Information processing device, information processing method, and program

Abstract

To detect abnormality of a livestock with high accuracy.SOLUTION: According to an embodiment, an information processing device is an information processing device for determining whether abnormality occurs in a livestock, and includes a posture estimation part for estimating a posture of the livestock on the basis of measurement data measured by a sensor attached to the livestock, a score calculation part for calculating a score representing an abnormality degree of the livestock on the basis of the posture and the measurement data, and a determination part for determining whether abnormality occurs in the livestock on the basis of the score.SELECTED DRAWING: Figure 7

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] It is known that fattening cattle are prone to develop abnormalities such as difficulty standing in the later stages of fattening. If this condition is left untreated, the cattle may die from suffocation, etc., so fattening cattle farmers patrol the cattle sheds to check whether the cattle are experiencing any abnormalities such as difficulty standing.

[0003] It is also known that breeding cows are prone to developing abnormalities such as difficulty standing before and after giving birth, so breeding farmers similarly patrol their cowsheds to check whether their cows are experiencing any abnormalities such as difficulty standing.

[0004] Furthermore, cows may have difficulty standing up due to bloat, etc. Bloat is a disease that can occur in ruminant livestock such as cows, and can develop when, for example, they overeat fermented feed such as green grass. When bloat develops, the cow's abdomen swells, making it difficult for the cow to stand up.

[0005] Methods for estimating the above-mentioned difficulty in standing up have been conventionally known (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0006] [Patent Document 1] International Publication No. 2019 / 039118 Summary of the Invention [Problem to be solved by the invention]

[0007] However, in the past, the accuracy of detecting abnormalities in livestock such as difficulty in standing up was not always high, and false positives occurred. For example, if a cow shook its head or scratched its body while lying down, it could be erroneously detected as having difficulty standing up.

[0008] One embodiment of the present invention has been made in consideration of the above points, and has an object to detect abnormalities in livestock with high accuracy. [Means for solving the problem]

[0009] In order to achieve the above-mentioned object, an information processing device according to one embodiment is an information processing device that determines whether or not an abnormality has occurred in a livestock, and has a posture estimation unit that estimates the posture of the livestock based on measurement data measured by a sensor attached to the livestock, a score calculation unit that calculates a score representing the degree of abnormality of the livestock based on the posture and the measurement data, and a determination unit that determines whether or not an abnormality has occurred in the livestock based on the score. Effect of the Invention

[0010] Abnormalities in livestock can be detected with high accuracy. [Brief description of the drawings]

[0011] [Figure 1] 1 is a diagram illustrating an example of an overall configuration of an anomaly detection system according to an embodiment of the present invention. [Diagram 2] FIG. 4 is a diagram showing an example of measurement data stored in a measurement data storage unit. [Diagram 3] FIG. 4 is a diagram illustrating an example of a functional configuration of a posture estimation processing unit according to the present embodiment. [Figure 4] FIG. 11 is a diagram for explaining an example of the relationship between each attitude and each coordinate axis. [Diagram 5] FIG. 2 is a diagram illustrating an example of a functional configuration of an abnormality detection processing unit according to the present embodiment. [Figure 6] 10 is a flowchart illustrating an example of a tag angle calculation process according to the present embodiment. [Figure 7] 10 is a flowchart illustrating an example of a posture estimation process according to the present embodiment. [Figure 8] 10 is a flowchart illustrating an example of an abnormality detection process according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] An embodiment of the present invention will be described below. In this embodiment, an abnormality detection system 1 that can detect abnormalities such as difficulty in standing up (e.g., difficulty in standing up that occurs in the later stage of fattening, difficulty in standing up that occurs before and after childbirth, difficulty in standing up due to bloat, etc.) as a target of cattle with high accuracy will be described. However, cattle are only one example of livestock, and livestock are not limited to cattle, and this embodiment can also be applied to various animals (e.g., horses, sheep, pigs, etc.) that can develop abnormalities such as difficulty in standing up like cattle.

[0013] <Overall configuration of anomaly detection system 1> The overall configuration of an anomaly detection system 1 according to this embodiment is shown in Fig. 1. As shown in Fig. 1, the anomaly detection system 1 according to this embodiment includes an anomaly detection device 10 that detects abnormalities in cows, one or more tags 20 attached to each cow, and a reference air pressure sensor 30 that measures reference air pressure. Note that the tag 20 is preferably attached and fixed around the neck of the cow by a collar, belt, or the like.

