JP2020058277A - Detection device, detection method and program - Google Patents

Abstract

To provide a delivery detection system which can detect the delivery of a cow anteriorly or posteriorly in the early stage.SOLUTION: A detection device 10 for detecting the delivery of livestock includes: storage means which stores acceleration data measured by a triaxial acceleration sensor included in a tag 20 attached to the livestock; acquisition means which acquires from the storage means a component value in a prescribed direction included in each of the plurality of pieces of acceleration data during the prescribed time; and detection means which detects the delivery of the livestock on the basis of the plurality of component values.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a detection device, a detection method and a program.

  It is one of the important tasks for a livestock farmer who raises cows to control the time of calving. This is because accidents such as death of mother cows and calves are likely to occur during delivery, and therefore delivery assistance may be necessary. For this reason, livestock farmers check the signs of calving by, for example, looking around the barn.

JP, 2010-227161, A JP, 2007-296312, A Japanese Patent No. 3938786 JP, 2005-110880, A JP, 2003-325077, A

Abe Ryo, "Agriculture Basic Seminar: Basics of Livestock Breeding", New Edition, Agricultural and Fisheries Village Cultural Association, April 2008, p.109, p.122-124.

  However, a large-scale livestock farmer who breeds several tens to several hundreds of cattle requires a lot of work to confirm signs of calving by looking around the barn. Therefore, there is a demand for a method capable of detecting calving in advance.

  Further, even when the signs of calving of the cow can be confirmed, it is difficult to accurately grasp the calving time, and therefore the calving of the cow may be overlooked. Therefore, for example, an accident such as death of a calf may occur. Therefore, there is also a demand for a method capable of detecting the calving early after the fact even when the calving is overlooked.

  The embodiment of the present invention has been made in view of the above points, and an object thereof is to detect calving of a cow.

  In order to achieve the above object, in the embodiment of the present invention, a detection device for detecting the delivery of livestock, a storage means for storing acceleration data measured by a triaxial acceleration sensor mounted on the livestock, and a predetermined Acquisition means for acquiring a component value in a predetermined direction contained in each of a plurality of acceleration data during the period of time from the storage means, and a detection means for detecting the delivery of the livestock based on the plurality of component values, It is characterized by having.

  According to the embodiment of the present invention, calving of a cow can be detected.

It is a figure which shows an example of the whole structure of the delivery detection system which concerns on this embodiment. It is a figure which shows an example of the measurement data memorize | stored in the measurement data memory | storage part. It is a figure which shows an example of the amount data of activity memorize | stored in the amount data storage part of activity. It is a figure showing an example of functional composition of the 1st labor detection processing part. It is a figure which shows an example of a function structure of the 2nd delivery detection processing part. It is a flow chart which shows an example of the 1st labor detection processing. It is a figure for explaining an example of a peak in the 1st labor detection processing. It is a figure for explaining an example of the 1st labor detection. It is a flow chart which shows an example of the 2nd labor detection processing. It is a figure for explaining an example of a peak in the 2nd labor detection processing.

  Hereinafter, embodiments of the present invention (hereinafter, also referred to as “this embodiment”) will be described with reference to the drawings. In this embodiment, as an example of livestock, a calving detection system 1 that detects calving in advance or after is explained. In particular, in the case of detecting the calving of a cow in advance, a cow (eg, a single-bred cow) whose feeding is regularly performed with a certain degree of accuracy is targeted. Livestock is not limited to cows.

<Overall configuration of labor detection system 1>
First, the overall configuration of the labor detection system 1 according to the present embodiment will be described with reference to FIG. FIG. 1 is a diagram showing an example of the overall configuration of a labor detection system 1 according to the present embodiment.

  As shown in FIG. 1, the calving detection system 1 according to the present embodiment includes a calving detection device 10 that detects calving in advance or after, and one or more tags 20 attached to each cow. Be done. The tag 20 is preferably fixed and attached to the neck of the cow. Hereinafter, the tag 20 is assumed to be fixedly attached to the neck of the cow.

  The tag 20 is a device attached to a cow. One tag 20 is attached to one cow. The tag 20 includes an acceleration sensor that measures acceleration (acceleration of three axes of x-axis, y-axis, and z-axis) of a cow wearing the tag 20. Here, for example, when the cow is standing, the tag 20 indicates that the right direction to the traveling direction of the cow is the positive direction of the x axis, the traveling direction is the positive direction of the y axis, and the gravity direction. It is mounted so that it is in the positive direction of the z axis.

