same part of each body, and the angular velocity vector is time-sequentially for each synchronized behavior detected subject. Obtaining the measured value data, converting the angular velocity vector for each measurement time t into an angular velocity scalar ω (t) according to Equation 3, and the angular velocity scalar ω (t) for each synchronization behavior detection target The step of calculating the cross-correlation r _ω (t ₀ ) related to the angular velocity at the same part of the body at each evaluation reference time t ₀ in the pair of synchronized behavior detection subjects using the equation 4 and the cross-correlation r _ω related to the angular velocity a step of evaluating the synchronization behavior based on (t ₀ ) every reference time t _{0 of the} evaluation.

According to a sixth aspect of the present invention, there is provided a device for detecting a synchronization behavior according to the synchronization behavior detection target for each synchronization behavior detection target, which is acquired in time series by measurement using a sensor attached to the same part of each body . Means for reading the measured value data of the angular velocity vector from the storage device; means for converting the angular velocity vector for each measurement time t into an angular velocity scalar ω (t) according to Equation 3; and the angular velocity scalar ω for each synchronization action detection target. means for calculating the cross-correlation r _ω (t ₀ ) related to the angular velocity at the same part of the body at each evaluation reference time t ₀ in the pair of synchronized behavior detection subjects using Equation (4) using (t), and the angular velocity Means for performing an evaluation of the synchro action at every reference time t ₀ of the evaluation based on the cross-correlation r _ω (t ₀ ).

According to a tenth aspect of the present invention, there is provided a synchronization behavior detection program for each synchronization behavior detection target for each synchronization behavior detection target, which is acquired in time series by measurement using a sensor attached to the same part of each body . Means for reading the measured value data of the angular velocity vector from the storage device, means for converting the angular velocity vector for each measurement time t into an angular velocity scalar ω (t) according to Equation 3, and the angular velocity scalar ω (t for each synchronization action detection target ), A means for calculating a cross-correlation r _ω (t ₀ ) related to the angular velocity at the same part of the body at each evaluation reference time t ₀ in the pair of subjects to be detected by the synchronization behavior, using Equation 4, a cross-correlation r related to the angular velocity _omega (t ₀₎ the evaluation of the synchro act on and cause the computer to function as means for performing for each reference time t ₀ of the evaluation That.

According to a third aspect of the present invention, there is provided a method for detecting a synchronized behavior, wherein a measurement is performed using a sensor attached to the same part of each body for each synchronized behavior detection target, and a magnetic field vector in time series for each synchronized behavior detection target. Using the measurement value data, converting the magnetic field vector for each measurement time t into an azimuth angle value θ (t), and the azimuth angle value θ (t) for each synchronization action detection target The step of calculating the cross-correlation r _θ (t ₀ ) related to the magnetic field at the same part of the body at each evaluation reference time t ₀ in the pair of synchronization behavior detection subjects by the equation 5 and the cross-correlation r _θ ( and a step of performing an evaluation of the synchronization behavior at every reference time t ₀ of the evaluation based on t ₀ ).

According to a seventh aspect of the present invention, there is provided the synchro behavior detection apparatus according to the synchro behavior detection target person acquired in time series by measurement using a sensor attached to the same part of each body for each sync action detection target person. Means for reading measured value data of a magnetic field vector from a storage device, means for converting the magnetic field vector for each measurement time t into an azimuth angle value θ (t), and the azimuth angle value θ ( means for calculating a cross-correlation r _θ (t ₀ ) related to the magnetic field at the same part of the body at each evaluation reference time t ₀ in the pair of subjects to be detected by the synchronization behavior using Equation 5; And a means for performing an evaluation of the synchronization behavior at every reference time t ₀ of the evaluation based on the correlation r _θ (t ₀ ).

According to another aspect of the present invention, there is provided a synchronization behavior detection program for each synchronization behavior detection target for each synchronization behavior detection target, which is acquired in time series by measurement using a sensor attached to the same part of each body . Means for reading magnetic field vector measurement value data from a storage device, means for converting the magnetic field vector for each measurement time t into an azimuth angle value θ (t), and the azimuth angle value θ (t) for each synchronization action detection target The means for calculating the cross-correlation r _θ (t ₀ ) related to the magnetic field in the same part of the body at each evaluation reference time t ₀ in the pair of subjects to be detected by the synchronization behavior using Equation 5, the cross-correlation r _θ related to the magnetic field Based on (t ₀ ), the computer is caused to function as a means for performing the evaluation of the synchro action at each evaluation reference time t ₀ .

Therefore, according to the detection method, detection apparatus, and detection program for the synchronization behavior, the time-series acceleration vector obtained by the measurement using the sensor is converted into the acceleration scalar and the time-series angular velocity vector is converted into the angular velocity scalar. The normalized cross-correlation is calculated at each evaluation reference time t ₀ using the acceleration scalar and the angular velocity scalar, and the evaluation of the synchronization behavior is evaluated based on the sum or product of the calculated normalized cross-correlation values. Since this is performed every reference time t _0, the evaluation of the synchronization behavior at the moment (or a slight time width) is performed continuously in time series.

  And the structure and the operation effect peculiar to the present invention for making it possible to detect the moment when the breaths come together as a synchronized action for general purposes are arranged as follows.

  First, the first point of the present invention is that attention is paid to “phase”. Although several studies have been conducted to grasp the relationship between people and people and people and things using acceleration sensors, etc., it is common to perform frequency analysis (or zero crossing analysis). The fine phase component has been almost ignored. This is because it has been considered that a fine phase component is nothing more than noise to capture the similarity of movement such as “walking together” and “talking”. However, in synchro behavior, it is important that the timing in the time direction matches, that is, the phases are aligned. For example, it cannot be said that breathing is tight with a single tightening that is slightly out of timing. In the present invention, the synchronization behavior is evaluated by paying attention to the “phase”, and therefore the degree of coincidence of timing in the time direction is evaluated.