[0014] The tag 20 is a device attached to a cow. One tag 20 is attached to each cow. The tag 20 includes an acceleration sensor that measures the acceleration (acceleration on three axes: x-axis, y-axis, and z-axis) of the cow to which the tag 20 is attached, and a pressure sensor that measures air pressure. Here, a coordinate system consisting of the x-axis, y-axis, and z-axis is fixed to the tag 20, and the tag 20 is attached to the cow when the cow is standing and its neck is not bent (straight) so that the direction opposite to the direction of gravity is the positive direction of the x-axis, the moving direction of the cow is the positive direction of the y-axis, and the rightward direction with respect to the moving direction of the cow is the positive direction of the z-axis. However, the directions of the axes (x-axis, y-axis, and z-axis) relative to the cow are not limited to these, and it is also possible to attach the tag 20 to the cow so that each axis is oriented in any direction while maintaining the relative positional relationship of each axis in the coordinate system.

[0015] The tag 20 transmits measurement data including an acceleration sensor value measured by the acceleration sensor and an atmospheric pressure sensor value measured by the atmospheric pressure sensor to the anomaly detection device 10 at predetermined time intervals Δ1 (for example, Δ1=2 [s], etc.). The measurement data transmitted to the anomaly detection device 10 is accumulated (stored) in a measurement data storage unit 300, which will be described later.

[0016] The reference air pressure sensor 30 is installed at a predetermined position in the cowshed (e.g., on the ground in the cowshed) and measures the reference air pressure. The reference air pressure sensor 30 transmits reference air pressure data including a reference air pressure sensor value indicating the measured air pressure to the anomaly detection device 10 every predetermined time Δ2 (e.g., Δ2=2 [s], etc.). The reference air pressure data transmitted to the anomaly detection device 10 is accumulated (stored) in a reference air pressure data storage unit 400, which will be described later. In the following, for simplicity, it is assumed that Δ2=Δ1 and the sampling times of the reference air pressure sensor value and the air pressure sensor value are synchronized. Note that if Δ2≠Δ1, or if the sampling times of the reference air pressure sensor value and the air pressure sensor value are not synchronized even if Δ2=Δ1, then resampling or the like may be performed as appropriate.

[0017] The abnormality detection device 10 is one or more computers that detects a cow's difficulty in standing up as an abnormality. The abnormality detection device 10 has a posture estimation processing unit 100, an abnormality detection processing unit 200, a measurement data storage unit 300, a reference atmospheric pressure data storage unit 400, a tag angle data storage unit 500, and a posture data storage unit 600.

[0018] The posture estimation processing unit 100 estimates the posture of the cow using the measurement data stored in the measurement data storage unit 300 and the reference atmospheric pressure data stored in the reference atmospheric pressure data storage unit 400. Here, the inventor of the present application has found, as a result of observing actual cows, that the postures of cows are roughly classified into "standing" and "lying", and the lying posture is further classified into lying and prone. In addition, the inventor has also found that the difference between lying and prone is particularly evident in the rotation angle of the neck around the axis of the forward direction of the face. That is, the inventor has found that the neck is not bent and is often in a straight state when standing and lying, whereas the neck is not straight and is often rotated to a certain extent around the axis of the forward direction of the face when lying. Based on these findings, the posture estimation processing unit 100 according to this embodiment estimates whether the cow's posture is lying or prone from the rotation angle of the neck when the cow is lying.

[0019] Incidentally, standing refers to the position of a cow standing up. On the other hand, both prone and lying refer to the position of a cow lying down, but generally speaking, prone is a position where the knees are lowered and the chest is raised (a so-called sitting position), whereas lying is a position where the cow is lying on its side with its limbs thrown out.

[0020] The abnormality detection processing unit 200 detects abnormalities in the cow (difficulty in standing up) using the posture estimated by the posture estimation processing unit 100 and the measurement data stored in the measurement data storage unit 300. Here, the inventor of the present application has found, as a result of observing actual cows, that even if they are in the same sleeping posture, lying prone is almost unlikely to cause difficulty in standing up, whereas lying sideways may cause difficulty in standing up. Based on this finding, the abnormality detection processing unit 200 increases a predetermined score only when the cow is lying sideways, and detects that the cow is having difficulty in standing up when this score exceeds a certain threshold value.