  The tag 20 transmits the measurement data including the acceleration sensor value measured by the acceleration sensor to the delivery detecting device 10 at every predetermined time (for example, every 2 seconds). The measurement data transmitted to the labor detection device 10 is accumulated (stored) in the measurement data storage unit 300 described later.

  The calving detection device 10 is a computer that detects calving of a cow in advance or after. The labor detection device 10 includes a first labor detection processing unit 100, a second labor detection processing unit 200, a measurement data storage unit 300, and an activity amount data storage unit 400.

  The first calving detection processing unit 100 detects calving in advance based on the measurement data stored in the measurement data storage unit 300 and the activity amount data stored in the activity amount data storage unit 400. To do. The first delivery detection processing unit 100 is realized by a process that causes a CPU (Central Processing Unit) or the like to execute one or more programs installed in the delivery detection device 10.

  Here, the activity amount data is data including the activity amount calculated from the acceleration sensor value. In addition, the activity amount is a value representing the intensity of movement of the cow during a predetermined time (for example, 10 minutes). The activity amount data is created by the first delivery detection processing unit 100 and stored in the activity amount data storage unit 400. The method of calculating the activity amount will be described later.

  The second calving detection processing unit 200 subsequently detects the calving of the cow based on the measurement data stored in the measurement data storage unit 300. The second labor detection processing unit 200 is realized by a process that causes the CPU or the like to execute one or more programs installed in the labor detection device 10.

  The measurement data storage unit 300 stores the measurement data received from the tag 20. The measurement data storage unit 300 can be realized by using an auxiliary storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). The measurement data storage unit 300 may be realized using a storage device or the like that is connected to the birth detection device 10 via a network.

  The activity amount data storage unit 400 stores the activity amount data created by the first delivery detection processing unit 100. The activity amount data storage unit 400 can be realized by using, for example, an auxiliary storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). The activity amount data storage unit 400 may be realized by using a storage device or the like connected to the birth detection device 10 via a network.

  The configuration of the labor detection system 1 shown in FIG. 1 is an example, and other configurations may be used. For example, the labor detection device 10 may be composed of a plurality of computers. In this case, for example, the first delivery detection processing unit 100 and the second delivery detection processing unit 200 may be realized by different computers. That is, the labor detection device 10 according to the present embodiment includes a first labor detection device having the first labor detection processing unit 100 and a second labor detection device having the second labor detection processing unit 200. It may be done.

  Further, for example, a part or all of the functions of the first delivery detection processing unit 100 and the second delivery detection processing unit 200 are connected to the delivery detection apparatus 10 via a network (for example, a cloud server or the like). May have.

  Furthermore, the method of transmitting the measurement data from the tag 20 to the birth detection device 10 is not limited. For example, the tag 20 may transmit the measurement data to the labor detection device 10 via a network such as the Internet, or may transmit the measurement data to the labor detection device 10 within the local network, or the short distance. The measurement data may be transmitted to the labor detection device 10 by wireless communication or the like.

<Measurement data stored in the measurement data storage unit 300>
Here, the measurement data stored in the measurement data storage unit 300 according to the present embodiment will be described with reference to FIG. FIG. 2 is a diagram showing an example of measurement data stored in the measurement data storage unit 300. When the delivery detection device 10 receives the measurement data from the tag 20, the delivery detection device 10 stores (stores) the measurement data in the measurement data storage unit 300 by the first delivery detection processing unit 100 or the second delivery detection processing unit 200. ) You can do it.

  As shown in FIG. 2, the measurement data storage unit 300 stores one or more pieces of measurement data for each tag ID, which is an example of identification information for identifying the tag 20. Since one tag 20 is attached to one cow, the tag ID may be identification information for identifying the cow (for example, individual identification information of the cow).

  Each measurement data includes the date and time and the acceleration sensor value. The date and time is, for example, the date and time when the tag 20 transmitted the measurement data. The date and time may be, for example, the date and time when the delivery detection device 10 received the measurement data.

The acceleration sensor value is a value of acceleration measured by the acceleration sensor included in the tag 20. The acceleration sensor value includes an x component indicating an acceleration component in the x axis direction, ay component indicating an acceleration component in the y axis direction, and az component indicating an acceleration component in the z axis direction. For example, the measurement data at the date and time “t ₀ ” includes the x component “x ₀ ”, the y component “y ₀ ”, and the z component “z ₀ ” of the acceleration sensor value. Similarly, for example, the measurement data of date and time “t ₁ ” includes the x component “x ₁ ”, the y component “y ₁ ”, and the z component “z ₁ ” of the acceleration sensor value. When Δt = t _{i + 1} −t _i (i is an integer of 0 or more), for example, Δt = 2 [seconds] because the acceleration sensor value is measured every 2 seconds.