  The second point of the present invention is that it is possible to perform detection that is not easily influenced by differences in sensor mounting position and physical characteristics. Even if the sensor is attached to any part of the body, such as an arm or leg, it is impossible for multiple people to accurately align the mounting position, and there are differences in movement trends such as arm length and swing size. . On the other hand, such differences in sensor mounting position, physical characteristics, and movement tendencies are considered to be unrelated to the “smooth” movement as a synchronized movement. A detection method that is less susceptible to inconsistencies is needed. And, in the present invention, the detection of synchro action that is not easily affected by inconsistencies in sensor mounting positions, physical characteristics, and movement tendencies is made. Is done.

  The third point of the present invention is that the disclosure range of the data itself measured by the sensor (so-called “raw data”) is narrow. The longer the sensor is worn on the body, the more private data it contains. Even if there is no problem that the person can browse such raw data, it is considered that there are many people who are reluctant to have others browse their own raw data. In the present invention, since it is not necessary to disclose the raw data to others, the data itself is not known to others.

  According to the detection method, the detection device, and the detection program for the synchronization behavior of the present invention, the synchronization behavior can be evaluated in a time series continuously in time (or a slight time width). The combined moment can be detected for general purposes, and the versatility can be improved as a method for detecting the synchro action.

It is a flowchart explaining an example of embodiment of the detection method of the synchronization action of this invention. It is a functional block diagram of the detection apparatus of a synchronization action implement | achieved by the said program in the case of implementing the detection method of the synchronization action of embodiment using the detection program of a synchronization action. It is a figure which shows the example of a waveform of the combination data of the measured value measured and acquired with the sensor.

  Hereinafter, the configuration of the present invention will be described in detail based on an example of an embodiment shown in the drawings.

  FIG. 1 to FIG. 3 show an example of embodiments of a detection method, a detection apparatus, and a detection program for synchro behavior according to the present invention.

The synchronization behavior detection method according to the present embodiment is a step of performing measurement using a sensor for each synchronization behavior detection target person and acquiring combination data of measurement values of the acceleration vector and the angular velocity vector in time series for each synchronization behavior detection target person. (S1) and the acceleration vector for each measurement time t is converted to an acceleration scalar a (t) by Equation 6-1 and the angular velocity vector for each measurement time t is converted to an angular velocity scalar ω (t) by Equation 6-2. a step (S2), the cross-correlation r _a (t ₀ regarding the acceleration of each reference time t ₀ of the evaluation in synchronous action detection subject pairs by equation 7 by using synchrotron action detection subject-specific acceleration scalar a (t) ) And using the angular velocity scalar ω (t) for each synchronization behavior detection target, the evaluation of the pair of synchronization behavior detection targets by Equation 8 Calculating a cross-correlation r _ω (t ₀ ) related to the angular velocity at each reference time t ₀ and calculating a sum R (t ₀ ) of these cross-correlations at each evaluation reference time t ₀ (S3); A step (S4) of performing an evaluation on the presence / absence of a synchronization action at each evaluation reference time t ₀ based on the total R (t ₀ ).

In addition, the synchronization behavior detection device of the present embodiment is a combination data of acceleration vector and angular velocity vector measurement values for each synchronization behavior detection target acquired in time series by measurement using a sensor for each synchronization behavior detection target. Is read from the data server 16 serving as a storage device, and an acceleration vector for each measurement time t is converted into an acceleration scalar a (t) by Equation 6-1 and a measurement time t by Equation 6-2. A measurement data conversion unit 11b serving as a means for converting another angular velocity vector into an angular velocity scalar ω (t), and an acceleration scalar a (t) for each synchronization behavior detection target, a pair of synchronization behavior detection subjects according to Equation (7). synchro behavior detected with calculating the cross-correlation r _a (t ₀₎ relating to the acceleration of each reference time t ₀ of the evaluation in Total R of the cross-correlation by calculating the cross-correlation r _omega (t ₀₎ for the reference time t ₀ for each of the angular velocity of the evaluation in synchronous action detection subject pairs by Equation 8 with another angular velocity scalar omega (t) The cross-correlation calculation unit 11c as means for calculating (t ₀ ) at every evaluation reference time t _0, and the evaluation of the presence / absence of the synchronization behavior based on the total R (t ₀ ) of the cross-correlation, etc. and a synchronization evaluation unit 11d as means for performing every t ₀ .

Furthermore, the synchronization behavior detection program of the present embodiment is a combination data of acceleration vector and angular velocity vector measurement values for each synchronization behavior detection target person acquired in time series by measurement using a sensor for each synchronization behavior detection target person. From the storage device, the acceleration vector for each measurement time t is converted to an acceleration scalar a (t) by Equation 6-1 and the angular velocity vector for each measurement time t is converted to an angular velocity scalar ω (t) by Equation 6-2. The cross correlation r _a (t ₀ ) relating to the acceleration at each evaluation reference time t ₀ in the pair of synchronized behavior detection subjects is calculated by Equation 7 using the conversion means, the acceleration scalar a (t) for each synchronized behavior detection subject. Using the angular velocity scalar ω (t) for each synchronization action detection target and calculating the pair of synchronization action detection targets according to Equation 8 Means for calculating the cross-correlation r _ω (t ₀ ) related to the angular velocity at each evaluation reference time t ₀ and calculating the sum R (t ₀ ) of these cross-correlations at each evaluation reference time t ₀ , the sum of the cross-correlations Based on R (t ₀ ), the computer is caused to function as a means for performing an evaluation on the presence / absence of a synchronization action at every evaluation reference time t ₀ .

  In executing the synchro action detection method, first, measurement by a sensor is performed (S1).

