JP5522338B2 - Situation judging device, situation judging system, method and program thereof - Google Patents

Description

  The present invention relates to a situation determination apparatus, a situation determination system, a method thereof, and a program.

  A related situation determination system is a system for determining a situation of a detection target from sensor data, and is extracted with a data holding unit that stores sensor data, and a data extraction unit that extracts data of a certain length of time from the data holding unit. A situation determination unit that determines the situation in the section from the data and a determination result output unit that outputs the determination result. The situation determination unit determines the detection target situation by calculating a representative value representing the detection target situation from the extracted data and comparing the obtained representative value with a preset threshold value. Here, the status of the detection target is a status of a person such as “stop” or “walk” when the detection target is a person.

Further, Patent Document 1 describes an action determination system that detects a human body movement in real time using a sensor. In this system, since behavior determination is performed focusing on only the periodicity of body movement, a sudden change in body movement cannot be detected.
JP 2006-262773 A

  The problem with the technology described above is that, when trying to determine the current status of the detection target, if high accuracy is to be maintained, there will be a time delay in the output of the determination result with respect to a sudden change in the status. It was. On the other hand, if an attempt is made to output the determination result without time delay, the accuracy of the determination result will decrease.

  Therefore, the present invention has been invented in view of the above-described problems, and the purpose thereof is a situation determination device capable of performing situation determination with high accuracy and performing determination without impairing time responsiveness, To provide a situation determination system, a method thereof, and a program.

  The present invention that solves the above-mentioned problem is based on the data of the data section of the predetermined time length of the detection target, determines the continuity of the status of the detection target in the data section of the predetermined time length, and A detection target continuity determination unit that determines a data section used for determination of the status of the detection target based on continuity, and a detection target status that determines the status of the detection target based on data in the determined data section A situation determination apparatus having a determination unit.

  The present invention for solving the above-mentioned problems is a sensor whose output changes depending on the situation, a data holding unit that acquires the output of the sensor, stores the acquired data in association with time, and is stored in the data holding unit. A data extraction unit that extracts data of a predetermined data section of a predetermined time length from the data, and the data in the data section of the predetermined time length based on the extracted data of the data section of the predetermined time length A detection target continuity determination unit that determines the continuity of the detection target situation and determines a data section used for the detection target situation determination based on the continuity of the detection target situation; and the determined data section A detection target status determination unit that determines the detection target status based on the data of the detection target, and an output unit that outputs the determined status of the detection target. It is a system.

  The present invention that solves the above-mentioned problem is based on the data of the data section of the predetermined time length of the detection target, determines the continuity of the status of the detection target in the data section of the predetermined time length, and In this situation determination method, a data section to be used for determining the status of the detection target is determined based on continuity, and the status of the detection target is determined based on the data in the determined data section.

  The present invention that solves the above-described problem is based on the data of the data section of the predetermined time length of the detection target, the continuity determination process that determines the continuity of the status of the detection target in the data section of the predetermined time length, Based on the continuity of the status of the detection target, a data interval determination process for determining a data interval to be used for determining the status of the detection target, and determining the status of the detection target based on the data of the determined data interval This is a program for causing an information processing apparatus to execute situation determination processing.

  The present invention can determine a situation with high accuracy and can make a determination without impairing time response.

  An embodiment of the present invention will be described.

  When determining the status of a detection target at a certain time, the status is determined using data in a certain section from a certain point in the past to the time at which the determination is to be made. Generally, if the situation is stationary in the data section used for the determination, the longer the data section, the higher the correct answer rate of the determination. On the other hand, when the situation changes in the data section used for judgment, the judgment result is affected by the data before the situation change as the data section is longer, and the current situation cannot be judged correctly. . The state of the detection target is, for example, an operation such as “stop” or “walking” when the detection target is a person, or environmental information such as temperature and humidity when the detection target is an object.

  In the present embodiment, in the data section of a predetermined time length obtained from the detection target, the continuity of the situation of the detection target is determined, and the data section of an appropriate length according to the continuity of the detection target situation is used. Determine the status of the detection target. Specifically, when the status of the detection target is steady, the situation is determined using a long data section, and when the situation changes, the length of the data section is determined so that the situation is determined using a short data section. To control. By doing in this way, situation determination can be performed with high accuracy, and determination can be performed without impairing time responsiveness.

  Hereinafter, specific embodiments will be described.

<First Embodiment>
Next, a first embodiment of the present invention will be described in detail with reference to the drawings.

  Referring to FIG. 1, the best mode for carrying out the first invention of the present invention includes a sensor 1, a situation determination device 2, and a determination result output device 3.