  As described above, the measurement data stored in the measurement data storage unit 300 according to the present embodiment stores (stores) the measurement data including the date and time and the acceleration sensor value for each tag ID. In the following, the x component of the acceleration sensor value will be referred to as the “x component acceleration sensor value”, the y component of the acceleration sensor value will be referred to as the “y component acceleration sensor value”, and the z component of the acceleration sensor value will be referred to as the “z component acceleration sensor value”. Represent

<Activity data stored in activity data storage unit 400>
Next, the activity amount data stored in the activity amount data storage unit 400 according to the present embodiment will be described with reference to FIG. FIG. 3 is a diagram showing an example of activity amount data stored in the activity amount data storage unit 400.

  As shown in FIG. 3, the activity amount data storage unit 400 stores activity amount data for the past week for each tag ID. It should be noted that the past one week is an example, and the activity amount data of an arbitrary past period may be stored in the activity amount data storage unit 400.

  Each activity amount data includes the date and time and the activity amount. The date and time is the date and time set by the first delivery detection processing unit 100. The activity amount is a value of the activity amount calculated by the first delivery detection processing unit 100.

For example, the activity amount data of date “s _1,0 ” includes the activity amount “v _1,0 ”. Similarly, the activity amount data “v _1,1 ” is included in the activity amount data of the date and time “s _1,1 ”. When Δs _j = s _{j, i + 1} −s _{j, i} (j = 1, ..., 7, i is an integer of 0 or more), for example, as described later, the activity amount is every 10 minutes. As calculated, Δs _j = 10 [min].

  As described above, the activity amount data stored in the activity amount data storage unit 400 according to the present embodiment stores the activity amount data for the past week including the date and time and the activity amount for each tag ID. Has been done. Although FIG. 3 shows the activity amount data of the tag ID “Tag1” as an example, the activity amount data storage unit 400 also stores activity amount data of other tag IDs.

<Functional configuration of the first labor detection processing unit 100>
Next, the functional configuration of the first labor detection processing unit 100 according to the present embodiment will be described with reference to FIG. FIG. 4 is a diagram showing an example of a functional configuration of the first labor detection processing unit 100.

  As shown in FIG. 4, the first delivery detection processing unit 100 according to the present embodiment includes an acquisition unit 101, a preprocessing unit 102, a first peak number calculation unit 103, an activity amount calculation unit 104, and It has an index value calculation unit 105, a comparison determination unit 106, a cumulative value calculation unit 107, and a first delivery detection unit 108.

  The acquisition unit 101 acquires the y-component acceleration sensor value of the measurement data stored in the measurement data storage unit 300. At this time, the acquisition unit 101 acquires the y-component acceleration sensor value within a predetermined time width (for example, 10 minutes) for each tag ID, for example. Hereinafter, it is assumed that the predetermined time width is 10 minutes.

  The preprocessing unit 102 performs preprocessing on the y component acceleration sensor value acquired by the acquisition unit 101. Examples of the pre-processing include loss interpolation (resampling) processing of the y-component acceleration sensor value. The pre-processing may include, for example, noise removal processing.

  The first peak number calculation unit 103 calculates the number of peaks of the preprocessed y-component acceleration sensor value in 10 minutes. Here, the peak in the process executed by the first delivery detection processing unit 100 (first delivery detection process) is the y component acceleration sensor value that locally becomes the maximum value (or y that becomes the minimum value locally). Component acceleration sensor value).

  The activity amount calculation unit 104 calculates the activity amount from the number of peaks calculated by the first peak number calculation unit 103. Then, the activity amount calculation unit 104 creates activity amount data including this activity amount and stores it in the activity amount data storage unit 400. Here, the activity amount calculation unit 104 calculates, for example, the total of the peak numbers in the past 90 minutes as the activity amount. However, the past 90 minutes is an example, and the total of the peak numbers of arbitrary past periods may be calculated as the activity amount.

  The index value calculation unit 105 calculates a predetermined index value using the activity amount data stored in the activity amount data storage unit 400. The index value calculation unit 105 calculates, for example, the average value and the standard deviation of the activity amount at the same time in the past one week, and then calculates the index value by the index value = average value + standard deviation × c. Here, c is a preset parameter, and may be set to, for example, c = 1.5. However, c can be set to an arbitrary value of 0 or more.

  The comparison determination unit 106 compares and determines the magnitude relationship between the activity amount calculated by the activity amount calculation unit 104 and the index value calculated by the index value calculation unit 105.