[0021] The posture estimation processing unit 100 and the anomaly detection processing unit 200 are realized, for example, by a processor such as a CPU (Central Processing Unit) executing one or more programs installed in the anomaly detection device 10. However, all or part of the functions of the posture estimation processing unit 100 and the anomaly detection processing unit 200 may be realized by a cloud server or the like connected to the anomaly detection device 10 via a communication network.

[0022] The measurement data storage unit 300 stores the measurement data received from each tag 20. The measurement data storage unit 300 stores the measurement data for each time Δ1 for each tag 20. Note that the measurement data received from the tag 20 is accumulated (stored) in the measurement data storage unit 300 by, for example, the posture estimation processing unit 100 or the anomaly detection processing unit 200.

[0023] The reference atmospheric pressure data storage unit 400 stores the reference atmospheric pressure data received from the reference atmospheric pressure sensor 30. The reference atmospheric pressure data for each time period Δ2 is stored in the reference atmospheric pressure data storage unit 400. The reference atmospheric pressure data received from the reference atmospheric pressure sensor 30 is accumulated (stored) in the reference atmospheric pressure data storage unit 400 by, for example, the posture estimation processing unit 100 or the anomaly detection processing unit 200.

[0024] The tag angle data storage unit 500 stores tag angle data for estimating whether the posture of the cow is lying or prone. Details of the tag angle data will be described later.

[0025] The posture data storage unit 600 stores posture data indicating the posture estimated by the posture estimation processing unit 100.

[0026] The measurement data storage unit 300, the reference atmospheric pressure data storage unit 400, the tag angle data storage unit 500, and the attitude data storage unit 600 can be realized using a storage device such as a hard disk drive (HDD) or a solid state drive (SSD). However, all or a part of the measurement data storage unit 300, the reference atmospheric pressure data storage unit 400, the tag angle data storage unit 500, and the attitude data storage unit 600 may be realized by a database server or the like connected to the anomaly detection device 10 via a communication network.

[0027] <Measurement Data Stored in Measurement Data Storage Unit 300> An example of the measurement data stored in the measurement data storage unit 300 according to this embodiment is shown in Fig. 2. As shown in Fig. 2, the measurement data storage unit 300 stores one or more measurement data for each tag ID, which is information for identifying a tag. Since one tag 20 is attached to one cow, the tag ID may be information for identifying the cow (for example, individual identification information of the cow). In the following, the tag ID and information for identifying the cow are regarded as the same thing, and it is assumed that the cow can be identified by the tag ID.

[0028] Each piece of measurement data includes a date and time, an acceleration sensor value, and an air pressure sensor value. The date and time is, for example, the date and time when the tag 20 transmits the measurement data. Note that the date and time may be the date and time when the anomaly detection device 10 receives the measurement data.

[0029] The acceleration sensor value is an acceleration value measured by an acceleration sensor included in the tag 20. The acceleration sensor value includes an x ​​component indicating an acceleration component in an x-axis direction, a y component indicating an acceleration component in a y-axis direction, and a z component indicating an acceleration component in a z-axis direction. For example, measurement data at date and time "t1" includes an x ​​component of the acceleration sensor value "x1", a y component of the acceleration sensor value "y1", and a z component of the acceleration sensor value "z1". Similarly, measurement data at date and time "t2" includes an x ​​component of the acceleration sensor value "x2", a y component of the acceleration sensor value "y2", and a z component of the acceleration sensor value "z2". Hereinafter, the x component of the acceleration sensor value is also referred to as the "x-component acceleration sensor value", the y component of the acceleration sensor value is also referred to as the "y-component acceleration sensor value", and the z component of the acceleration sensor value is also referred to as the "z-component acceleration sensor value".

[0030] The air pressure sensor value is the value of air pressure measured by an air pressure sensor included in the tag 20. For example, the measurement data at date and time "t1" includes the air pressure sensor value "p1". Similarly, for example, the measurement data at date and time "t2" includes the air pressure sensor value "p2".

[0031] In this way, the measurement data stored in the measurement data storage unit 300 includes the date and time, the acceleration sensor value, and the air pressure sensor value for each tag ID. i+1 -t i (i is an integer greater than or equal to 1).

[0032] <Functional configuration of the posture estimation processing unit 100> The functional configuration of the posture estimation processing unit 100 according to this embodiment is shown in Fig. 3. As shown in Fig. 3, the posture estimation processing unit 100 according to this embodiment includes a tag angle calculation unit 101, a basic angle calculation unit 102, and a posture estimation unit 103.