  The situation determination device 2 includes a data acquisition / update unit 21, a data holding unit 22, a data extraction unit 23, a continuity determination index calculation unit 24, a continuity determination unit 25, a time window control unit 26, a situation A determination index calculation unit 27 and a situation determination unit 28 are provided.

  For example, when the detection target is a person, the sensor 1 is a temperature that is attached to a mechanical component when the detection target is a machine, such as an acceleration sensor or a heart rate sensor carried by a person, or a human sensor arranged in the environment. A device such as a sensor or a contact sensor whose output changes depending on the condition of a detection target.

  The data acquisition / update unit 21 acquires the sensor output that changes with time, and transmits the sensor output to the data holding unit 22 together with the time information of the measured sensor output.

  The data holding unit 22 stores the sensor output and time information in association with each other.

  The data extraction unit 23 reads a data section having a preset length from the data holding unit 22 and supplies it to the continuity determination index calculation unit 24 and the time window control unit 26.

  The continuity determination index calculation unit 24 detects, for example, a variance value, an average value, a maximum value, a minimum value, an amplitude, a kurtosis, a skewness, and an FFT power spectrum from the data section supplied from the data extraction unit 23. A stationarity determination index that is an index for determining the stationarity of the target situation is calculated.

  The stationarity determination unit 25 determines the stationarity of the status of the detection target in the data section from the stationarity determination index given from the stationarity determination index calculation unit 24, and is used for determining the status of the detection target. Determine the data interval. For example, when an FFT power spectrum is used as a steadiness determination index, an appropriate data section used for determination of a situation to be detected is determined, such as a section including n frequency component waves of the strongest power spectrum. .

  The time window control unit 26 extracts the data set of the data section used for the detection target situation determination from the data extraction unit 23 according to the result given from the continuity determination unit 25.

  The situation determination index calculation unit 27 uses the data set given from the time window control unit 26 to calculate a situation determination index indicating the degree of the type of situation to be detected. For example, by calculating indices such as the variance value, average value, maximum value, minimum value, amplitude, kurtosis, skewness, FFT power spectrum of the data set given from the time window control unit 26, the type of the situation is changed. The situation determination index to be expressed is calculated.

  When the detection target is a person, for example, by comparing the magnitude relationship with a preset threshold value from the situation determination index given from the situation determination index calculation unit 27, for example, “ Get judgment results such as “stop” and “walk”.

  The determination result output device 3 includes, for example, a display device that displays as character information or image information, a communication device that transmits to other display devices or recording devices, and the like, and outputs the determination result obtained from the situation determination unit 28. .

  Next, the operation of the first embodiment will be described in detail with reference to FIG. 1 and FIG.

  The sensor output that changes over time is transmitted to the data holding unit 22 by the data acquisition / update unit 21 together with the time information at the time of acquisition and stored (step A2).

  From the sensor data stored in the data holding unit 22, a sufficiently long data section set in advance is read by the data extracting unit 23 and supplied to the continuity determination index calculating unit 24 (step A3). For example, if the preset data interval is 1 second, the newest data interval for 1 second is read from the data holding unit 22 and supplied to the continuity determination index calculation unit 24.

  The continuity determination index calculation unit 24 checks whether there is a change in the situation in the data section supplied from the data extraction unit 23 (step A4). Therefore, from the data section supplied from the data extraction unit 23, for example, variance, average value, maximum value, minimum value, amplitude, kurtosis, skewness, FFT power spectrum, etc., the continuity of the situation in the data section Is calculated, and the result is given to the continuity determination unit 25.

  The stationarity determination unit 25 compares, for example, the magnitude relationship between a preset threshold and the obtained stationarity determination index in accordance with the stationarity determination index obtained from the stationarity determination index calculation unit 24. A data section suitable for use in situation determination is determined, and the result is given to the time window control unit 26 (step A5). When it is determined from the stationarity determination index obtained from the stationarity determination index calculation unit 24 that the situation is stationary in the data section, the data section length that is set in advance in the data extraction unit 23, for example, 1 second To the time window control unit 26 as it is. However, if it is determined that a change has occurred in the situation, an appropriate data section length to be used for the situation determination of the detection target is reset, for example, a section of 0.5 second in the second half as a half section of one second. The determined data section length is given to the time window control unit 26.

  The time window control unit 26 extracts a data set of the data section based on the data section length newly set by the continuity determination unit 25 from the data section extracted by the data extraction unit 23, and sends it to the situation determination index calculation unit 27. Supply (step A6).

  The situation determination index calculation unit 27 uses a current set such as a variance value, an average value, a maximum value, a minimum value, an amplitude, a kurtosis, a skewness, and an FFT power spectrum from the data set supplied from the time window control unit 26. The situation determination index for determining the type of the situation of the detection target in is calculated, and the result is given to the situation determination unit.