  When the comparison determination unit 106 determines that the activity amount is larger than the index value, the cumulative value calculation unit 107 accumulates the difference between the activity amount and the index value. On the other hand, when the comparison determination unit 106 determines that the activity amount is equal to or less than the index value, the cumulative value calculation unit 107 clears the cumulative value so far to zero.

  The first labor detection unit 108 determines whether the cumulative value calculated by the cumulative value calculation unit 107 satisfies the first labor detection condition. When the first calving detection unit 108 determines that the first calving detection condition is satisfied, the first calving detection unit 108 notifies that the calving is detected, assuming that the corresponding cow has a calving sign. Here, the first labor detection condition may be, for example, that the cumulative value calculated by the cumulative value calculating unit 107 exceeds a predetermined threshold value.

<Functional configuration of second labor detection processing unit 200>
Next, the functional configuration of the second delivery detection processing unit 200 according to the present embodiment will be described with reference to FIG. FIG. 5 is a diagram illustrating an example of a functional configuration of the second delivery detection processing unit 200.

  As shown in FIG. 5, the second delivery detection processing unit 200 according to the present embodiment includes an acquisition unit 201, a preprocessing unit 202, a second peak number calculation unit 203, and a second delivery detection unit 204. Have and.

  The acquisition unit 201 acquires the y-component acceleration sensor value of the measurement data stored in the measurement data storage unit 300. At this time, the acquisition unit 201 acquires the y-component acceleration sensor value for 10 minutes for each tag ID, for example.

  The preprocessing unit 202 performs preprocessing on the y-component acceleration sensor value acquired by the acquisition unit 201. Examples of the pre-processing include loss interpolation (resampling) processing of the y-component acceleration sensor value. The pre-processing may include, for example, noise removal processing.

  The second peak number calculation unit 203 calculates the number of peaks of the preprocessed y-component acceleration sensor value in 10 minutes. Here, the peak in the process executed by the second labor detection processing unit 200 (second labor detection process) is the y component acceleration sensor value that locally becomes the maximum value and the y component that locally becomes the minimum value. Among the acceleration sensor values, the acceleration sensor values satisfying a predetermined condition.

  The second delivery detecting unit 204 determines whether or not the number of peaks calculated by the second peak number calculating unit 203 and the pre-processed y component acceleration sensor value satisfy the second delivery detecting condition. . Then, when the second calving detection unit 204 determines that the second calving detection condition is satisfied, the second calving detection unit 204 notifies that the calving has been carried out and that calving has been detected. Here, the second labor detection condition is, for example, that the number of peaks calculated by the second peak number calculation unit 203 is a predetermined number or more, and the average of the y-component acceleration sensor values after preprocessing in 10 minutes. The value may be equal to or greater than a predetermined first value or equal to or less than a predetermined second value.

  Since the acquisition unit 101 and the acquisition unit 201 have the same function, the acquisition unit 101 and the acquisition unit 201 may be realized by the same program or module. Similarly, since the preprocessing unit 102 and the preprocessing unit 202 have the same function, the preprocessing unit 102 and the preprocessing unit 202 may be realized by the same program or module.

<First labor detection processing>
Hereinafter, the first calving detection process (that is, the process in the case of detecting calving of a cow in advance) will be described with reference to FIG. 6. FIG. 6 is a flowchart showing an example of the first labor detection process. The process shown in FIG. 6 is repeatedly executed, for example, every 10 minutes. However, 10 minutes is an example, and may be repeatedly performed at any time. Hereinafter, the index indicating the number of repetitions will be k. Therefore, k is also a time index that represents a certain time (for example, the current time) at the start point of the repetition. In this case, for example, k-1 represents a time 10 minutes before the certain time, and k-2 represents a time 20 minutes before the certain time.

  First, the acquisition unit 101 acquires, for each tag ID, the y-component acceleration sensor value included in the measurement data for the past 10 minutes from the measurement data storage unit 300 based on the current time (step S101). As described above, the acquisition unit 101 stores, for each cow (tag ID), the y-component acceleration sensor value included in the measurement data in the past predetermined time width (for example, the past 10 minutes) based on the current time as the measurement data storage. Obtained from the section 300. Note that such a predetermined time width is not limited to 10 minutes, and the user of the labor detection device 10 can set an arbitrary time width, for example.

Hereinafter, as an example, the description will be continued assuming that the y-component acceleration sensor values y ₀ , y ₁ , ..., Y _M of the tag ID “Tag1” in the past 10 minutes have been acquired. Note that, for example, when the acceleration sensor measures the acceleration every 2 seconds, M ≦ 299.