  In the present invention, the acceleration vector and the angular velocity vector are continuously measured in time by a sensor attached to the body of the subject person who detects the synchronization behavior. In the present embodiment, it is assumed that three users, a user B, a user B, and a user C, are wearing sensors as subjects of detection of the synchronization behavior according to the present invention.

  Note that the sensor used in the present invention is not limited to a specific device as long as it can be attached to the body and can obtain data relating to movement / operation attributes / specifications at the wearing site. Specifically, for example, when using measurement data of acceleration vectors and angular velocity vectors as in this embodiment, it is built into a small three-axis acceleration sensor, pedometer or activity meter built into various portable terminals. It can be constituted by an acceleration sensor or the like.

  In the present invention, the body part to which the sensor is attached is not limited to a specific part as long as it is a part where human movement can be measured. Specifically, for example, the wrist, ankle, head, waist Etc. Here, the target person (in this embodiment, the user A, the user B, and the user C) of the detection of the synchronization behavior according to the present invention wears the sensor on the same part of each body.

  Note that the number of sensors to be mounted in the present invention is not limited to a specific number, and may be one (for example, only one of the left and right wrists, only the head, etc.) or a plurality of sensors ( For example, both left and right wrists, wrists and ankles).

  In addition, it is preferable that the greater the number of sensors attached and the more motion information obtained from one sensor, since it is generally possible to detect synchronization behavior with higher accuracy.

  FIG. 3 shows a waveform example of combination data of time-series measurement values measured and acquired continuously in time using sensors attached to the users A, B, and C, respectively.

  Next, the sensor measurement data obtained by the process of S1 is converted into a feature amount (S2).

  Here, the processing from S2 to S4 in the synchronization behavior detection method of the present invention can be executed by the synchronization behavior detection device of the present invention.

  And the process from S2 to S4 in the synchronization behavior detection method and the synchronization behavior detection apparatus for executing the processing can also be realized by executing the synchronization behavior detection program of the present invention on a computer. In the present specification, a case will be described in which a synchronization behavior detection apparatus is realized by executing a synchronization behavior detection program on a computer and the processing from S2 to S4 in the synchronization behavior detection method is performed.

  FIG. 2 shows an overall configuration of a computer 10 (which is also a synchronization behavior detection device 10 in the present embodiment) for executing the synchronization behavior detection program 17. The computer 10 (synchronous behavior detection device 10) includes a control unit 11, a storage unit 12, an input unit 13, a display unit 14, and a memory 15, and is connected to each other by a signal line such as a bus. Further, a data server 16 as a storage device is connected to the computer 10 via a signal line such as a bus, and signals such as data and control commands are transmitted and received (that is, input / output) through the signal line. .

  The control unit 11 performs the calculation related to the control of the entire computer 10 and the detection process of the synchronization behavior by the synchronization behavior detection program 17 stored in the storage unit 12, and is, for example, a CPU (Central Processing Unit).

  The storage unit 12 is a device that can store at least data and programs, and is, for example, a hard disk.

  The memory 15 serves as a memory space that is a work area when the control unit 11 executes various controls and operations, and is a RAM (Random Access Memory), for example.

  The input unit 13 is an interface for giving at least an operator's command to the control unit 11, and is, for example, a keyboard.

  The display unit 14 performs drawing / display of characters, graphics, and the like under the control of the control unit 11 and is, for example, a display.

  In the present embodiment, the acceleration vector and the angular velocity vector relating to each of the user A, the user B, and the user C, which are the synchronization behavior detection target persons measured and acquired in the measurement (S1) process by the above-described sensor, are obtained. The combination data (measured as time series data) of the measurement values is stored (saved) in advance in the data server 16 as the measurement value database 18. The combination data of the measurement values is recorded in association with the identifier for each synchronization behavior detection target person and in association with the measurement time.

  In addition, the recording of the combination data of the measurement values measured using the sensor in the measurement value database 18 stored (saved) in the data server 16 is, for example, a mechanism (specifically, a sensor transmits the measurement value data). Is provided with a transmission mechanism using, for example, radio waves or infrared rays, and a reception mechanism paired with the transmission mechanism is connected to the data server 16 so that the received data are sequentially recorded and stored in the measurement value database 18. Alternatively, the sensor may be provided with a mechanism for recording the measurement value data (specifically, for example, an internal memory or a removable storage medium), and the sensor and the data server 16 may be connected by wire. The combination data of the measurement values in the internal memory of the storage medium is recorded in the measurement value database 18 or the storage medium attached to the sensor. It may be connected to the data server 16 by (attached) or so as to record the combination data of the measurement values in the storage medium in the measurement database 18.

  In the present invention, the database or data file in which the combination data of the measurement values obtained by the measurement using the sensor is stored (saved) is the control unit of the computer 10 (synchronous behavior detection device 10). 11 may be any storage device that can be accessed, and is not limited to the data server 16. Specifically, for example, it may be stored in the storage unit 12 that is an internal storage device, or may be stored in various storage media such as an optical storage medium or an external storage device.

Then, the control unit 11 of the computer 10 (which is also the sync action detection device 10 in this embodiment) executes the sync action detection program 17, so that the user A who is the sync action detection target user, the use Combination of measured values of acceleration vector and angular velocity vector for each of user A, user B, and user C, who are synchronization behavior detection targets acquired in time series by measurement using sensors for each of user B and user C The data reading unit 11a as means for reading data from the data server 16 as a storage device, the acceleration vector for each measurement time t is converted into an acceleration scalar a (t) by Equation 9-1, and the measurement time t by Equation 9-2. Measurement data conversion unit 11b as means for converting another angular velocity vector into an angular velocity scalar ω (t), a synchronization behavior detection target The user A, the user B, and the user C who are the synchronization behavior detection target user A, the user B, and the user C using the acceleration scalar a (t) for each of the user A, the user B, and the user C. And calculating _a cross-correlation ra (t ₀ ) related to the acceleration at each evaluation reference time t ₀ in the pair of synchronization behavior detection subjects selected by the user A, user B, who are the synchronization behavior detection subjects, By using the angular velocity scalar ω (t) for each user C, the cross-correlation r _ω (t ₀ ) related to the angular velocity at each evaluation reference time t ₀ in the pair of subjects to be synchronized is detected by Equation 11 to calculate these cross-correlations. The cross-correlation calculation unit 11c as means for calculating the total R (t ₀ ) at each evaluation reference time t ₀ , and the evaluation of the presence / absence of the synchronization behavior is evaluated based on the total R (t ₀ ) of the cross-correlation As a means to be performed at every reference time t ₀ A synchronization evaluation unit 11d is configured.