[0033] The tag angle calculation unit 101 calculates the tag angle of each cow every predetermined time Δ3 (for example, Δ3=1 [m]). At this time, the tag angle calculation unit 101 extracts only the gravitational acceleration component from each of the x-component acceleration sensor values ​​and z-component acceleration sensor values ​​contained in the current measurement data of the cow, and calculates the tag angle from the gravitational acceleration component in the x-axis direction and the gravitational acceleration component in the z-axis direction. Then, the tag angle calculation unit 101 stores the tag angle data including the date and time, tag ID, and tag angle in the tag angle data storage unit 500. Hereinafter, the gravitational acceleration component in the x-axis direction is also referred to as the "x-component gravitational acceleration value," and the gravitational acceleration component in the z-axis direction is also referred to as the "z-component gravitational acceleration value."

[0034] Here, the tag angle is the angle that indicates how much the z-axis has rotated with the direction of gravity as the reference (i.e., 0) and the y-axis as the axis of rotation. Ideally, as shown in Figure 4, when the cow is standing or lying down, the positive direction of the x-axis is opposite to the direction of gravity, and the positive direction of the z-axis is to the right of the cow's front. Therefore, in this case, the tag angle is θ, and the x-component gravitational acceleration value is g x , the z-component gravitational acceleration value is g z Then, θ=tan -1 (g x / g z )=-π / 2. On the other hand, when lying down, the positive direction of the x-axis is to the right (or left) of the front of the cow, and the positive direction of the z-axis is the direction of gravity (or the opposite direction to the direction of gravity). Therefore, in this case, θ=tan -1 (g x / g z )=nπ (where n=-1, 0, 1).

[0035] However, FIG. 4 above shows an ideal situation in which each tag 20 is attached to the same position on the neck of each cow, and the collar or belt for attaching the tag 20 does not shift (or the influence of these can be ignored). In reality, the position of the tag 20 may differ slightly depending on the individual, and the collar or belt may shift depending on the movement of the cow. Therefore, in this embodiment, an angle called the basic angle is introduced, and the basic angle of each cow is calculated every predetermined time Δ4 (for example, Δ4=10 [m]), and the posture of the cow is estimated to be either prone or lying from the difference between the tag angle and the basic angle. Note that Δ4 ≧ Δ3 is required, and it is particularly preferable that it can be expressed as Δ4 = N1 × Δ3 for a certain natural number N1.

[0036] The basic angle calculation unit 102 calculates the basic angle of each cow for each predetermined time Δ4. At this time, the basic angle calculation unit 102 calculates the basic angle of the cow using past tag angle data of the cow among the tag angle data stored in the tag angle data storage unit 500 and past posture data of the cow among the posture data stored in the posture data storage unit 600. Specifically, the basic angle calculation unit 102 calculates the average value of tag angles that satisfy a predetermined condition among the tag angles of the cow for the past T1 hours (for example, T1=12 [h] or T1=24 [h]) as the basic angle of the cow. Here, the tag angle that satisfies the predetermined condition is, for example, the tag angle at the time when the cow is standing, but in addition to this, a condition that the amount of fluctuation in the tag angle is small may be set.

[0037] The posture estimation unit 103 estimates whether the posture of the cow is standing or lying down by air pressure analysis every predetermined time Δ4. Here, air pressure analysis is a method of estimating whether the posture of the cow is standing or lying down from the difference between the air pressure sensor value included in the measurement data of the cow and the reference air pressure sensor value included in the reference air pressure data at the same time on the same day. For details of the air pressure analysis, please refer to, for example, JP 2019-97475 A, JP 2019-103442 A, JP 2020-198828 A, JP 2020-198829 A, etc.

[0038] Furthermore, if the posture estimation unit 103 estimates that the posture of the cow is lying down, it further estimates whether the cow is lying prone or lying down using the current tag angle and basic angle of the cow. Then, the posture estimation unit 103 stores posture data including the date and time, tag ID, and estimated posture in the posture data storage unit 600.

[0039] <Functional configuration of the abnormality detection processing unit 200> The functional configuration of the anomaly detection processing unit 200 according to this embodiment is shown in Fig. 5. As shown in Fig. 5, the anomaly detection processing unit 200 according to this embodiment includes a score calculation unit 201 and an anomaly detection unit 202.

[0040] The score calculation unit 201 calculates a predetermined score at a predetermined time Δ5 (for example, Δ5=Δ4) using the measurement data of the cow among the measurement data stored in the measurement data storage unit 300, according to the posture represented by the latest posture data of the cow among the posture data stored in the posture data storage unit 600. Note that it is preferable that Δ5≧Δ4, and it is more preferable that it can be expressed as Δ5=N2×Δ4 for a certain natural number N2, and in particular, Δ5=Δ4 is ​​most suitable.