  The situation determination unit 28 compares the magnitude relationship between the situation determination index obtained from the situation determination index calculation unit 27 and, for example, a preset threshold value. Depending on the result, for example, the detection target is human. In some cases, information on a specific situation such as “run” or “walk” is output and provided to the determination result output device 3 (step A7).

  The determination result output device 3 displays the result obtained by the situation determination unit 28 as, for example, character information or image information, or transmits the result to another display device or recording device.

  In the present embodiment, a continuity determination index for determining the continuity of the situation is calculated, the changed situation is sufficiently included according to the calculated continuity determination index, and high determination accuracy can be ensured. Set the data interval appropriately. For this reason, it is possible to improve the time delay of the determination output without reducing the correct answer rate of the situation determination.

  Example 1 will be described in detail with reference to the drawings. Example 1 corresponds to the first embodiment.

  Assume that the detection target is a person and the determination situation is the action of the detection target person, and the person's “stop” action and “walk” action are distinguished and determined. As an example of the sensor 1, the person to be detected wears a mobile phone with a built-in triaxial acceleration sensor, and acceleration sensor data is stored in a storage device inside the mobile phone. Using this data, the situation determination device 2 mounted on the mobile phone determines the behavior of the mobile phone wearer and outputs the result to the display of the mobile phone (corresponding to the determination result output device 3).

  An example of acceleration sensor data stored in the data holding unit of the cellular phone is shown in FIG. In FIG. 3, acceleration sensor data for each of the three axes and time information are stored in association with each other. Here, the three acceleration axes correspond to the front / rear, left / right and up / down directions of the person's body.

  It is assumed that the action of this person has changed from “stop” to “walk”. Assume that the acceleration data graph obtained at this time is as shown in FIG. This person's behavior changes from “stop” to “walk” between time 7 seconds and time 8 seconds.

  Now, the operation of the situation determination device 4 at time 8 seconds shown in the graph of FIG. 4 will be described.

  At present, in order to distinguish between “stop” and “walk”, attention is paid only to the acceleration in the vertical axis direction of the person's body in the acceleration data.

  In the data extraction unit, for example, from a measurement result of acceleration of walking behavior collected in advance, a data interval of 1 second is determined in advance as a time length including at least one step in the data interval, and the data holding unit 23, the vertical acceleration data from the time 7 seconds to the time 8 seconds in FIG. 4 is extracted and supplied to the continuity determination index calculation unit 24.

  The stationarity determination index calculation unit 24 calculates a stationarity determination index for determining the stationarity of the situation from the supplied data section. First, the average value of acceleration over the entire data section is calculated. Assume that this value is 1 [G], for example. Next, an average value of deviation from the overall average value of the acceleration for the first half of 0.5 second and an average value of deviation from the overall average value of the acceleration data for the second half of 0.5 second are calculated. As a result, for example, it is assumed that the average value of the deviation in the first half is 0.03 [G] and the average value of the deviation in the second half is 0.13 [G]. These two values are given to the continuity determination unit 25.

  In the continuity determination unit 25, for example, a threshold value used for continuity determination and a data interval length to be selected when it is determined that the situation is not stationary are set in advance based on measurement results collected in advance. . In this example, the threshold used for the continuity determination set in the continuity determination unit 25 is 0.1 [G]. Then, the continuity determination unit 25 determines that the situation is not steady if the absolute value of the difference between the two average values given from the continuity determination index calculation unit 24 is 0.1 [G] or more. In addition, the data section length to be selected when it is determined that the situation is not steady is determined from the last part of the data section for one second, and the data section length is determined when the data temporarily stops during walking. The length of the data section when the absolute value of the difference between the average values is 0.1 [G] is 0.5 seconds with reference to 0.5 seconds considered as the shortest time required for almost complete stop. And Each time the absolute value increases by 0.01 [G], a value obtained by subtracting 0.05 seconds from the standard 0.5 seconds is used as the data section length, and the minimum data section length is 0.3 seconds. To do. That is, when the absolute value is 0.11 [G], the data interval length is 0.45 seconds, and when the absolute value is 0.12 [G], the data interval length is 0.4 seconds. When the absolute value is 0.13 [G], the data interval length is 0.35 seconds, and when the absolute value is 0.14 [G], the data interval length is 0.3 seconds. is there. When the absolute value is 0.14 [G] or more, the data interval length is 0.3 seconds.