  And the data reading part 11a comprised by the control part 11 of the computer 10 (synchronous action detection apparatus 10) by executing the synchronous action detection program 17 reads the combination data of the measurement value according to a synchronous action detection subject. I do.

  Specifically, the data reading unit 11a is measured and acquired by using a sensor for each synchronization behavior detection target person in the process of S1, and is recorded in the measurement value database 18 stored in the data server 16. The combination data of the measured values of the acceleration vector and the angular velocity vector for each person to be detected is read from the data server 16. Then, the data reading unit 11 a stores the combination data of the measurement values in the memory 15. The identifier and the measurement time for each synchronization behavior detection target associated with the measurement data combination data are read and the measurement value combination data is stored in the memory 15 in association with the identifier and the measurement time.

  Subsequently, the measurement data conversion unit 11b converts the combination data of the measurement values.

  Specifically, the measurement data conversion unit 11b reads the combination data of the measurement values of the acceleration vector and the angular velocity vector stored in the memory 15 by the data reading unit 11a from the memory 15, converts the acceleration vector into an acceleration scalar, and converts the angular velocity. Convert vector to angular velocity scalar. The measurement value combination data conversion process (S2) is repeatedly performed for each synchronization behavior detection target person for each measurement time associated with the measurement value combination data stored in the memory 15.

Performed by Equation 1 3 and converted into the angular velocity scalar converting the angular velocity vector of the acceleration scalar acceleration vector. Specifically, when the acceleration vector and the angular velocity vector at the measurement time t are defined as Equations 12-1 and 12-2, respectively, an acceleration scalar a (t) [m / s ² ] and an angular velocity scalar ω (t ) [Rad / s] is expressed as Equation 13-1 and Equation 13-2, respectively.

  By scalarizing the vector quantity in this way, it is possible to eliminate the influence of the difference in the initial value due to the difference in the mounting position of the sensor and the difference in the direction of the gravitational acceleration. If it is worn on the right arm, etc., the same data will be obtained except for the amplitude when each person performs the same operation. Note that the amplitude is affected by, for example, the length of the arm and the height, so that even if each person performs the same operation, the data is not necessarily the same as the scalar value.

  The measurement data conversion unit 11b converts the acceleration vector into an acceleration scalar a (t) using Equation 13-1, and further converts the angular velocity vector into an angular velocity scalar ω (t) using Equation 13-2. Then, the measurement data conversion unit 11b associates the converted combination data of the acceleration scalar a (t) and the value of the angular velocity scalar ω (t) with the identifier and measurement time for each synchronization behavior detection target person in the memory 15. Remember.

  Next, the cross-correlation calculation unit 11c calculates the sum of the cross-correlation using the acceleration scalar and the angular velocity scalar obtained by the process of S2 (S3).

  Specifically, the cross-correlation calculation unit 11c uses the combination data of the acceleration scalar a (t) and the angular velocity scalar ω (t) for each measurement time for each synchronization behavior detection target person stored in the memory 15 in the process of S2. Read from the memory 15 and use this combination data to calculate the total cross-correlation. It should be noted that the calculation of the sum of cross correlations (S3) is repeated for each combination of all the combinations of the two selected from the synchronized behavior detection target group stored in the memory 15. In other words, in the present embodiment, there are three users A, user B, and user C, and the user A-user B, user A-user C, and user B. -Repeat for each combination of user C.

  Calculation of the sum of the cross-correlation using the acceleration scalar a (t) and the angular velocity scalar ω (t) is performed by Expressions 14 to 16 and Expressions 17 to 21.

In the case of the formulas 14 to 16, specifically, for the pair (in other words, the combination) of the user A and the user B, the acceleration scalar calculated from the measured value of the sensor of the user A is a ₁ , the angular velocity scalar. Ω ₁ , an acceleration scalar calculated from the measurement value of the sensor of the user B is a ₂ , an angular velocity scalar is ω ₂ , and the evaluation reference time is t ₀ , the normalized cross-correlation r _a (t ₀₎ was calculated by equation 14 relates to the acceleration of each reference time t ₀ of the evaluation based on the measurement values of the sensor of the user B, the sensor of the user a and the sensor of the user B The normalized cross-correlation r _ω (t ₀ ) related to the angular velocity at each evaluation reference time t ₀ based on the measured value is calculated by Equation 15.

The average value of the acceleration scalar and the average value of the angular velocity scalar are the average values of acceleration and angular velocity within the target time for the synchronization behavior evaluation. That is, it is an average value in the past range from the evaluation reference time t ₀ , and is continuously updated with the progress of the time (that is, the evaluation reference time t ₀ ). The past range for which the average value is to be calculated is not limited to a specific range. For example, the entire range from the measurement start time to the evaluation reference time t ₀ may be the target. from the reference time t ₀ may be a certain period of time. Further, in the case where the past of a certain period from the evaluation reference time t _{0 is} the target of the average value calculation, the certain period is not limited to a specific period. Specifically, for example, the past several seconds or several tens of seconds may be used. It may be in the past several minutes or several tens of minutes, and can be set as appropriate depending on the content and purpose of the synchronization behavior evaluation.