[0041] The abnormality detection unit 202 detects an abnormality in the cow (i.e., difficulty in standing up) by using the score calculated by the score calculation unit 201. At this time, the abnormality detection unit 202 detects that an abnormality has occurred in the cow, for example, when the score exceeds a predetermined threshold value.

[0042] <Tag angle calculation process> The tag angle calculation process according to this embodiment will be described with reference to Fig. 6. Note that the following steps S101 to S103 are repeatedly executed for each cow at predetermined time intervals Δ3. Below, a case where a tag angle for a certain cow is calculated will be described.

[0043] First, the tag angle calculation unit 101 obtains the x-component gravitational acceleration value and the z-component gravitational acceleration value from the x-component acceleration sensor value and the z-component acceleration sensor value contained in the latest measurement data for the cow (i.e., the latest measurement data including the tag ID of the tag 20 attached to the cow) from the measurement data stored in the measurement data memory unit 300 (step S101).

[0044] Here, the acceleration sensor value of each component is a composite value of the acceleration sensor value due to the cow's movement and the acceleration sensor value due to gravitational acceleration. Furthermore, when the acceleration sensor value due to the cow's movement and the acceleration sensor value due to gravitational acceleration are converted into a frequency spectrum, the acceleration sensor value due to the cow's movement appears in the high frequency components, while the acceleration sensor value due to gravitational acceleration appears in the low frequency components. For this reason, the tag angle calculation unit 101 can obtain (extract) the x-component gravitational acceleration value from the x-component acceleration sensor value and the z-component gravitational acceleration value from the z-component acceleration sensor value, for example, by using a low pass filter that attenuates the acceleration sensor value of a frequency equal to or higher than a predetermined frequency.

[0045] Next, the tag angle calculation unit 101 calculates the tag angle using the x-component gravitational acceleration value and the z-component gravitational acceleration value acquired in the above step S101 (step S102). x , the z-component gravitational acceleration value is g z Then, the tag angle calculation unit 101 calculates θ=tan -1 (g x / g z ) to calculate the tag angle θ.

[0046] Then, the tag angle calculation unit 101 creates tag angle data including the current date and time, the tag ID of the tag 20 attached to the cow, and the tag angle θ calculated in step S102 above, and stores this tag angle data in the tag angle data storage unit 500 (step S103). As a result, the tag angle data of each cow is stored in the tag angle data storage unit 500 for every predetermined time Δ3.

[0047] <Posture estimation processing> The posture estimation process according to this embodiment will be described with reference to Fig. 7. Note that the following steps S201 to S211 are repeatedly executed for each cow at predetermined time intervals Δ4. Below, a case will be described where the posture of a certain cow is estimated.

[0048] First, the basic angle calculation unit 102 calculates the basic angle of the cow using the past tag angle data of the cow from the tag angle data stored in the tag angle data memory unit 500 and the past posture data of the cow from the posture data stored in the posture data memory unit 600 (step S201).

[0049] Specifically, the basic angle calculation unit 102 extracts tag angles that satisfy a predetermined condition from the tag angles included in the tag angle data for the past T1 hours of the cow, and calculates the average value of the extracted tag angles as the basic angle of the cow. As described above, the tag angle that satisfies the predetermined condition may be, for example, the tag angle at the time when the cow is standing, or in addition, it may be a tag angle with little fluctuation in each time width of Δ4 (that is, for example, a tag angle with an amount of fluctuation that falls within a certain predetermined range). Below, the basic angle of the cow is represented by φ.

[0050] Next, the posture estimation unit 103 estimates whether the posture of the cow is standing or lying down by air pressure analysis (step S202). That is, the posture estimation unit 103 estimates whether the posture of the cow is standing or lying down by the following (1-1) to (1-3). In the following, for simplicity, it is assumed that there is no data loss and that noise removal is not required. For more details, please refer to, for example, JP 2019-97475 A, JP 2019-103442 A, JP 2020-198828 A, JP 2020-198829 A, etc.

