JP2016042843A - Mounting action detector, mounting action detection method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a novel technique capable of easily detecting a mounting action by a quadruped.SOLUTION: A mounting action detector, when determining a mounting action of a second quadruped different from a first quadruped by an infrared sensor attached to a back of the first quadruped, determines whether or not a reception period of data of the detection exceeds a predetermined threshold value according to the result of the detection and a detection timing of the infrared sensor, and determines as a mounting allowable action when the reception period exceeds the threshold value.SELECTED DRAWING: Figure 2A

Description

  The present invention relates to an animal riding behavior detection technique.

  Conventionally, various methods have been proposed for techniques for recognizing and discriminating animal behavior. For example, in Patent Document 1, in order to recognize human behavior, motion information such as acceleration, angular acceleration, velocity, angular velocity, position, rotation, and bending angle is combined with biological information such as pulse and blood pressure. Recognizing state changes is disclosed.

  Japanese Patent Application Laid-Open No. 2004-228561 discloses that in order to recognize a human action, the human action is determined using all the accelerations on the three orthogonal axes and the angular velocities around the three orthogonal axes. Moreover, in Patent Document 3, in order to accurately detect the estrus period without applying stress to the cow, the riding is detected by the inclination of the cow body, and between the cow to be boarded and the cow to be boarded. The use of infrared as communication is disclosed.

JP-A-10-113343 JP 2006-340903 A Japanese Patent Laid-Open No. 2005-210927

  However, the method of Patent Document 1 has problems such as complicated processing because operation information and ecological information are combined. In the method of Patent Document 2, the acceleration on three orthogonal axes and the angular velocities around the three orthogonal axes are sampled, and the behavior of the target is determined using all the sampled accelerations and angular velocities. Similarly, there is a problem that the processing becomes complicated. Furthermore, in the method described in Patent Document 3, it is necessary to specify the cow to be boarded and the cow to be boarded, and there is a problem that the processing becomes complicated.

  An object of this invention is to provide the novel technique which can detect the riding behavior of a cow easily.

  According to one aspect of the present invention, when the riding behavior of the second limb walking animal different from the first limb walking animal is detected by the infrared sensor attached to the back of the first limb walking animal, Based on the detection result and the detection timing of the infrared sensor, it is determined whether or not the reception period of the detection data exceeds a predetermined threshold value. There is provided a riding behavior detecting apparatus characterized in that it is determined as an allowable behavior. It is determined whether a reception period of the data exceeds a predetermined threshold value, and when it does not exceed the threshold value, it is determined as a riding avoidance action.

  Furthermore, an acceleration acquisition unit that acquires horizontal acceleration, an angular velocity acquisition unit that acquires a horizontal angular velocity, and an action pattern determination unit that determines a behavior pattern of a limb walking animal, the behavior pattern determination unit includes: , Based on the first value indicating the magnitude of the animal's longitudinal motion obtained based on the horizontal acceleration and the horizontal angular velocity, and on the horizontal acceleration and the horizontal angular velocity. The behavior pattern is discriminated based on the second value indicating the magnitude of the movement of the animal in the front-rear and left-right directions.

The first value is represented by | angular velocity (X direction) | / | acceleration (Y direction) | (formula 1), and the second value is represented by | angular velocity (X direction) | / | acceleration (Y direction) | × | Angular velocity (Y direction) | / | Acceleration (X direction) | (Expression 2)
However, the X direction is the animal side direction, and the Y direction is the animal front direction. It can be applied to biped walking animals (including humans) instead of limb walking.

  The behavior pattern determination unit is a positive correlation obtained when the logarithm of the first value and the logarithm of the second value obtained during the behavior of the animal are taken in a two-dimensional xy graph. Based on the above, the behavior pattern of the animal is discriminated. The behavior pattern determination unit depends on a behavior pattern obtained when a logarithm of the first value and a logarithm of the second value obtained at the time of animal behavior are taken in a two-dimensional xy graph. The behavior pattern of the animal is discriminated based on the range of xy to be performed.

  The present invention is also a terminal device attached to the back of a first limb walking animal, comprising an infrared sensor, a timer, and a communication unit, and the first limb walking animal is defined by the infrared sensor. When the riding behavior of a different second limb walking animal is detected, a terminal that transmits the detection result and the time of detection timing by the timer to a remote information processing apparatus as a set of data It is a riding action detection apparatus characterized by including an apparatus.