  Then, as in Equations 14 and 15, the average value in the comparison interval is reduced and divided by the standard deviation, so that the similarity is not the magnitude of the values of the acceleration scalar and the angular velocity scalar but the relative ratio of the values. Will be judged. Accordingly, even if the entire signal pattern measured and acquired using the sensor is raised or the overall contrast fluctuates, it is possible to perform collation according to the similarity of the patterns. Specifically, even in the same movement, the overall movement of the other signal pattern is raised with respect to one with the exaggerated movement and the modest movement, but even in that case, if the movement mode is consistent, it is only a pattern It is possible to collate and determine the similarity of each other, and if the length of the arm is different, the overall contrast will fluctuate. It becomes possible to collate and determine similarity.

Then, the sum R (t ₀ ) of the normalized cross-correlation is calculated for each evaluation reference time t ₀ using Equation 16.
(Equation 16) R (t ₀ ) = r _a (t ₀ ) + r _ω (t ₀ )

Then, the sum R (t ₀ ) of normalized cross-correlations in a fixed time width (τ) calculated by Expression 16 is defined as the synchronization rate at the evaluation reference time (t ₀ + τ).

Note that the evaluation reference time t ₀ = measurement time t. In this embodiment, the evaluation time is from “evaluation reference time t ₀ ” to “evaluation reference time t ₀ + time width τ”, but the evaluation time is “evaluation reference time t ₀ −τ / 2”. To “evaluation reference time t ₀ + τ / 2”.

  The synchronization rate in the present invention expressed by the mathematical formulas 14 to 16 directly calculates the correlation between waveforms, so that the values are small when the phases are shifted even if the waveforms are similar, and zero in the opposite phase. Therefore, it is appropriate as an index for evaluating a “breathing” state. Furthermore, the synchronization rate in the present invention expressed by the mathematical formulas 14 to 16 is normalized for correlation, so that the size of each waveform, that is, the size of the body, the length of the arm, etc., has a small effect on the synchronization rate. Therefore, the degree of matching of the waveform phase is given the highest priority as the evaluation of the degree of synchronization, and the similarity as the shape becomes dominant next. It is appropriate as an indicator to

  Here, when the time width τ for calculating the correlation is short, only the latest sync operation is detected as the sync rate. On the other hand, when the time width τ for calculating the correlation is long, the past operation is used as the sync rate. Will be taken into account. The time width τ for calculating the correlation is not limited to a specific value, and may be set as appropriate based on the purpose of the synchronization operation evaluation. Specifically, for example, about 1 [second]. It is possible to do.

In consideration of human forgetting, the weight is set such that the closer the evaluation time is to the reference time t ₀ , the lower the synchronization rate (in other words, the higher the synchronization time is toward the end of the evaluation time (t ₀ + τ)). You may make it attach.

Note that Equations 14 to 16 use the acceleration scalar a and the angular velocity scalar ω calculated from the measured values of one sensor each of the user A and the user B as the synchronization rate. An equation for calculating the sum R (t ₀ ) of normalized cross-correlation. When each user wears a plurality of sensors, the acceleration scalar a and the angular velocity calculated from the measured values of the plurality of sensors All of the scalar ω may be used. When the measurement value data of a plurality of sensors is used, all normalized cross-correlations of the acceleration scalar a and the angular velocity scalar ω calculated from the measurement values of the plurality of sensors are added together.

For example, user A wears sensor α ₁ and sensor α ₂ and user B wears sensor β ₁ and sensor β _2, and user A's sensor α ₁ and user B's sensor β ₁ is attached to the same part of the body, and the sensor α ₂ of the user A and the sensor β _{2 of the} user B are attached to the same part of the body.

The acceleration scalar calculated from the measured value of the sensor α ₁ of the user A is a ₁ , the angular velocity scalar is ω ₁ , the acceleration scalar calculated from the measured value of the sensor α ₂ is a ′ ₁ , and the angular velocity scalar is ω. ' ₁ and the acceleration scalar calculated from the measured value of the sensor β ₁ of the user B is a ₂ , the angular velocity scalar is ω _2, and the acceleration scalar calculated from the measured value of the sensor β ₂ is a' ₂ , the angular velocity scalar and ω _'2.

In this case, the normalized cross-correlation r a — ₁ (t ₀ ) related to the acceleration based on the measured values of the sensor A ₁ of the user A and the sensor β _{1 of the} user B is calculated by the equation 17, and the sensor of the user A alpha ₂ and the user B of the sensor beta ₂ normalized cross-correlation relates to the acceleration based on the measured value of r _{a_2} a (t ₀₎ as well as calculated by equation 18, the sensor of the user B and the sensor alpha ₁ of the user a beta _The normalized cross-correlation r _{ω — 1} (t ₀ ) related to the angular velocity based on the measured value of ₁ is calculated by Equation 19, and the normality related to the angular velocity based on the measured values of the sensor A ₂ of the user A and the sensor β _{2 of the} user B is calculated. The cross-correlation r _{ω_2} (t ₀ ) is calculated by Equation 20.

Then, the sum R (t ₀ ) of normalized cross-correlations as the synchronization rate is calculated for each evaluation reference time t ₀ using Equation 21.
( _Expression 21) R (t ₀ ) = r _{a_1} (t ₀ ) + r _{a_2} (t ₀ ) + r _{ω_1} (t ₀ ) + r _{ω_2} (t ₀ )

The cross-correlation calculation unit 11c calculates the sum R (t ₀ ) of normalized cross-correlation using Expressions 14 to 16 and Expressions 17 to 21. Then, the cross-correlation calculation unit 11c stores the calculated normalized cross-correlation sum R (t ₀ ) as a synchronization rate in association with the identifier and the evaluation reference time t ₀ for each synchronization action detection target person combination. 15 is stored.

  Next, the synchronization evaluation unit 11d evaluates the presence / absence of synchronization behavior using the synchronization rate obtained by the process of S3 (S4).