[0051] (1-1) Among the measurement data stored in the measurement data storage unit 300, the barometric pressure sensor value included in the measurement data from Δ4 years ago to the present for the cow in question and the reference barometric pressure sensor value included in the reference barometric pressure data from Δ4 years ago to the present stored in the reference barometric pressure data storage unit 400 are acquired. Hereinafter, the index representing the date and time from Δ4 years ago to the present is defined as j=1, . . . , J, and the set of barometric pressure sensor values ​​acquired here is referred to as {P j ;j=1,...,J}, and the set of reference pressure sensor values ​​is {Q j ;j=1,···,J}.

[0052] (1-2) Calculate the differential pressure sensor value by subtracting the reference pressure sensor value from the pressure sensor value at the same date and time. That is, the set of differential pressure sensor values ​​{R j :=P j -Q j ;j=1, ,J}.

[0053] (1-3) Differential air pressure sensor value R j An index value, such as the average value of (j=1, ,J), is calculated, and if this index value is higher than a predetermined threshold th1 (<0), the cow's posture is presumed to be lying down, and if not, the cow's posture is presumed to be standing.

[0054] Next, posture estimation section 103 determines whether or not the posture estimated in above step S202 is a lying down posture (step S203).

[0055] If it is not determined in step S203 that the cow is in a lying position, the position estimation unit 103 estimates that the cow's position is standing (step S204).

[0056] On the other hand, if it is determined in step S203 that the cow is in a lying position, the position estimation unit 103 compares the tag angle θ included in the latest tag angle data of the cow among the tag angle data stored in the tag angle data storage unit 500 with the basic angle φ calculated in step S201 (step S205). That is, the position estimation unit 103 calculates Δθ:=|θ-φ|.

[0057] Next, the attitude estimation unit 103 determines whether or not the tag angle θ is sufficiently close to the basic angle φ as a result of the comparison in the above step S205 (step S206). That is, the attitude estimation unit 103 determines whether or not Δθ is smaller than a predetermined threshold ε.

[0058] If it is determined in step S206 that the tag angle θ is sufficiently close to the base angle φ, the attitude estimation unit 103 calculates the degree of fluctuation of the tag angle in a certain period T2 in the past (for example, T2=Δ4, etc.) (step S207). There are various index values ​​that represent the degree of fluctuation of the tag angle, and any index value may be calculated, but for example, variance, standard deviation, etc. can be used.

[0059] For example, if Δ3=1[m], Δ4=10[m], and T2=Δ4, the tag angles for a certain period T2 in the past are 10 tag angles θ1(=θ), θ2, . . . , θ 10 Therefore, in this case, these 10 tag angles θ1, θ2, . . . , θ 10 The variance, standard deviation, etc. of the tag angle can be calculated as the degree of fluctuation in the tag angle.

[0060] Next, the attitude estimation unit 103 determines whether or not the degree of fluctuation calculated in the above step S207 is large (step S208). That is, the attitude estimation unit 103 determines whether or not the degree of fluctuation of the tag angle is larger than a predetermined threshold value th2.

[0061] If it is not determined in step S208 above that the tag angle is significantly unstable, the posture estimation unit 103 estimates that the posture of the cow is lying down (step S209).

[0062] On the other hand, if it is determined in step S206 above that the tag angle θ is not sufficiently close to the basic angle φ, or if it is determined in step S208 above that the tag angle varies greatly, the posture estimation unit 103 estimates that the posture of the cow is lying down (step S210). In this way, even if the tag angle θ is sufficiently close to the basic angle φ, if the tag angle varies greatly, the posture of the cow is estimated to be lying down. This is because even if the cow has difficulty standing up, it may raise its head several times, and in this case the tag angle θ may be sufficiently close to the basic angle φ.

[0063] Then, the posture estimation unit 103 stores posture data including the current date and time, the tag ID, and the posture estimated in the above step S204, step S209, or step S210 in the posture data storage unit 600 (step S211). As a result, the posture data of each cow is stored in the posture data storage unit 600 for each predetermined time Δ4.

[0064] It is not necessary to execute steps S207 to S208, and for example, if it is determined in step S206 that the tag angle θ is sufficiently close to the basic angle φ, the cow's posture may be estimated as lying down. However, executing steps S207 to S208 can improve the accuracy of detecting difficulty in standing up.

[0065] <Anomaly detection processing> The abnormality detection process according to this embodiment will be described with reference to Fig. 8. Note that the following steps S301 to S306 are repeatedly executed for each cow at predetermined time intervals Δ5. Below, a case where an abnormality in a certain cow is detected will be described.