  According to another aspect of the present invention, when the riding behavior of the second limb walking animal different from the first limb walking animal is detected by the infrared sensor attached to the back of the first limb walking animal. Determining whether the detection data reception period exceeds a predetermined threshold based on the detection result and the detection timing of the infrared sensor; and There is provided a method for detecting a riding behavior characterized in that the method includes a step of determining that the behavior is a riding tolerance behavior. Furthermore, it is characterized by determining whether or not a reception period of the data exceeds a predetermined threshold value, and when the data reception period does not exceed the threshold value, it is determined to be a riding avoidance action.

  The present invention may be a program for causing a computer to execute the behavior determination method described above, or a computer-readable recording medium for recording the program.

  According to the present invention, it is possible to easily and accurately detect the riding behavior of the limb walking animal based on the sensing data.

It is a functional block diagram which shows the example of 1 structure of the system containing the action determination apparatus by embodiment of this invention. It is a figure which shows a mode that the 1st apparatus (terminal device) was mounted on the flat back part of a cow. It is a functional block diagram which shows the 1st system configuration example. It is a functional block diagram which shows the 2nd system configuration example. It is a functional block diagram which shows the 3rd system configuration example. It is a functional block diagram which shows the 4th system configuration example. FIG. 10 is a functional block diagram illustrating a fifth system configuration example. It is a functional block diagram which shows the 6th system configuration example. It is a figure which shows the meaning of (1) Formula. It is a figure which shows the meaning of (1) Formula. It is a figure which shows the meaning of (2) Formula. It is a figure which shows the meaning of (2) Formula. FIG. 11 is a flowchart illustrating an example of information processing flow in the information processing apparatus. It is a flowchart figure which shows the flow of the riding behavior detection process in information processing apparatus. It is a flowchart figure which shows the flow of the prediction process of an estrus start. It is a figure which shows the example of the discrimination | determination data shown taking the cow as an example about an example of the data for determining the action pattern of an animal. It is a figure which shows the example of the discrimination | determination data shown taking the cow as an example about an example of the data for determining the action pattern of an animal. It is a figure which shows the example of the discrimination | determination data shown taking the cow as an example about an example of the data for determining the action pattern of an animal. It is a figure which shows the example of the discrimination | determination data shown taking the cow as an example about an example of the data for determining the action pattern of an animal. It is a figure which shows the example of the discrimination | determination data shown taking the cow as an example about an example of the data for determining the action pattern of an animal. It is a figure which shows the example of the discrimination | determination data shown taking the cow as an example about an example of the data for determining the action pattern of an animal. FIG. 13 is a diagram corresponding to FIG. 12, and in addition to FIG. 12, by using the method for discriminating whether to allow riding or to refuse riding according to the second embodiment, FIG. It is a figure which shows the example of a recording of the utterance time of the riding tolerance action in the cow riding detection technique by the 2nd Embodiment of this invention. It is a figure which shows an example of the detection rate of the riding behavior shown in FIG.

  Hereinafter, an animal behavior discrimination technique according to an embodiment of the present invention will be described in detail with reference to the drawings. In the following, a technique for discriminating the behavior of a cow is taken as an example, but the present invention is not limited to a cow as long as it is a limb walking animal that walks on the limb. Examples of classification include livestock.

  In Japan, cows are made pregnant by artificial insemination and embryo transfer, except for some meat varieties. Since these breeding techniques are based on the date of estrus, it is necessary to understand the estrous behavior of cattle. In particular, when artificial insemination is performed, accurate estimation of the optimal fertilization period is important for improving the fertility rate, and it is necessary to grasp the estrus start and end times and the time zone of the estrus maximum. Estrus behavior in cattle, especially the duration of tolerance for riding, is as short as 14-21 hours, whereas estrus behavior only occurs at 21-day intervals, so the estimated loss due to one missed estrus is tens of thousands of yen It also extends. Therefore, finding estrus reliably is extremely important for stable calf production and good breeding management, and a technique for discriminating behavior that can estimate estrus with an inexpensive sensor or the like is required.