Specifically, the synchronization evaluation unit 11d determines the synchronization rate by the reference time t _{0 for} each combination of synchronization behavior detection subjects stored in the memory 15 in the process of S3 (that is, the sum R of normalized cross-correlations). The value of (t ₀ )) is read from the memory 15, and the presence / absence of a synchronization action is evaluated using this value. Note that the synchronization behavior evaluation (S4) is performed for each combination of synchronization behavior detection subjects associated with the value of the synchronization rate (R (t ₀ )) stored in the memory 15. t ₀ repeated separately.

The evaluation of the synchronization behavior may be performed by, for example, displaying the combination of users and the reference time t _{0 of the} evaluation side by side in descending order of the synchronization rate, or displaying a time-series synchronization rate value for each combination of users in a line graph. The user is synchronized in behavior at the reference time t ₀ of the combination / evaluation of users whose synchronization rate is higher than the threshold by setting a threshold for the synchronization rate (in other words, , And there is a synchro action). Note that the threshold of the synchronization rate for determining the presence / absence of the synchronization behavior is not limited to a specific value, and the actual calculation result of the synchronization rate (that is, the sum R (t ₀ ) of normalized cross-correlations). This should be set as appropriate. Specifically, for example, when the sum of the cross-correlations is calculated by using the formulas 14 to 16, it may be set in the range of 1.0 to 1.8, and the sum of the cross-correlations is calculated by using the formulas 17 to 21. In this case, it is conceivable to set in the range of 2.0 to 3.6.

The synchro evaluation unit 11d, for example, or is displayed on the display unit 14 side by side reference time t ₀ of the combination and evaluation of the user in the order value of the synchronous rate is high, the synchronous rate time series for each combination of the user or to display a line graph of the values on the display unit 14, or to display the reference time t ₀ of the combination and evaluation of the user sync rate is larger than the threshold value of the synchronous rate on the display unit 14.

In addition, by performing the processing up to S3 for the waveform example of the combination data of the measurement values shown in FIG. 3, the synchronization behavior is detected between the user A and the user C (the range indicated by the symbol S in FIG. 3). ). In FIG. 3, the acceleration scalar value (upper line graph) obtained by converting the acceleration vector every 40 [milliseconds] and the angular velocity scalar value (lower line graph) obtained by converting the angular velocity vector are displayed. In addition, the correlation is calculated for the past 2 seconds from the evaluation reference time t ₀ (ie, τ = 2 seconds).

  And the control part 11 complete | finishes the calculation which concerns on the detection process of a synchronous action (END).

  According to the detection method, the detection apparatus, and the detection program for the synchronization behavior of the present invention having the above-described configuration, it is possible to continuously evaluate the synchronization behavior at a moment (or time width τ) in time series. Therefore, it is possible to detect the moment of breathing for a general purpose, and it is possible to improve the versatility as a method for detecting a synchronization behavior.

  In addition, although the above-mentioned form is an example of the suitable form of this invention, it is not limited to this, A various deformation | transformation implementation is possible in the range which does not deviate from the summary of this invention. For example, in the above-described embodiment, the synchronization rate is calculated by adding the normalized cross-correlation of the acceleration scalar a and the normalized cross-correlation of the angular velocity scalar ω. However, the present invention is not limited to this. The synchronization rate may be set by only one of the normalized cross-correlation and the normalized cross-correlation of the angular velocity scalar ω.

  Further, in the above-described embodiment, the measurement value data of the acceleration vector and the measurement value data of the angular velocity vector are used to determine the presence / absence of the synchronization action, but the types of measurement value data that can be used in the present invention are as follows. The data is not limited to these as long as the data is related to the attributes / specifications of body movements. Specifically, for example, measured value data of a magnetic field vector obtained by measurement using a geomagnetic sensor may be used. In addition, it becomes possible to consider the direction of body movement by using the magnetic field vector in addition to the acceleration vector and the angular velocity vector. Therefore, when it is not necessary to consider the direction of the body and movement (specifically, for example, when determining the presence or absence of synchronized behavior when talking face to face or facing each other), using magnetic field vectors is useful. However, the use of magnetic field vectors is highly useful when evaluating synchronized behavior including the direction of motion when the body is aligned.

When using the magnetic field vector, for example, measurement is performed using a sensor for each synchronization behavior detection target person, and measurement value data of the magnetic field vector is acquired in time series for each synchronization behavior detection target person (and data The reading unit 11a reads), and calculates the azimuth angle value θ (0 to 2π) from the magnetic field vector for each measurement time t (in other words, converts the magnetic field vector into the azimuth angle value θ) (the measurement data conversion unit 11b converts it). The cross-correlation r _θ (t ₀ ) related to the magnetic field at each reference time t ₀ of the evaluation of the pair of synchronization behavior detection subjects is calculated by Equation 22 using the azimuth angle value θ for each synchronization behavior detection subject ( correlation calculation unit 11c calculates), the cross-correlation relates to magnetic field r _theta (t ₀₎ in the evaluation of synchronous behavior for each reference time t ₀ of the evaluation on the basis of (synchro evaluation unit 11d does) Unisuru. The method for calculating the azimuth angle value θ from the magnetic field vector is appropriately selected according to the geomagnetic sensor that has measured and acquired the magnetic field vector.

  Further, the number of axes (dimensions) of various vector data used in the present invention may be one axis, two axes, or three axes. However, the 2-axis rather than the 1-axis and the 3-axis rather than the 2-axis are preferable in terms of being able to evaluate a finer similarity in body motion.