  Next, the preprocessing unit 102 performs preprocessing on the y component acceleration sensor value acquired in step S101 (step S102). That is, the pre-processing unit 102 performs a loss interpolation (resampling) process on the y-component acceleration sensor value. The defect interpolation process is a process of interpolating a defect when the y component acceleration sensor value cannot be measured (or cannot be collected from the tag 20) in order to improve the accuracy of labor detection.

As a result, the y-component acceleration sensor value after preprocessing is renumbered to obtain the y-component acceleration sensor values y ₀ , y ₁ , ..., Y _N. Note that, for example, when the y component acceleration sensor values y ₀ , y ₁ , ..., Y _M are subjected to defect interpolation (resampling) processing at 2-second intervals, N = 299, and the y component acceleration sensor value y ₀ , y ₁ , ..., Y ₂₉₉ are obtained.

  Note that the preprocessing unit 102 may perform the loss interpolation process after performing the noise removal process, for example. By performing the noise removal process, for example, the y-component acceleration sensor value indicating the movement of the cow that may cause noise can be removed.

  Next, the first peak number calculation unit 103 calculates the number of peaks of the preprocessed y-component acceleration sensor value in 10 minutes (step S103). Here, the peak in the first labor detection process is, as described above, the y-component acceleration sensor value that locally becomes the maximum value (or the y-component acceleration sensor value that locally becomes the minimum value). .

Specifically, when the y component acceleration sensor value that locally becomes the maximum value is a peak (that is, when the maximum value of the y component acceleration sensor value is a peak), the first peak number calculation unit 103 For example, the number of peaks is calculated with the y-component acceleration sensor value y _i (i = 1, ..., N) satisfying the following expression (1) as a peak.

y _i −y _i−1 > δ, and y _i −y _{i + 1} > δ (1)
Further, when the y-component acceleration sensor value that is locally the minimum value is the peak (that is, when the minimum value of the y-component acceleration sensor value is the peak), the first peak number calculation unit 103 may, for example, The number of peaks is calculated with the y-component acceleration sensor value y _i (i = 1, ..., N) satisfying the equation (2) of (1) as a peak.

y _i-1 −y _i > δ, and y _{i + 1} −y _i > δ (2)
Here, δ in the above equations (1) and (2) is a preset parameter. In the above equations (1) and (2), δ may be different from each other.

  FIG. 7 shows an example of a peak when the maximum value of the y-component acceleration sensor value is set as the peak. FIG. 7 is a diagram for explaining an example of a peak in the first labor detection process.

  As shown in FIG. 7, when the y-component acceleration sensor values are connected by a line segment of the extreme value (maximum value or minimum value), the y-component acceleration sensor value at the top of the peak has a peak.

Here, the number of peaks calculated by the first peak number calculation unit 103 is stored in a storage device such as an auxiliary storage device or a RAM (Random Access Memory). In the following, it represents the peak number and p _k. The past peak numbers _pk-K , _{pk-K + 1} , ..., _pk-1 are also stored in this storage device. Here, it is preferable that K ≧ 9.

  In step S103 described above, the number of peaks is calculated by using either the maximum value or the minimum value of the y-component acceleration sensor value as a peak, but the present invention is not limited to this. For example, the maximum value and the minimum value of the y-component acceleration sensor value are calculated. The peak number may be calculated by using both of the values as peaks.

  Next, the activity amount calculation unit 104 calculates the activity amount from the number of peaks calculated in the above step S103 (step S104). That is, the activity amount calculation unit 104 calculates the total of the peak numbers in a predetermined past period (for example, the past 90 minutes) as the activity amount.

Specifically, when the activity amount is v8 _{, k} , the activity amount calculation unit 104 calculates the activity amount by, for example, v8 _{, k} = _pk-1 + _pk-2 + ... + _pk-9. calculate. In addition, the activity amount calculation unit 104 may calculate the activity amount by, for example, v8 _{, k} = _pk + _pk-1 + _pk-2 + ... + _pk-9 .

Then, the activity amount calculation unit 104 creates activity amount data including this activity amount and stores it in the activity amount data storage unit 400. At this time, the activity amount calculation unit 104 creates, for example, activity amount data including the date and time s _{8, k} and the activity amount v _{8, k} and stores the activity amount data storage unit 400.