  Conventionally, there are two types of methods for discriminating cattle behavior using sensors: (1) a method for monitoring the number of steps, and (2) a method for monitoring the amount of behavior.

(1) Method of monitoring changes in the number of steps Ushiho (registered trademark of Comtech) has been commercialized, and a pedometer that also serves as a transmitter is attached to the forelimbs to receive step data in one-hour units via PC communication. This system monitors changes in the number of steps.
(2) Method of monitoring changes in behavior amount Doctor Cowbell (registered trademark of My Media) and Hatsuhatsu (PA Technology) are commercialized. Dr. Cowbell uses a vibration sensor on the cow's neck and Hatsuhatsu uses an acceleration sensor. It is a device that attaches a used behavior meter and monitors changes in behavior and body temperature.

  In these methods, the behavior of the animal cannot be determined as “reliable recumbent”, “recumbent or standing upright”, “with motion”, “moving”, or the like. In addition, the increase in the number of steps and the amount of action is not necessarily estrus specific, and high-precision prediction of estrus cannot be applied.

(First embodiment)
In the present embodiment, the behavior of the animal can be discriminated as “reliable recumbent”, “recumbent or standing upright”, “with motion”, and “moving” with a simple configuration.

  FIG. 1A is a functional block diagram illustrating a configuration example of a system including a behavior determination device according to an embodiment of the present invention. The behavior determination device includes at least a part of an information processing device including a behavior pattern determination unit.

  The system A according to the present embodiment includes, for example, a device (1) 1 such as a terminal device and a remote device (2) 3 such as an information processing device. FIG. 1B is a diagram showing a state in which the terminal device (device (1)) 1 is placed on the flat back portion of the cow. The mounting position of the terminal device 1 is a position on the back of a limb walking animal such as a cow that satisfies the conditions such as “a position that cannot be touched by the nose of the individual to which the apparatus is attached”, “a position where the installation is as stable as possible” It is. Here, the terminal device is also referred to as a first device (1) and refers to a device that is placed on the back of a cow.

  Hereinafter, various configuration examples of the system (behavior determination device) A according to the present embodiment will be exemplarily described. FIG. 1A basically corresponds to the configuration of FIGS. 2B to 2F below. All the functional units described below may not be required. In addition, each component in the first to third devices may be provided in any case where it can be replaced.

  The acceleration sensor 1-1 is a biaxial or triaxial acceleration sensor, and the angular velocity sensor 1-2 is also a biaxial or triaxial angular velocity sensor. In the case of a triaxial sensor, each sensor is installed so that the X axis is in the cow side (lateral) direction, the Y axis is in the cow head direction, and the Z axis is in the upward direction (vertical direction). Hereinafter, they are referred to as an X direction, a Y direction, and a Z direction. Therefore, in this embodiment, it is a requirement that the sensor is a biaxial sensor. The infrared sensor 1-3 can determine the presence or absence of an object in the upward direction. The memory | storage part 1-4 memorize | stores the program for controlling each function part by CPU, the sensing result by a sensor, etc. FIG. Further, in addition to the arithmetic processing program, action determination data 1-4-1 for determining action is stored. The control unit 1-5 is a CPU that controls each functional unit. The timer 1-6 provides the time when the sensing data is acquired, and can associate the time with the sensing data. The calculation unit 1-7 creates determination data 1-4-1 for action determination based on data such as sensing data and time.

  The action pattern determination unit 1-8 refers to the determination data 1-4-1 for action determination based on the values calculated by the calculation unit 1-7 based on the sensing data, etc., and determines the behavior pattern of the animal. judge. The display unit 1-10 displays the obtained action pattern and the like. Instead of the display unit or together with the display unit, an output unit for printing data and results may be provided.

FIG. 2B is a functional block diagram illustrating a second system configuration example of the behavior determination apparatus (system).
As shown in FIG. 2B, the second system includes a device (1) 1 and a device (2) 3. For example, the device (1) 1 is a terminal device attached to a cow, and the device (2) 3 is a terminal device such as a tablet that performs output.
The device (1) 1 includes an acceleration sensor 1-1 to a behavior pattern determination unit 1-8 and a communication unit 1-9. The device (2) 3 includes a communication unit 3-1 and a display unit 3. -2 is provided. The communication unit 1-9 and the communication unit 3-1 exchange data and the like between the device (1) 1 and the device (2) 3. Other processing is basically the same as that of the first system.