  As described above, the continuity determination unit 25 in which the threshold value and the data section length are set determines the absolute value of the difference between the two average values given from the continuity determination index calculation unit 24. Since the given two average values are 0.03 [G] and 0.13 [G], the absolute value of the difference between the two average values is 0.1 [G]. As a result, the continuity determination unit 25 determines that a change in behavior has occurred within the 1 second data interval because the absolute value 0.1 [G] of the difference between the two average values is greater than or equal to the threshold value. Then, as the data section length, 0.5 seconds that is the data section length when the absolute value of the difference between the two average values is 0.1 [G] is selected, and the last data section for one second is selected. A data interval of 0.5 seconds from the portion is set as an interval used for situation determination, and the value is given to the time window control unit 26.

  First, the time window control unit 26 obtains a data interval of 1 second from the data extraction unit 23. Next, the continuity determination unit 25 obtains a value that the data section to be used for the situation determination is the second half 0.5 seconds. Therefore, the time window control unit 26 supplies data in the data section for the second half of 0.5 seconds to the situation determination index calculation unit 27 excluding data for the first half of 0.5 seconds that is not used for situation determination.

The situation determination index calculation unit 27 calculates a variance value (situation determination index) as an index for determining the type of situation from the data in the data section obtained from the time window control unit 26. FIG. 5 shows a graph in which the dispersion value is calculated every second using the data of FIG. 4, and the horizontal axis represents time and the vertical axis represents the dispersion value in logarithm. Now, it is assumed that the variance value obtained from the data in the supplied data section is 0.002 [G ² ]. From the result of the graph of FIG. 5 obtained in advance, the situation determination unit 28 is set to 0.001 [G ² ] as a threshold value for determining “stop” and “walk” in advance. When the variance value is equal to or greater than this value, the person's action is determined to be “walking”, and when the variance value is less than this value, it is determined to be “stop”. Now, since the dispersion value of 0.002 [G ² ] obtained from the situation determination index calculation unit 27 is larger than the threshold value 0.001 [G ² ], it is assumed that the action of this person at the present time is “walking”. The determination is made, and the determination result is given to the determination result output device 3.

  On the display of the mobile phone which is the determination result output device 3, the determination result obtained from the situation determination unit is displayed as a character, the determination result “walking”.

  Further, when the action of a person is as shown in the graph of FIG. 4, the action of the person changes from “walking” to “running” between time 13 seconds and time 14 seconds.

  At this time, as in the above-described operation, the continuity determination index calculation unit 24 calculates a continuity determination index for determining the continuity of the situation from the supplied data section. First, the average value of acceleration over the entire data section is calculated. It is assumed that this value is, for example, 0.99 [G]. Next, an average value of deviation from the overall average value of the acceleration for the first half of 0.5 seconds and an average value of deviation from the overall average value of the acceleration for the second half of 0.5 seconds are calculated. As a result, for example, it is assumed that the average value of the deviation in the first half portion is 0.08 [G] and the average value of the deviation in the second half portion is -0.04 [G]. These two values are given to the continuity determination unit 25.

  The continuity determination unit 25 obtains the absolute value of the difference between the two values given from the continuity determination index calculation unit 24. Now, this value is 0.12 [G]. Since the absolute value of the difference between the two average values is 0.1 [G] or more, the continuity determination unit 25 determines that the situation is not stationary.

  Furthermore, since the difference between the two average values is 0.12 [G], the stationarity determination unit 25 selects 0.4 seconds as the data section length, and starts from the last part of the data section for one second. A data interval of 0.4 seconds is set as an interval used for situation determination, and the value is given to the time window control unit 26.

  The time window control unit 26 obtains a data interval of 1 second from the data extraction unit 23. Next, the continuity determination unit 25 obtains a value that the data section to be used for the situation determination is the second half 0.4 seconds. Therefore, the time window control unit 26 supplies the data of the data section of the second half 0.4 seconds to the situation determination index calculation unit 27 excluding the data of the first 0.6 seconds not used for the situation determination.

The situation determination index calculation unit 27 calculates a variance value (situation determination index) as an index for determining the type of situation from the data in the data section obtained from the time window control unit 26. It is assumed that the variance value obtained from the data in the supplied data section is 0.19 [G ² ]. From the result of the graph of FIG. 5 obtained in advance, the situation determination unit 28 is set with 0.1 [G ² ] as a threshold for determining “walking” and “running” in advance. When the variance value is greater than or equal to this value, the person's action is determined to be “running”, and when the variance value is less than this value, it is determined to be “walking”. Now, since the variance value of 0.19 [G ² ] obtained from the situation determination index calculation unit 27 is larger than the threshold value 0.1 [G ² ], it is assumed that this person's action is “run” at the present time. The determination is made, and the determination result is given to the determination result output device 3.