In this way, the activity amount calculation unit 104 calculates the total of the peak numbers in a predetermined past period (for example, the past 90 minutes) as the activity amount. Note that, for example, when calculating the total of the number of peaks in the past 90 minutes as the amount of activity, if the number of peaks p _k-1 , p _k-2 , ..., P _k-9 has not been calculated yet, The first labor detection processing unit 100 repeatedly executes the above steps S101 to S103 to calculate the number of these peaks.

  Next, the index value calculation unit 105 calculates a predetermined index value using the activity amount data stored in the activity amount data storage unit 400 (step S105). That is, the index value calculation unit 105 calculates the index value by, for example, index value = average value + variable value after calculating the average value of the activity amount and the predetermined variable value at the same time in the past one week. Here, the variable value may be, for example, standard deviation × c or the like. Note that c is a preset parameter, and may be set to c = 1.5, for example. However, c can be set to an arbitrary value of 0 or more.

Specifically, the index value is V _k , and the activity amount at the same time in the past week (that is, the activity amount of the same index k in the past week) is v _{1, k} , v _{2, k} , ... In the case of v _{7, k} , the index value calculation unit 105 calculates the average value of these activity amounts and the standard deviation of these activity amounts, and then V _k = average value + standard deviation × 1.5. Calculate the index value by. It should be noted that this index value V _k becomes the delivery detection model at the index k indicating the time.

Next, the comparison determination unit 106 compares determines the amount of activity _{v 8, k} calculated in step S104 described above, the magnitude relation between the index values _{V k} calculated in step S105 described above (step S106).

Activity _{v 8} in step _{S106, k>} If it is determined that the index value _{V k,} the accumulated value calculating section 107 accumulates the difference between the amount of activity and the index value (step S107). That is, when the accumulated value and is W, the accumulated value calculating section 107 _{adds v} to _8, k -V _k to W. The initial value of W is 0.

Next, the first delivery detecting unit 108 determines whether or not the cumulative value W calculated in the above step S107 satisfies the first delivery detecting condition (step S108). That is, the first delivery detecting unit 108 determines whether the cumulative value W exceeds a predetermined threshold Th ₁ . Note that the threshold Th ₁ can be set to an arbitrary value, and for example, Th ₁ = 500 can be mentioned.

When it is determined in step S108 that the cumulative value W satisfies the first calving detection condition (that is, when the cumulative value W> threshold Th ₁ ), the first calving detection unit 108 determines that the cow (tag ID Assuming that there is a sign of parturition in the cow (on which the tag 20 of "Tag1" is attached), the fact that parturition is detected is notified (step S109). Note that examples of the notification destination include various terminal devices such as a PC (personal computer), a smartphone, and a tablet terminal that are registered in advance. As a result, delivery is detected in advance.

When it is determined in step S106 that the activity amount v _{8, k} ≦ index value V _k , the cumulative value calculation unit 107 clears the cumulative value W to 0 (step S110).

  As described above, the calving detection device 10 according to the present embodiment can detect the calving of the corresponding cow in advance. Here, in the first calving detection, as shown in FIG. 8, when the cumulative value of the difference between the activity amount of the cattle and the index value exceeds a predetermined threshold value, there is a sign of calving, Detect delivery. As a result, the calving detection device 10 according to the present embodiment can detect calving in advance with high accuracy.

  Note that, as shown in FIG. 8, since the amount of activity increases near the time when feeding is regularly performed, the delivery detecting device 10 according to the present embodiment may reduce the accuracy of delivery detection near this time. There is. For this reason, if there is any event in which the amount of activity other than feeding is large, it is necessary to register the time zone of the event in the labor detection device 10 as a time zone excluding labor detection.

<Second labor detection processing>
In the following, the second calving detection process (that is, the process when the calving of the cow is detected afterwards) will be described with reference to FIG. 9. FIG. 9 is a flowchart showing an example of the second labor detection process. The process shown in FIG. 9 is repeatedly executed, for example, every 10 minutes. However, 10 minutes is an example, and may be repeatedly performed at any time. Hereinafter, the index indicating the number of repetitions will be k. Therefore, k is also a time index that represents a certain time (for example, the current time) at the start point of the repetition. In this case, for example, k-1 represents a time 10 minutes before the certain time, and k-2 represents a time 20 minutes before the certain time.

  First, the acquisition unit 201 acquires, for each tag ID, the y-component acceleration sensor value included in the measurement data for the past 10 minutes from the measurement data storage unit 300 based on the current time (step S201). In this way, the acquisition unit 201 is included in the measurement data of a predetermined time period in the past (for example, the past 10 minutes) based on the current time, for each cow (tag ID), as in step S101 of FIG. The y component acceleration sensor value is acquired from the measurement data storage unit 300.