FIG. 2C is a functional block diagram illustrating a third system configuration example of the behavior determination device (system).
As shown in FIG. 2C, the third system includes a device (1) 1 and a device (2) 3. For example, the device (1) 1 is a terminal device attached to a cow, and the device (2) 3 is a terminal device such as a tablet that performs output.
The device (1) 1 includes an acceleration sensor 1-1 to a timer 1-6 and a communication unit 1-9. The device (2) 3 includes a communication unit 3-1, a storage unit 3-3, and a control unit. A unit 3-4, a timer 3-5, a calculation unit 3-6, an action pattern determination unit 3-7, and a display unit 3-2 are provided. Determination data 3-3-1 is stored in the storage unit 3-3.
The communication unit 1-9 and the communication unit 3-1 exchange data and the like between the device (1) 1 and the device (2) 3.

  In the third system, sensing data and time are acquired on the device (1) 1 side, and calculation processing and behavior determination processing are performed on the device (2) 3 side. Other processing is basically the same as that of the first system.

FIG. 2D is a functional block diagram illustrating a fourth system configuration example of the behavior determination apparatus (system).
As shown in FIG. 2D, the fourth system includes a device (1) 1, a device (2) 3, and a device (3) 5. For example, the device (1) 1 is a terminal device attached to a cow, the device (2) 3 is a relay device such as an information processing device that performs output, and the device (3) 5 is a terminal such as a tablet that performs output. Device.
The device (1) 1 includes an acceleration sensor 1-1 to a timer 1-6 and a communication unit 1-9. The device (2) 3 includes a communication unit 3-1, a storage unit 3-3, and a control unit. 3-4, timer 3-5, calculation unit 3-6, behavior pattern determination unit 3-7, communication unit 3-8, device (3) 5, communication unit 5-1 and display unit 5-2 Is provided. Determination data 3-3-1 is stored in the storage unit 3-3. The communication unit 1-9 and the communication unit 3-1 exchange data and the like between the device (1) 1 and the device (2) 3. The communication unit 3-8 and the communication unit 5-1 exchange data and the like between the device (2) 3 and the device (3) 5.

  In the fourth system, sensing data and time are acquired on the device (1) 1 side, and calculation processing and action determination processing are performed on the device (2) 3 side. Output such as display is performed by the device (3) 5. Other processing is basically the same as that of the first system.

FIG. 2E is a functional block diagram illustrating a fifth system configuration example of the behavior determination device (system).
As shown in FIG. 2E, the fifth system includes a device (1) 1 and a device (2) 3. For example, the device (1) 1 is a terminal device attached to a cow, and the device (2) 3 is a terminal device such as a tablet that performs output.
The device (1) 1 includes the acceleration sensor 1-1 to the calculation unit 1-7 and the communication unit 1-9. The device (2) 3 includes the communication unit 3-1, the storage unit 3-3, and the control. A unit 3-4, a timer 3-5, an action pattern determination unit 3-7, and a display unit 3-2 are provided. Determination data 3-3-1 is stored in the storage unit 3-3. The communication unit 1-9 and the communication unit 3-1 exchange data and the like between the device (1) 1 and the device (2) 3.

  In the fifth system, the processing is performed after the sensing data and time are acquired on the device (1) 1 side, and the calculation result is received on the device (2) 3 side, and the behavior determination processing is performed. Other processing is basically the same as that of the first system.

FIG. 2F is a functional block diagram illustrating a sixth system configuration example of the behavior determination apparatus (system).
As shown in FIG. 2F, the sixth system includes a device (1) 1, a device (2) 3, and a device (3) 5. For example, the device (1) 1 is a terminal device attached to a cow, the device (2) 3 is a relay device such as an information processing device that performs output, and the device (3) 5 is a terminal such as a tablet that performs output. Device.
The device (1) 1 includes the acceleration sensor 1-1 to the calculation unit 1-7 and the communication unit 1-9. The device (2) 3 includes the communication unit 3-1, the storage unit 3-3, and the control. Unit 3-4, timer 3-5, action pattern determination unit 3-7, communication unit 3-8, and device (3) 5 are provided with communication unit 5-1 and display unit 5-2. . Determination data 3-3-1 is stored in the storage unit 3-3.
The communication unit 1-9 and the communication unit 3-1 exchange data and the like between the device (1) 1 and the device (2) 3. The communication unit 3-8 and the communication unit 5-1 exchange data and the like between the device (2) 3 and the device (3) 5.