[0066] First, the score calculation unit 201 determines whether the posture represented by the latest posture data of the cow among the posture data stored in the posture data memory unit 600 (i.e., the current posture of the cow) is standing, prone, or lying (step S301).

[0067] If it is determined in step S301 above that the current posture of the cow is lying down, the score calculation unit 201 does nothing. If it is determined in step S301 above that the current posture of the cow is standing up, the score calculation unit 201 resets the score S to 0 (step S302).

[0068] On the other hand, if the current posture of the cow is determined to be lying in step S301 above, or following step S302, the score calculation unit 201 calculates the score s (step S303). That is, the score calculation unit 201 calculates the score s according to the following (2-1) to (2-4). Here, the score s is an index value for detecting abnormalities in the cow, and is calculated from the acceleration sensor value included in the measurement data of the cow from Δ5 years ago to the present. In the following, for simplicity, it is assumed that there is no data loss and that noise removal is not required. For details of the calculation method of the score s, please refer to, for example, JP 2019-97475 A, JP 2020-198828 A, JP 2020-198829 A, etc.

[0069] (2-1) Among the measurement data stored in the measurement data storage unit 300, the acceleration sensor values ​​(x-component acceleration sensor value, y-component acceleration sensor value, and z-component acceleration sensor value) included in the measurement data of the cow from Δ5 years ago to the present are acquired. Hereinafter, an index indicating the date and time from Δ5 years ago to the present is k=1, . . . , K, and the set of acceleration sensor values ​​acquired here is expressed as {(X k ,Y k ,Z k );k=1, ,K}. Note that X k is the x component acceleration sensor value, Y k is the y-component acceleration sensor value, Z k represents the z-component acceleration sensor value.

[0070] (2-2) Acceleration sensor value (X k ,Y k ,Z k ) is calculated. That is, for each k = 1, , K, k =(|X k |2 +|Y k | 2 +|Z k | 2 ) 1 / 2 Calculate.

[0071] (2-3) Calculate the standard deviation of the L2 norm for a given time unit Δ6 (where Δ6<Δ5). In the following, we will refer to the calculated standard deviation as σ d As a result, the set of standard deviations {σ d ;d=1, ,D} is obtained. Note that it is preferable that Δ6 can be expressed as Δ5=N3×Δ6 for some natural number N3.

[0072] For example, if Δ1 = 2 [s], Δ5 = 10 [m], and Δ6 = 1 [m], then in (2-2) above, there are 300 L2 norms L k (k=1, , 300) is obtained, and 30 L2 norm L k Therefore, there are 30 L2 norms L k one standard deviation σ d is calculated, and the result is a set of 10 standard deviations {σ d ;d=1,···,10} is obtained.

[0073] (2-4) And the set of standard deviations {σ d ;d=1,...,D}, the number of times that the standard deviation exceeded a predetermined upper limit and then fell below a predetermined lower limit, and the number of times that the standard deviation fell below a predetermined lower limit and then exceeded the predetermined upper limit, from the past Δ5 years to the present, are counted, and the sum of these is calculated as the score s. Note that these upper and lower limits are preset parameters and are determined, for example, based on empirical rules, etc.

[0074] For example, if the upper limit is sup and the lower limit is inf, and the standard deviations are arranged in chronological order, then D In this case, for example, for some d, sup<σ d After that, for some d' (>d), σ d'When it becomes <inf, this is counted as once. Similarly, for example, for a certain d, σ d After it becomes <inf, for a certain d' (>d), sup<σ d' When it becomes, this is counted as once.

[0075] Note that the above score s is the sum of the number of times the standard deviation between Δ6 of the L2 norm of the acceleration sensor value falls below the lower limit value after exceeding the upper limit value, and the number of times it exceeds the upper limit value after falling below the lower limit value. Therefore, when the cow is continuously making strong movements, the value becomes high. This is generally because when it is difficult to stand up, the cow struggles and thus has a tendency to continuously make strong movements (in other words, a tendency to thrash around continuously while lying down). In this sense, it can also be said that the score s is an index value representing the degree of thrashing of the cow during the most recent past Δ5.

[0076] Next, the score calculation unit 201 adds the score s calculated in step S303 above to the score S (step S304).

[0077] Next, the abnormality detection unit 202 determines whether the score S exceeds a predetermined threshold value th3 (step S305). Note that the threshold value th3 is a parameter set in advance, and its value can be set to any value as appropriate. For example, it can be set to about 15 to 35, and particularly preferably set to about 25.