  In the above-described embodiment, the synchronization rate is calculated by adding the normalized cross-correlation of the acceleration scalar a and the normalized cross-correlation of the angular velocity scalar ω. However, the present invention is not limited to this. The synchronization rate may be calculated by multiplying the normalized cross-correlation and the normalized cross-correlation of the angular velocity scalar ω. When the synchronization rate is defined by the sum of normalized cross-correlation, the synchronization rate increases if the degree of coincidence is high even for some of the sensor data, while the synchronization rate is calculated by the product of normalized cross-correlation. When defined, when there is sensor data with a high correlation and sensor data with a low correlation, the synchronization rate decreases as a result, and a strict evaluation is performed as a determination of the presence or absence of a synchronization action.

  In the above-described embodiment, the synchronization rate is calculated by adding the cross-correlation of the acceleration scalar a and the cross-correlation of the angular velocity scalar ω. However, the synchronization rate for evaluating / determining the presence / absence of the synchronization action is used. The calculation of is not limited to this. That is, first, as described above, the data that can be used in the present invention is not limited to these two (specifically, as described above, for example, it is considered to use the cross-correlation of the azimuth angle value θ of the magnetic field vector). ). In addition, when calculating the synchronization rate by adding or multiplying the cross-correlations of multiple indicators, the combination of the two indicators is not limited to the combination of acceleration and angular velocity. The number may be three or more. Specifically, for example, the cross correlation of acceleration and the cross correlation of magnetic field may be combined (sum, product), the cross correlation of angular velocity and the cross correlation of magnetic field may be combined, You may make it combine the cross correlation of acceleration, the cross correlation of angular velocity, and the cross correlation of a magnetic field.

  In the above-described embodiment, the combination data of the measured values of the acceleration vector and the angular velocity vector measured and acquired using the sensor is recorded in the measured value database 18 stored (saved) in the data server 16 and the measured value. The example has been described on the assumption that the data read from the database 18 is read and the control unit 11 of the synchronization behavior detection device 10 performs a calculation related to the synchronization behavior detection processing. Temporary storage in the data server 16) in the above embodiment is not an essential configuration in the present invention. That is, the combination data of the measurement values measured using the sensor is directly input to the synchronization behavior detection device 10, and the control unit 11 of the synchronization behavior detection device 10 performs an operation related to the synchronization behavior detection processing immediately and simultaneously. May be performed. In this case, for example, the sensor is provided with a mechanism for transmitting measurement value data (specifically, for example, a transmission mechanism using radio waves, infrared rays, etc.), and a reception mechanism that is paired with the transmission mechanism is synchronized. The measurement value data received by the behavior detection device 10 is directly input to the synchronization behavior detection device 10 and the control unit 11 of the synchronization behavior detection device 10 immediately and simultaneously performs the synchronization behavior. The calculation related to the detection process may be performed.

  Further, in the above-described embodiment, the example has been described on the assumption that the sensor and the synchro action detection device 10 having the display unit 14 are separate from each other, but the configuration of the present invention is not limited to this. The sensor may include a display unit. In this case, the sensor further includes a mechanism for transmitting the measurement value data and a mechanism for receiving a calculation result related to the synchronization action detection process performed by the synchronization action detection device 10 and a synchronization action detection device. 10 control units 11 perform operations related to the detection process of the synchronization action immediately and simultaneously, so that the synchronization action detection target person wearing the sensor can immediately and simultaneously detect the presence or absence of the synchronization action and the synchronization. It becomes possible to know the other party of the action.

  Further, in the above-described embodiment, the case where the synchronization behavior detection target persons are three users A, B, and C has been described. However, the number of the synchronization behavior detection target group to which the present invention can be applied is three. The present invention is not limited, and the present invention can be applied to a group of two or more synchronized behavior detection subjects.

  Further, in the above-described embodiment, an example has been described on the assumption that a human wears a sensor to detect a synchronization behavior between humans, but the application target of the present invention is not limited to a human, A sensor may be attached to a person and an animal to detect a synchronization action between the person and the animal, or a sensor may be attached to an animal to detect a synchronization action between the animals. good.

DESCRIPTION OF SYMBOLS 10 Synchro action detection apparatus 11 Control part 12 Storage part 13 Input part 14 Display part 15 Memory 16 Data server 17 Synchro action detection program 18 Measurement value database

Claims (12)

A step of performing measurement using a sensor attached to the same part of each body for each synchronization behavior detection target person to acquire measurement value data of the acceleration vector in time series for each synchronization behavior detection target person, and the following formula 1

By using the step of converting the acceleration vector for each measurement time t into an acceleration scalar a (t) and the acceleration scalar a (t) for each synchronization action detection target,

Wherein calculating the cross-correlation r _a (t ₀₎ relating to the acceleration at the same site of the synchronous action detecting subjects for each reference time t ₀ of the evaluation in pairs of the body, the acceleration correlation regarding r _a by (t ₀₎ And performing a synchronization behavior evaluation at each reference time t _{0 of the} evaluation. Performing measurement using sensors attached to the same part of each body for each synchronization behavior detection target person to obtain measurement value data of angular velocity vectors in time series for each synchronization behavior detection target person;

Using the step of converting the angular velocity vector for each measurement time t into the angular velocity scalar ω (t) and the angular velocity scalar ω (t) for each synchronization action detection target,

Calculating a cross-correlation r _omega (t ₀₎ relating to the angular velocity at the same site of the synchronous action detecting subjects for each reference time t ₀ of the evaluation in pairs of the body by the cross-correlation r _omega concerning the angular velocity (t ₀₎ And performing a synchronization behavior evaluation at each reference time t _{0 of the} evaluation. Performing measurement using a sensor attached to the same part of each body for each synchronization behavior detection target person to acquire measurement value data of magnetic field vectors in time series for each synchronization behavior detection target person, and for each measurement time t The step of converting the magnetic field vector into an azimuth angle value θ (t) and the azimuth angle value θ (t) for each synchronization action detection target person