  In the sixth system, calculation processing is performed after the sensing data and time are acquired on the device (1) 1 side, and the operation determination processing is performed on the device (2) 3 side by receiving the calculation processing result. Output such as display is performed by the device (3) 5. Other processing is basically the same as that of the first system.

  Various other variations are included in the present invention.

FIG. 3A to FIG. 3D are diagrams showing an image of cow behavior determination. In the following, the contents of the processing will be described by taking the configuration as shown in FIGS. 1A and 2C as an example, but the subject of the processing may be changed in accordance with the change in the configuration from FIGS. 2A to 2F.
FIG. 4 is a flowchart illustrating an example of a flow of information processing in the information processing apparatus 3. FIG. 7 to FIG. 12 are diagrams showing examples of discrimination data 3-3-1 showing an example of data for determining an animal behavior pattern using a cow as an example.

  First, an example of the discrimination data 3-3-1 will be described in detail. Such discrimination data can be obtained, for example, by comparing sensing data from the terminal device 1 obtained by using the system according to the present embodiment with the behavior of the cow photographed by a camera or the like. Once such data can be obtained, it can be used continuously in subsequent action discrimination. The data may be corrected as appropriate according to the breeding environment.

A device (1) 1 having a triaxial acceleration sensor 1-1 and a triaxial angular velocity sensor (installed so that the Y axis of each sensor is in the head direction and the Z axis is in the upward direction) 1-2 (FIGS. 1A and 2C) The behavior pattern of a cow is analyzed by analyzing the acceleration (X • Y: m / s ² ) and angular velocity (X • Y: deg / s) obtained from the device (1) 1 by the following procedure. Can be determined as “reliable recumbent”, “recumbent or standing upright”, “with motion”, and “moving”. In addition, as described later, it can be applied to prediction of estrus if it is attached to “the position of the back of the cow that can detect the riding behavior of other cows”. Although a 3-axis sensor is used here, it is not necessary to use data in the Z-axis direction, and a 2-axis sensor can be used.

The behavior discrimination method is shown below using the case of cattle as an example.
1) As a method for discriminating the behavior of a cow, formula (1) is used as a value indicating the magnitude of movement (front-rear direction).
Value indicating magnitude of motion (front-rear direction: x value) = | Angular velocity (X direction) | / | Acceleration (Y direction) |
The obtained value is set as the first value.

2) As a method for discriminating the behavior of the cow, (2) was used as a value indicating the magnitude of the movement (front-rear / left-right direction: y value).
Value indicating the size of the motion (front / back / left / right) =
| Angular velocity (X direction) | / | Acceleration (Y direction) | × | Angular velocity (Y direction) | / | Acceleration (X direction) |
The obtained value is set as the second value.

The meaning of action discrimination will be described in detail below.
In equation (1), | angular velocity (X direction) | (unit: deg / s) means that the installation position of the terminal device 1 is P1, and as shown in FIG. (Symbol 11).

Further, | acceleration (Y direction) | (unit: m / s ² ) in the expression (1) means that if the installation position of the terminal device 1 is P1, the position P1 is as shown in FIG. 3A (b). This is the acceleration toward the front and rear 13a and 13b.

The concept of the equation (1) is as follows (dimensional transformation).
a1) The “speed to tilt back and forth” per “back and forth acceleration” is
a2) deg / s ÷ m / s ² = deg · s / m
a3) Since s is analyzed at intervals of 1 second, it is canceled,
a4) The unit is deg / m.
a5) That is, the expression (1) is the change angle of the back line per movement amount in the front-rear direction.

As shown in FIG. 3B, the change angle of the back line per movement amount in the front-rear direction may be considered to be the inclination θ ₁ (pitch) of the back line L1 per movement amount 13a, 13b in the straight traveling direction.

On the other hand, the value indicating the magnitude of motion in the formula (2) (front / rear / left / right direction) = | angular velocity (X direction) | / | acceleration (Y direction) | × | angular velocity (Y direction) | / | acceleration (X direction) ) | In the second term, | angular velocity (Y direction) | (unit: deg / s) is a velocity that tilts left and right around the position P1, as shown in FIG. 3C (a). .