[0078] If it is determined in step S305 that the score S exceeds a predetermined threshold th3, the abnormality detection unit 202 detects that an abnormality (i.e., difficulty in standing up) has occurred in the cow (step S306). The abnormality detection unit 202 may notify the abnormality detection result to, for example, a terminal (e.g., a smartphone used by the farmer) connected to the abnormality detection device 10 via a communication network. At this time, the abnormality detection result includes information such as the date and time when the abnormality occurred, information for identifying the cow in which the abnormality occurred (e.g., a tag ID, etc.), and information indicating the type of abnormality (e.g., a code indicating difficulty in standing up, etc.). This makes it possible for the terminal to, for example, issue an alert indicating that an abnormality has occurred in the cow.

[0079] If it is not determined in step S305 that the score S exceeds the predetermined threshold th3, the anomaly detection unit 202 does nothing.

[0080] <Effects of this embodiment> As described above, in the abnormality detection system 1 according to this embodiment, the cow's posture is classified into standing and lying posture, and the lying posture is further classified into prone and sideways posture, and the score S, which captures the continuous strong movement of the cow (i.e., the cow's continuous restlessness), is increased only when the cow is lying down. On the other hand, when the cow is standing, the score S is reset, and when the cow is lying down, the value of the score S is maintained. This prevents the score S from being increased for normal movements performed in a lying position (for example, shaking the head or scratching the body), thereby preventing the occurrence of false positives, and as a result, it becomes possible to realize highly accurate abnormality detection.

[0081] <Modification> In the above embodiment, the difficulty in standing up is detected based on the knowledge that, in general, when a cow has difficulty in standing up, the cow continues to make strong movements (i.e., the cow thrashes about in a lying position). However, if the cow is left in a state of difficulty in standing up despite the detection of an abnormality, the cow may become limp and the abnormality may not be detected thereafter. In addition, in very rare cases, there may be cows that have difficulty in standing up while remaining in a limp state.

[0082] Therefore, in order to be able to detect the above-mentioned difficulty in standing up in a lethargic state as an abnormality, the abnormality detection unit 202 may detect an abnormality, for example, when the duration that the cow remains lying down exceeds a predetermined threshold value th4. The duration that the cow remains lying down may be calculated from the date and time and the posture included in the posture data of the cow among the posture data stored in the posture data storage unit 600.

[0083] The present invention is not limited to the above-described embodiments specifically disclosed, and various modifications, changes, combinations with known technologies, etc. are possible without departing from the scope of the claims. [Explanation of symbols]

[0084] 1. Anomaly detection system 10. Anomaly detection device 100 Attitude estimation processing unit 101 Tag angle calculation unit 102 Basic angle calculation section 103 Posture estimation section 200 Abnormality detection processing unit 201 Score Calculation Department 202 Anomaly detection unit 300 Measurement data storage unit 400 Reference atmospheric pressure data storage unit 500 Tag angle data storage unit 600 Attitude data storage unit

Claims

1. An information processing device for determining whether an abnormality has occurred in livestock, a posture estimation unit that estimates a posture of the livestock based on measurement data measured by a sensor attached to the livestock; A score calculation unit that calculates a score representing a degree of abnormality of the livestock based on the posture and the measurement data; A determination unit that determines whether or not an abnormality has occurred in the livestock based on the score; An information processing device having the above configuration.

2. The score calculation unit is The information processing device according to claim 1, wherein the score is calculated depending on whether the posture is standing, indicating that the livestock is standing, lying, indicating that the livestock has its knees lowered and its chest raised, or lying, indicating that the livestock is lying on its side with its limbs stretched out.

3. The score calculation unit is The information processing apparatus according to claim 2 , wherein the score is calculated as a value representing a degree of violence of the livestock in the prone or recumbent state.

4. An information processing device for determining whether an abnormality has occurred in livestock, a posture estimation step of estimating a posture of the livestock based on measurement data measured by a sensor attached to the livestock; a score calculation step of calculating a score representing a degree of abnormality of the livestock based on the posture and the measurement data; a determination step of determining whether or not an abnormality has occurred in the livestock based on the score; An information processing method for performing the above.

5. An information processing device for determining whether an abnormality has occurred in livestock, a posture estimation step of estimating a posture of the livestock based on measurement data measured by a sensor attached to the livestock; a score calculation step of calculating a score representing a degree of abnormality of the livestock based on the posture and the measurement data; a determination step of determining whether or not an abnormality has occurred in the livestock based on the score; A program that executes the following.

Images