Hereinafter, as an example, the description will be continued assuming that the y-component acceleration sensor values y ₀ , y ₁ , ..., Y _M of the tag ID “Tag1” in the past 10 minutes have been acquired. Note that, for example, when the acceleration sensor measures the acceleration every 2 seconds, M ≦ 299.

  Next, the preprocessing unit 202 performs preprocessing on the y-component acceleration sensor value acquired in step S201 (step S202). That is, the pre-processing unit 202 performs the defect interpolation (resampling) process of the y-component acceleration sensor value, as in step S102 of FIG.

As a result, the y-component acceleration sensor value after preprocessing is renumbered to obtain the y-component acceleration sensor values y ₀ , y ₁ , ..., Y _N. Note that, for example, when the y component acceleration sensor values y ₀ , y ₁ , ..., Y _M are subjected to defect interpolation (resampling) processing at 2-second intervals, N = 299, and the y component acceleration sensor value y ₀ , y ₁ , ..., Y ₂₉₉ are obtained.

  Note that the preprocessing unit 202 may perform the loss interpolation process after performing the noise removal process, for example. By performing the noise removal process, for example, the y-component acceleration sensor value indicating the movement of the cow that may cause noise can be removed.

  Next, the second peak number calculation unit 203 calculates the number of peaks of the pre-processed y-component acceleration sensor value in 10 minutes (step S203). Here, the peak in the second labor detection process is an acceleration sensor value satisfying a predetermined condition, among the y-component acceleration sensor value that locally becomes the maximum value and the y-component acceleration sensor value that locally becomes the minimum value. That is. In addition, the predetermined condition is an acceleration sensor value at which the difference from (the candidate of) the immediately preceding peak is less than a predetermined value (for example, less than 500).

  That is, in the second labor detection process, both the maximum value and the minimum value of the acceleration sensor value are used as peak candidates, and candidates having a difference from the previous candidate that is equal to or more than a predetermined value are excluded. The example shown in FIG. 10 shows a case in which peaks at times 12:57 to 12:59 are specified. As shown in FIG. 10, among the peak candidates at times 12:57 to 12:59, the third and sixth candidates are excluded. As a result, the candidate excluding the third and sixth candidates becomes a peak. Such exclusion is performed, for example, to exclude the action of temporarily raising the head, which is seen in foraging behavior and the like.

Then, the second peak number calculation unit 203 calculates the peak number of the identified peak. Hereinafter, this peak number is represented by q _k .

Next, the second delivery detecting unit 204, the number of peaks q _k calculated in the above step S203, and the y component acceleration sensor values y ₀ , y ₁ , ..., After the pre-processing in the above step S202. It is determined whether y _N satisfies the second delivery detection condition (step S204). Here, the second labor detection condition is, for example, that the number of peaks q _k calculated by the second number-of-peaks calculation unit 203 is a predetermined threshold Th ₂ or more, and the y component acceleration sensor value y after preprocessing is The average value of ₀ , y ₁ , ..., Y _N is either a predetermined threshold value Th ₃ or more or a predetermined threshold value Th ₄ or less.

The thresholds Th ₂ , Th ₃ and Th ₄ can be set to arbitrary values, and examples thereof include Th ₂ = 25, Th ₃ = 250 and Th ₄ = −250.

If it is determined in step S204 that the peak number q _k and the pre-processed y component acceleration sensor values y ₀ , y ₁ , ..., Y _N satisfy the second delivery detection condition, the second delivery detection is performed. The section 204 notifies that the calving has been detected, assuming that the relevant cow (the cow to which the tag 20 having the tag ID “Tag1” is attached) has calved (step S205). Note that examples of the notification destination include various terminal devices such as a PC (personal computer), a smartphone, and a tablet terminal that are registered in advance. As a result, delivery is detected after the fact.

As described above, the calving detection device 10 according to the present embodiment can detect the calving of the cow in question. Here, in the second delivery detection, as described above, the peak number q _k and the pre-processed y-component acceleration sensor values y ₀ , y ₁ , ..., Y _N define the second delivery detection condition. If the condition is satisfied, it is detected that delivery has been performed. This is to detect a licking operation (an operation in which a calf licks a calf) performed after calving. That is, the second labor detection condition is a condition for detecting a motion of moving the head finely in the front-rear direction while tilting the head in the lower front direction (this is a characteristic motion as a licking motion). .