Also, | acceleration (X direction) | (unit: m / s ² ) is acceleration to the left and right 17a and 17b with the position P1 as the center, as shown in FIG. 3C (b).

The concept of equation (2) is as follows (dimensional transformation).
b1) The “velocity to tilt left / right” per “acceleration to the left / right” is
b2) deg / s ÷ m / s ² = deg · s / m
b3) Since s is analyzed at intervals of 1 second, it is canceled,
b4) The unit is deg / m.
b5) That is, the second term of the equation (2) is the rotation angle of the back axis per the movement amount in the left-right direction.
The rotation angle of the dorsal line axis per movement amount in the horizontal direction, as shown in FIG. 3D, a rotation angle (roll) θ ₂ of the back line axis L3 per the lateral direction movement amount 17a · 17b.

Expression (2) = Change angle of the back line / movement amount in the front-rear direction × rotation angle of the back line axis / movement amount in the left-right direction. here,
・ Change angle of the back line: Recumbent <standing <movement / movement amount in the front / rear direction: Recumbent ≤ standing << movement, rotation angle of the back line axis: recumbent ≤ standing << moving amount in the left / right direction: recumbent <standing <movement Therefore, the behavior of the animal can be determined from the change angle of the back line, the rotation angle of the back line axis, the movement amount in the front-rear direction, and the movement amount in the left-right direction.

3) The relationship between the values obtained by logarithmically converting the first value x and the second value y, which are the calculated values according to the equations 1 and 2, is classified into the sensor values at the time of lying, standing, and moving. As shown in FIG. 7, it was found to be in a positive linear relationship. According to FIG. 7, it can be seen that the behavior of the cow tends to increase as the x value (horizontal axis of the graph) and the y value (vertical axis of the graph) increase. In the y <1 motion (small) region, cows' behavior patterns have a lot of recumbency and standing, and in y> 1 motions (large), the cow's behavior patterns have a lot of movement.

4) As shown in FIG. 8, according to the frequency distribution obtained by logarithmically converting the value obtained from the equation (2) at the time of standing action, 99% or more standing actions are distributed in the range of y ≧ 0.02850. It can be seen that is less than that.

5) As shown in FIG. 9, from the frequency distribution obtained by logarithmically converting the value obtained from the equation (2) at the time of the moving action, 90% or more of the moving action is distributed in the range of y> 1.885739. It can be seen that is less than that.

6) As shown in FIG. 10, from the frequency distribution obtained by logarithmically converting the value obtained from the expression (2) at the time of recumbent action, 90% or more recumbent actions are distributed in the range of y <1.887018. It can be seen that the operation, turning over, and grooming are more than that.

7) As shown in FIG. 11, about 99% of recumbent behavior is distributed in the range of x <4.4420 from the frequency distribution obtained by logarithmically converting the value obtained by the equation (1) at the time of recumbent behavior. It turns out that it becomes the above.

8) From the results of FIGS. 8-11, as shown in FIG. 12, by using the formulas (1) and (2), the behavior of the cow is changed to “reliable recumbent”, “recumbent or standing upright”, “ It can be classified almost certainly into “operation present” and “movement”.

That is, the determination conditions are as follows.
“Reliable Recumbent”: y <0.03
“Recumbent or standing upright”: 0.03 ≦ y ≦ 2.0
“With motion”: y> 2.0, x ≦ 4.5
“Move”: y> 2.0, x> 4.5
(3) Formula

  A process for discriminating an action using the discrimination data 3-3-1 will be described in detail with reference to FIGS. 2C and 4. First, the process is started (step S1), and in step S2, the communication unit 3-1 of the device (2) 3 receives the acceleration sensor 1-1 and the angular velocity sensor obtained from the communication unit 1-9 of the device (1) 1. Sensing data α and ω in 1-2 are received. In step S3, the computing unit 3-3 calculates the magnitude x of the motion in the front-rear direction and the magnitude y of the motion in the front-rear / left-right direction based on the above formulas (1) and (2). The first value, the second value, etc. are calculated.