  On the display of the mobile phone that is the determination result output device 3, the determination result obtained from the situation determination unit is displayed as a character, the determination result “run”.

  The above operation is repeated at a predetermined cycle such as twice per second, and the determination result is sequentially displayed on the mobile phone display.

<Second Embodiment>
The second embodiment will be described in detail with reference to FIG.

  Referring to FIG. 1 and FIG. 6, the second embodiment uses the index calculated by the stationarity determination index calculator as the stationarity of the situation as compared to the first embodiment shown in FIG. It is different in that it is used not only for the determination of the situation but also for the determination of the type of situation.

  In the second embodiment, the continuity determination index is an index for determining the continuity of the situation and also an index for determining the type of the situation.

  The situation determination unit 281 is given a continuity determination index from the continuity determination index calculation unit 241 when the continuity determination unit 25 determines that the situation is stationary. The situation determination unit 281 compares the magnitude of the index with, for example, a preset threshold value, and when the detection target is a human according to the result, “running”, “walking”, etc. Is output to the determination result output device 3.

  In the second embodiment, the situation determination index calculation unit 271 and the situation determination unit 282 are the same as the situation determination index calculation unit 27 and the situation determination unit 28 in the first embodiment. Do not reset the data section used for judgment. That is, when the continuity determination unit 25 determines that the situation is steady, no processing is performed.

  Next, the operation of the best mode for carrying out the second invention of the present invention will be described in detail with reference to FIGS.

  The operations of the sensor 1, the data acquisition / update unit 21, the data holding unit 22, the data extraction unit 23, and the continuity determination unit 25 in the present embodiment indicated by steps A1 to A4 in FIG. Since the operations of the sensor 1, the data acquisition / update unit 21, the data holding unit 22, the data extraction unit 23, and the continuity determination unit 25 are the same, the description thereof will be omitted.

  In the first embodiment, the continuity determination unit 25 determines an appropriate data section according to the continuity determination index calculated by the continuity determination index calculation unit 24. In the second embodiment, the continuity determination index calculated by the continuity determination index calculation unit 241 is an index for determining the continuity of the situation and an index for determining the type of the situation. Choose a representative value.

  At this time, if the continuity determination unit 25 determines that the situation is stationary in the data section supplied from the data extraction unit and that no change has occurred, the index calculated by the continuity determination index calculation unit 241 To the situation determination unit 281 (step B1), and the situation determination can be performed directly (step A72).

  When the continuity determination unit 25 determines that the situation is not stationary, the time window control unit 26 extracts an appropriate data section (step A6), as in the first embodiment, and the situation determination index. The situation determination unit 282 performs situation determination according to the continuity determination index calculated by the calculation unit 271 (A71), and the determination result output device 3 displays the result (step A8).

  In the second embodiment, in addition to the effects of the first embodiment, when it is determined that the situation is steady, the time window setting (step A5) and the data extraction for the time window (step A6) ) And the amount of calculation and power consumption can be reduced as compared with the first embodiment.

  Example 2 will be described in detail with reference to the drawings. This Example 2 corresponds to the second embodiment.

As in the first embodiment, consider a case where “stop” and “walk” are determined. In the second embodiment, the stationarity determination index calculation unit 241 calculates a variance value of the entire data in the data section as an index for determining the stationarity of the situation. Now, it is assumed that the action of this person has “stopped” for a sufficiently long time, and the continuity determination index of the action of this person at that time, that is, the value of the variance value of the entire data section is 0.0001 [G ² ]. Suppose. For example, from the data as shown in FIG. 5, the continuity determination unit 25 preliminarily sets 0.001 [a value that is 10 times or more larger than the average dispersion value when both feet are on the ground. A threshold value G ² ] is set. The stationarity determination unit 25 determines that the situation is not stationary in this data section when the variance value obtained from the stationarity determination index calculation unit is equal to or greater than this value, and when the variance value is less than this value, the situation is stationary. Judge that there is.

Now, since the variance value of 0.0001 [G ² ] obtained from the stationarity determination index calculation unit 241 is smaller than the threshold value 0.001 [G ² ], the stationarity determination unit 25 uses this person within this data section. Is determined to be stationary (step B1), and the variance value calculated by the continuity determination index calculation unit 241 is directly given to the situation determination unit 281.

The situation determination unit 281 uses, for example, 0.0007 [G ² ] in advance as a value obtained by, for example, averaging the maximum value of acceleration dispersion values when a person is stopped from data as illustrated in FIG. Value is set. Then, the situation determination unit 281 determines that the person's action is “walking” in the data section when the given index is equal to or greater than the value, and determines that the action is “stop” when the index is less than this value.