Calculating a cross-correlation r _theta (t ₀₎ relating to the magnetic field at the same site of the body of each reference time t ₀ of evaluation in the synchronous action detection subject pairs by cross-correlation r _theta regarding the magnetic field (t ₀₎ And performing a synchronization behavior evaluation at each reference time t _{0 of the} evaluation. The cross-correlation r _a (t ₀ ) related to the acceleration of claim 1, the cross-correlation r _ω (t ₀ ) related to the angular velocity of claim 2, and the cross-correlation r _θ (t ₀ ) related to the magnetic field of claim 3. Are further calculated at each evaluation reference time t ₀ , and the synchronization behavior is evaluated at each evaluation reference time t ₀ based on the sum or product of the cross-correlations. A method for detecting synchro-characteristic features. Means for reading, from a storage device, measurement value data of the acceleration vector for each synchronization behavior detection target obtained in time series by measurement using a sensor attached to the same part of each body for each synchronization behavior detection target; Equation 6 below

By using the means for converting the acceleration vector for each measurement time t into an acceleration scalar a (t) and the acceleration scalar a (t) for each synchronization action detection target,

It said means for calculating a synchronous action detection subject evaluation in pairs of reference time t ₀ for each of the cross-correlation relates to the acceleration at the same part of the body r _a (t _0), the acceleration correlation regarding r _a by (t ₀₎ And a means for performing an evaluation of the synchronization behavior at every reference time t ₀ of the evaluation based on the above. Means for reading, from a storage device, measurement value data of the angular velocity vector for each synchronization action detection target person acquired in time series by measurement using a sensor mounted on the same part of each body for each synchronization action detection target person; Equation 8 below

Using the means for converting the angular velocity vector for each measurement time t into an angular velocity scalar ω (t) and the angular velocity scalar ω (t) for each synchronization action detection target,

The means for calculating the cross-correlation r _omega (t ₀₎ relating to an angular velocity in the same area of the body of each reference time t ₀ of evaluation in the synchronous action detection subject pairs, the cross-correlation r _omega concerning the angular velocity (t ₀₎ And a means for performing an evaluation of the synchronization behavior at every reference time t ₀ of the evaluation based on the above. Means for reading, from a storage device, measurement value data of the magnetic field vector for each synchronization behavior detection target obtained in time series by measurement using a sensor attached to the same part of each body for each synchronization behavior detection target; Using the means for converting the magnetic field vector for each measurement time t into an azimuth angle value θ (t) and the azimuth angle value θ (t) for each person to be detected by the synchronization behavior, the following formula 10

The means for calculating the cross-correlation r _theta (t ₀₎ relating to the magnetic field at the same site of the body of each reference time t ₀ of evaluation in the synchronous action detection subject pairs, the cross-correlation r _theta about the magnetic field (t ₀₎ And a means for performing an evaluation of the synchronization behavior at every reference time t ₀ of the evaluation based on the above. The cross-correlation r _a (t ₀ ) related to the acceleration of claim 5, the cross-correlation r _ω (t ₀ ) related to the angular velocity of claim 6, and the cross-correlation r _θ (t ₀ ) related to the magnetic field of claim 7. Are further calculated at each evaluation reference time t ₀ , and the synchronization behavior is evaluated at each evaluation reference time t ₀ based on the sum or product of the cross-correlations. A device for detecting synchro behavior. Means for reading, from a storage device, measured value data of the acceleration vector for each synchronization behavior detection target acquired in time series by measurement using a sensor attached to the same part of each body for each synchronization behavior detection target; Equation 11

Using the means for converting the acceleration vector for each measurement time t into an acceleration scalar a (t), and the acceleration scalar a (t) for each synchronization behavior detection target,

It said means for computing a synchronous action detection evaluation in the subject's to reference time t ₀ for each of the cross-correlation relates to the acceleration at the same part of the body r _a (t _0), the cross-correlation r _a (t ₀₎ relating to the acceleration by A synchronization behavior detection program for causing a computer to function as means for performing synchronization behavior evaluation at each evaluation reference time t ₀ based on the evaluation time. Means for reading, from a storage device, angular velocity vector measurement value data for each synchronization behavior detection target person acquired in time series by measurement using a sensor attached to the same part of each body for each synchronization behavior detection target person; Equation 13

Using the means for converting the angular velocity vector for each measurement time t into an angular velocity scalar ω (t), and the angular velocity scalar ω (t) for each synchronization action detection target,

The synchronous action detection subject means for calculating the cross-correlation r _omega (t ₀₎ relating to the angular velocity at the same site of the reference time t ₀ for each of the body of the evaluation in pairs, the cross-correlation r _omega (t ₀₎ related to the angular velocity by A synchronization behavior detection program for causing a computer to function as means for performing synchronization behavior evaluation at each evaluation reference time t ₀ based on the evaluation time. Means for reading, from a storage device, measurement value data of the magnetic field vector for each synchronization behavior detection target obtained in time series by measurement using a sensor attached to the same part of each body for each synchronization behavior detection target Using the means for converting the magnetic field vector for each time t into the azimuth angle value θ (t), and the azimuth angle value θ (t) for each person to be detected by the synchro action, the following formula 15

The synchronous action detection subject means for calculating the cross-correlation r _theta (t ₀₎ relating to the magnetic field at the same site of the reference time t ₀ for each of the body of the evaluation in pairs, the cross-correlation r _theta (t ₀₎ related to the magnetic field by A synchronization behavior detection program for causing a computer to function as means for performing synchronization behavior evaluation at each evaluation reference time t ₀ based on the evaluation time. Among the cross-correlation r θ _{(t 0)} relating to the field of cross-correlation r ω _{(t 0)} and claim 11 relates to the angular velocity of the cross-correlation r _a (t ₀₎ and claim 10 relates to the acceleration of claim 9 Is further calculated at each evaluation reference time t ₀ , and the synchronization behavior is evaluated at each evaluation reference time t ₀ based on the cross-correlation sum or product. A detection program for synchro action to make a computer function.

Images