  As described above, the calving detection device 10 according to the present embodiment can detect the calving of the cow ex-post highly accurately by detecting the licking operation. In addition, when the calving of the cow is to be detected after the fact, for example, it may be determined whether the cow is performing the licking operation based on the behavior of the cow specified by another method. For example, only when it is determined that the cow is not performing foraging behavior by another method, it is determined whether the licking operation is performed by the second calving detection process described above, and the calving is performed. You may detect. This allows the delivery to be detected with higher accuracy.

  In particular, the labor detection device 10 according to the present embodiment has the device inserted into the vagina or the birth canal as compared with the conventional techniques disclosed in Patent Documents 2, 3, and 4, for example. Without it, the calving of the cow can be detected, and the calving of the cow can be detected in a hygienic manner. In addition, the labor detection device 10 according to the present embodiment detects acceleration of labor in advance based on the cumulative value of the difference between the activity amount and the index value and the subsequent labor detection based on the detection of the licking operation to accelerate acceleration. Compared with other labor detection technologies that use sensors (for example, the conventional technology disclosed in Patent Document 1) and other labor detection technologies that use a vibrometer (for example, the conventional technology disclosed in Patent Literature 5) Then, the delivery detection (pre-delivery detection or post-delivery detection) can be performed with high accuracy. Furthermore, in the present embodiment, the case where the tag 20 is fixed and attached to the neck of the cow has been described, but in this case, the tag 20 is attached to the cow in comparison with the case where the tag 20 is attached to another place (for example, the tail of the cow). It is possible to reduce a situation in which a person is dropped due to an operation.

  The present invention is not limited to the above specifically disclosed embodiments, and various modifications and changes can be made without departing from the scope of the claims.

1 Delivery Detection System 10 Delivery Detection Device 100 First Delivery Detection Processing Unit 101 Acquisition Unit 102 Pre-Processing Unit 103 First Peak Number Calculation Unit 104 Activity Level Calculation Unit 105 Index Value Calculation Unit 106 Comparison Determination Unit 107 Cumulative Value Calculation Unit 108 1st delivery detection part 200 2nd delivery detection processing part 201 Acquisition part 202 Pre-processing part 203 2nd peak number calculation part 204 2nd delivery detection part 300 Measurement data storage part 400 Activity data storage part

Claims (8)

A detection device for detecting the delivery of livestock,
Storage means for storing acceleration data measured by the three-axis acceleration sensor attached to the livestock,
An acquisition unit that acquires a component value in a predetermined direction included in each of a plurality of acceleration data during a predetermined time from the storage unit,
Based on the plurality of component values, detection means for detecting the delivery of the livestock,
A detection device comprising: A first peak number calculating means for calculating the number of peaks showing a locally maximum component value or a locally minimum component value among the plurality of component values;
An activity amount calculating means for calculating the total number of the peaks in a predetermined period in the past as an activity amount;
Using the past activity amount calculated by the activity amount calculation means, an index value calculation means for calculating a predetermined index value,
Comparing means for determining the magnitude relationship between the activity amount and the index value,
When it is determined that the activity amount is larger than the index value, a cumulative value calculating unit that calculates a cumulative value by accumulating a difference between the activity amount and the index value,
And having the detection means,
The detection device according to claim 1, wherein the delivery of the livestock is detected when the cumulative value is equal to or larger than a predetermined first threshold value. The index value calculating means,
Using the average value of the past activity amount, the standard deviation of the past activity amount, and a predetermined constant, by adding the average value to the product of the standard deviation and the constant, the The detection device according to claim 2, wherein an index value is calculated. The cumulative value calculating means,
The detection device according to claim 2, wherein the cumulative value is cleared to 0 when it is not determined that the activity amount is larger than the index value. A second peak for calculating the number of peaks showing a component value that locally becomes maximum and a component value that locally becomes minimum among the plurality of component values and that satisfies a predetermined condition. Has a number calculation means,
The detection means is
When the number of the peaks is a predetermined number or more and the average value of the plurality of component values is a predetermined second threshold value or more or a third threshold value or less, the delivery of the livestock is detected. The detection device according to claim 1. The predetermined condition is
When the locally maximum component value and the locally minimum component value are the peak candidates, the difference from the immediately preceding candidate is less than or equal to a predetermined value. The detection device according to claim 5. A detection device that detects the delivery of livestock
A storage procedure for storing in the storage means acceleration data measured by the triaxial acceleration sensor attached to the livestock,
An acquisition procedure for acquiring a component value in a predetermined direction included in each of a plurality of acceleration data during a predetermined time from the storage means,
Based on the plurality of component values, a detection procedure for detecting the delivery of the livestock,
The detection method characterized by performing.   A program for causing a computer to function as each unit in the detection device according to claim 1.

Images