Now, since the variance value of 0.0001 [G ² ] obtained from the stationarity determination index calculation unit 241 is smaller than the threshold value 0.0007 [G ² ], the situation determination unit 281 has the data value in this data section. The person's action is determined to be “stop” (step B2), and the result is given to the determination result output device 3.

On the other hand, when the variance value obtained from the stationarity determination index calculation unit 241 is larger than the threshold value 0.001 [G ² ], the stationarity determination unit 25 determines that the action of this person is not stationary in this data section. Then, the same operation as in the first embodiment is performed.

<Third Embodiment>
A third embodiment will be described in detail with reference to FIG.

  Referring to FIG. 1 and FIG. 8, the third embodiment is different from the first embodiment shown in FIG. Holding unit 22, data extraction unit 23, continuity determination index calculation unit 24, time window control unit 26, continuity determination unit 251, situation determination index selection / calculation unit 272, threshold selection / situation determination unit 283 It differs in having a situation determination device 5 provided with.

  Unlike the continuity determination unit 25 in the first embodiment shown in FIG. 1, the continuity determination unit 251 uses the reset data interval length to select a situation determination index / in addition to the time window control unit 26. The calculation unit 272 and the threshold selection / situation determination unit 283 are also provided.

  The situation determination index selection / calculation unit 272 is based on the data section length obtained from the stationarity determination unit 251 and, for example, an appropriate calculation method from different situation determination index calculation methods such as a variance value and an FFT power spectrum. Is selected, and the situation determination index is calculated. Similarly, the threshold selection / situation determination unit 283 appropriately selects a threshold value used for situation determination based on the obtained data section length, and performs situation determination.

  Next, the operation of the third embodiment will be described in detail with reference to FIGS.

  The sensor 1, the data acquisition / update unit 21, the data holding unit 22, the data extraction unit 23, the continuity determination index calculation unit 24, and the determination in the present embodiment indicated by steps A1 to A5, A6, and A8 in FIG. Since the operation of the result output device 3 is the same as the operation of the sensor 1, the data acquisition / update unit 21, the data holding unit 22, and the data extraction unit 23 of the first embodiment, the description thereof is omitted.

  The data section length appropriate for the situation determination determined by the continuity determination section 251 is given to the time window control section 26, the situation determination index selection / calculation section 272, and the threshold selection / situation determination section 283. From the given data section length, the situation determination index selection / calculation unit selects a situation determination index calculation method suitable for determination. The threshold selection / situation determination unit 283 selects a determination method and a determination threshold according to the calculation method of the selected situation determination index (step C1). Then, from the data set supplied from the time window control unit 26, the situation determination index selection / calculation unit 272 calculates the selected situation determination index, gives the result to the threshold selection / situation determination unit 283, and the threshold selection / situation determination unit In step 283, the state is determined according to the selected determination method and determination threshold (step A73).

  In the third embodiment, in addition to the effects described above, the type of situation determination index, the determination method, and the threshold value are changed according to the data section length set by the continuity determination unit 251. More types of situations than forms can be determined.

  Example 3 will be described in detail with reference to the drawings. Example 3 corresponds to the third embodiment.

  Consider a case where “walking” and “running” are determined, and it is assumed that the action of the person to be measured is “running” for a sufficiently long time.

  The stationarity determination unit 251 determines that the situation is stationary in the data section of 1 second, and sets the data section length to 1 second. The information is given to the time window control unit 26, the situation determination index selection / calculation unit 272, and the threshold selection / situation determination unit 283.

  When the data section length is 0.5 seconds or more, the situation determination index selection / calculation unit 272 calculates the power spectrum of the data section from the FFT and obtains the frequency indicating the largest power spectrum at 1 Hz or more. A threshold is set in advance so that is used as a situation determination index. In addition, when the data section length is less than 0.5 seconds, the variance value of the data section is obtained and set as a situation determination index. That is, the situation determination index selection / calculation unit 272 has two methods for calculating the situation determination index. Then, the situation determination index selection / calculation unit 272 is configured to select one of the calculation methods based on the data section length, and to calculate the situation determination index using the selected calculation method.

The threshold selection / situation determination unit 283 also includes an FFT power spectrum of acceleration data collected in advance when the data section length is 0.5 seconds or longer, for example, when a person is walking and running. The frequency at which the maximum value is taken is determined, and when the given frequency is 2.5 Hz or higher as an intermediate frequency between both obtained frequencies, the current human action is determined to be “run” In addition, a threshold is set. In addition, when the data section length is less than 0.5 seconds, the variance values of the acceleration data collected in advance when the person is walking and running are respectively obtained, and the variance values of both obtained are obtained. As an intermediate value or the like, when a given variance value is 0.1 [G ² ], it is set to determine that the current human action is “run”. That is, the threshold selection / situation determination unit 283 has determination criteria corresponding to the calculation method described above. The threshold selection / situation determination unit 283 is configured to select a determination criterion corresponding to the calculation method selected by the situation determination index selection / calculation unit 272 and determine the detection target situation.

  The time window control unit 26 extracts a data set for one second from the data extraction unit 23 according to the same standard as that shown in the first embodiment, and supplies this to the situation determination index selection / calculation unit 272. The situation determination index selection / calculation unit 272 calculates a frequency indicating the largest power spectrum, which is the selected situation determination index, and this frequency is now 3 Hz. Since the value of 3 Hz obtained from the situation determination index selection / calculation unit 272 is larger than the threshold value 2.5 Hz, the threshold selection / situation determination unit 272 determines that this person's action at the present time is “running”, and the result Is given to the determination result output device 3.

<Fourth embodiment>
A fourth embodiment will be described in detail with reference to the drawings.

  Referring to FIG. 10, the fourth embodiment includes a sensor 1, a situation determination device 6, and a determination result output device 3, as in the first, second, and third embodiments.

  The situation determination device 6 controls the data acquisition / update unit 21, the data holding unit 22, the data extraction unit 23, and the continuity determination index calculation in the first, second, and third embodiments under the control of the situation determination program 7. The same processing as the processing by the unit 24, the continuity determination unit 25, the time window control unit 26, the situation determination index calculation unit 27, the situation determination unit 28, and the like is executed.

<Fifth embodiment>
The fifth embodiment will be described in detail with reference to FIG. Compared with the structure in 1st Embodiment, 5th Embodiment is the mobile telephone terminal 8 provided with the sensor 1, the data acquisition update part 21, and the determination result output device 3, the data holding part 22, and the data extraction part. 23, and the situation determination server 9 including the stationarity determination index calculation unit 24, the stationarity determination unit 25, the time window control unit 26, the situation determination index calculation unit 27, and the situation determination unit 28 is different.

  Consider a case where the target person is wearing a mobile phone with a built-in triaxial acceleration sensor. The acceleration sensor data generated by the action of the person is acquired by the data acquisition / update unit 21 inside the mobile phone, and then transmitted to the situation determination server 9 using the public mobile phone network and stored in the data holding unit 22. The

  The processing in the data holding unit 22, the data extraction unit 23, the continuity determination index calculation unit 24, the continuity determination unit 25, the time window control unit 26, the situation determination index calculation unit 27, and the situation determination unit 28 in the situation determination server 9 is as follows. Since it is the same as the process in 1st Embodiment, description is abbreviate | omitted.

  The situation determination result obtained by the situation determination server 9 is transmitted again to the mobile phone terminal 8 carried by the person to be measured using the public mobile phone network, and is displayed on the display screen of the mobile phone as the determination result output device 3. The determination result is output.

  According to the present invention, for example, an action at the time of going out can be acquired and information corresponding to the action of the person can be distributed at an appropriate timing. Moreover, it is applicable also to uses, such as a monitoring system of an elderly person or an infant.

FIG. 1 is a block diagram showing the configuration of the first embodiment. FIG. 2 is a flowchart showing the operation of the first embodiment. FIG. 3 is a diagram illustrating an example of sensor data stored in the data holding unit 22. FIG. 4 is a graph showing an example of acceleration sensor data acquired from human behavior. FIG. 5 is a logarithmic representation of an example of vertical acceleration sensor data acquired from a human action. FIG. 6 is a block diagram showing the configuration of the second embodiment. FIG. 7 is a flowchart showing the operation of the second embodiment. FIG. 8 is a block diagram showing the configuration of the third embodiment. FIG. 9 is a flowchart showing the operation of the third embodiment. FIG. 10 is a block diagram showing the configuration of the fourth embodiment. FIG. 11 is a block diagram showing the configuration of the fifth embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Sensor 2 Situation determination apparatus 3 Judgment result output apparatus 4 Situation determination apparatus 5 Situation determination apparatus 6 Situation determination apparatus 7 Situation determination program 8 Mobile phone terminal 9 Situation determination server 21 Data acquisition / update part 22 Data holding part 23 Data extraction part 24 Steadyness determination index calculation unit 25 Steadyness determination unit 26 Time window control unit 27 Situation determination index calculation unit 28 Situation determination unit 241 Steadyness determination index calculation unit 251 Steadyness determination unit 271 Situation determination index calculation unit 272 Situation determination index selection / Calculation unit 281 Situation determination unit 282 Situation determination unit 283 Threshold selection / Situation determination unit

Claims (13)

Based on the data of the data section of the predetermined time length of the detection target, determine the continuity of the status of the detection target in the data section of the predetermined time length, and based on the continuity of the status of the detection target, A detection target stationarity determining unit that determines a data section used for the situation determination of
A detection object situation determination unit that determines the situation of the detection object based on the data of the determined data section;
Have
When the detection target continuity determination unit determines that there is no change in the status of the detection target, the status determination apparatus determines the data section having the predetermined time length as a data section used for the status determination of the detection target .   The situation determination apparatus according to claim 1, wherein the detection target situation determination unit determines the situation of the detection target based on a situation determination criterion corresponding to the determined data section length.   A sensor whose output changes depending on the situation,
  A data holding unit for acquiring the output of the sensor and storing the acquired data in association with time;
  A data extraction unit that extracts data of a data section having a predetermined time length set in advance from data stored in the data holding unit;
  Based on the data of the extracted data section of the predetermined time length, the continuity of the status of the detection target in the data section of the predetermined time length is determined, and based on the continuity of the status of the detection target, A detection target continuity determination unit that determines a data section used for determination of the status of the detection target;
  Based on the data of the determined data section, a detection target status determination unit that determines the status of the detection target;
  An output unit for outputting the determined status of the detection target;
Have
  When the detection target continuity determination unit determines that there is no change in the status of the detection target, the detection target continuity determination unit determines the data interval of the predetermined time length as a data interval used for the status determination of the detection target
Situation judgment system.   The situation determination system according to claim 3, wherein the detection target situation determination unit determines the situation of the detection target based on a situation determination criterion according to the determined data section length.   The detection target continuity determination unit
  Based on the data extracted by the data extraction unit, a continuity determination index calculation unit that calculates a continuity determination index that is an index for determining the continuity of the status of the detection target in the data section;
  Based on the stationarity determination index calculated by the stationarity determination index calculation unit, the stationarity of determining the stationarity of the status of the detection target in the data section and determining the data section used for the status determination of the detection target Judgment department and
Have
  The detection target situation determination unit
  A situation determination index calculation unit that calculates a situation determination index that is an index for determining the situation of the detection target, from the data of the data section determined by the stationarity determination unit;
  A status determination unit that determines the status of the detection target based on the status determination index calculated by the status determination index calculation unit;
The situation determination system according to claim 3.   The situation determination unit determines the status of the detection target from the continuity determination index calculated by the continuity determination index calculation unit when the continuity determination unit determines that there is no change in the detection target status. The situation determination system according to claim 5.   The situation determination index calculation unit prepares at least one or more situation determination index calculation methods, selects and selects the situation determination index calculation method based on the data section length determined by the stationarity determination unit Based on the calculated calculation method, the situation determination index is calculated from the data of the data section,
  The situation determination unit prepares a determination criterion corresponding to the calculation method of the at least one situation determination index, selects a determination criterion corresponding to the calculation method selected by the situation determination index calculation unit, and selects the selected determination criterion The situation determination system according to claim 5 or 6, wherein the situation to be detected is determined based on the situation determination index calculated by the situation determination index calculation unit.   The stationarity determination index calculation unit uses at least one of the variance value, average value, maximum value, minimum value, amplitude, kurtosis, skewness, FFT power spectrum of the data extracted by the data extraction unit, The situation determination system according to claim 5, wherein the continuity determination index is calculated.   The situation determination index calculation unit obtains at least one of data dispersion value, average value, maximum value, minimum value, amplitude, kurtosis, skewness, FFT power spectrum of the data section determined by the continuity determination unit. The situation determination system according to claim 5, wherein the situation determination index is used to calculate the situation determination index.   Based on the data of the data section of the predetermined time length of the detection target, determine the continuity of the situation of the detection target in the data section of the predetermined time length,
  When it is determined that there is no change in the status of the detection target, the data interval of the predetermined time length is determined as a data interval used for the status determination of the detection target,
  A situation determination method for determining a situation of the detection target based on the data of the determined data section.   The situation determination method according to claim 10, wherein the situation of the detection target is judged based on a situation judgment criterion corresponding to the determined data section.   Continuity determination processing for determining continuity of the status of the detection target in the data section of the predetermined time length based on data of the data section of the predetermined time length of the detection target;
  In the continuity determination process, when it is determined that there is no change in the status of the detection target, a data interval determination process that determines the data interval of the predetermined time length as a data interval used for the status determination of the detection target;
  A status determination process for determining the status of the detection target based on the data of the determined data section;
For causing an information processing apparatus to execute the program.   The program according to claim 12, wherein the situation determination process determines the situation of the detection target based on a situation determination criterion corresponding to the determined data section.

Images