Next, in step S4, the discrimination data 3-3-1 in the storage unit 3-3 is read and referenced.
Next, in step S5, the behavior pattern determination unit 3-7 determines the behavior pattern of the cow from the calculation data of the calculation unit 3-6 and the discrimination data 3-3-1. For example, the behavior pattern can be determined by the above equation (3) depending on which region in FIG.

  In step S6, the determination result is output by the display unit 3-2. In step S7, when the process is ended (Yes), the process proceeds to step S8 (end), and when the process is continued, the process returns to step S2.

  With the above processing, it is possible to accurately determine the behavior pattern of the cow based on the sensing data. Unlike observation with a camera or the like, there is an advantage that determination can be made automatically. In addition, since it is possible to accurately determine the behavior even if the amount of sensing data is reduced, the burden of information processing can be reduced.

(Second Embodiment)
Next, a behavior discrimination technique according to the second embodiment of the present invention will be described.
In Japan, except for some meat varieties, cows are made pregnant by artificial insemination and embryo transfer. Since these breeding techniques are based on the date of estrus, it is essential to understand the estrous behavior of cattle. In particular, in artificial insemination, accurate estimation of the optimal fertilization period is important in order to improve the fertility rate, and it is necessary to grasp the estrus start and end times and the time zone of the estrus maximum. In the estrus cow, as a specific behavior, it shows the tolerance of riding that accepts the riding of the back of other cows and stops. In addition, although not specific, riding to other cattle, increased activity, isolation from the group, gun call and flaming up the upper lip, etc. were observed, hyperemia of the vulva and mucus outflow from the vulva Accompanied by. Furthermore, estrus behavior in cattle, especially the duration of riding tolerance, is as short as 14 to 21 hours, whereas estrus behavior occurs only at 21-day intervals, so there is an estimated loss of tens of thousands due to missed estrus once. It extends to a circle. Therefore, finding estrus reliably is extremely important for stable calf production and good breeding management.

  At production sites, reports from owners such as “difficult to find estrus” and “no cattle estrus” are increasing. As a major factor related to the decrease in conception rate, it has been pointed out that in recent breeding cattle, there is a possibility that shortening of estrus duration, weak increase of estrus behavior, etc. are occurring. Therefore, in order to overcome the current situation where the conception rate continues to decline, the development of a new estrus discovery technology that can grasp estrus behavior with high accuracy is expected.

  As a method for detecting the riding behavior, a heat mount detector (Kamar, USA) as a method using a pad and a tail paint (KIWIKIT, New Zealand) as a method using a paint are sold both in Japan and overseas. The heat mount detector has a structure in which a red ink tube is loaded in a white plastic pad and is attached to the back side of the waist with an adhesive. In cows that show that they are allowed to ride, ink leaks out of the tube due to excessive weight when riding from other cows, and the pad turns red. Tail paint, on the other hand, is a fluorescent paint that is applied to the back of the waist. In cattle that have been allowed to ride, the paint applied is peeled off due to strong contact with the cattle that have been boarded.

  Although it is possible to confirm that any of the methods has allowed riding behavior, it is not possible to accurately estimate the start or end time of the behavior, which is essential information for determining the appropriate period of insemination. Further, it is necessary to observe the state of the pad and the paint at least twice a day for 30 minutes or more, and the decrease in the observation frequency causes a great decrease in the accuracy of estrus detection. Furthermore, since either method is a method for finding an estrus cow that is restricted to a boarding-allowed behavior, the effectiveness cannot be expected for a faint or estrus cow. In addition to these problems, there are the following problems for each method.

The problems of the heat mount detector are as follows.
It is a disposable instrument and cannot be used repeatedly.
Under hot weather, it may change red due to rupture due to an increase in ink tube internal pressure.
Individuals who repeatedly carry riding allowance frequently, may be dropped or disappear due to bird damage such as crows or tolerant behavior is weak.

The problems with tail paint are as follows.
The determination of whether or not to ride is performed by classifying the state where the paint is peeled into five stages, and therefore lacks objectivity.
It is difficult to apply and judge in individuals with severely dirty cows.

  In the present embodiment, as shown in FIGS. 2A to 2F (here, FIG. 2C is taken as an example), the device (1) 1 is provided by the infrared sensor 1-3 provided in the device (1) 1. The movement of the second cow different from the first cow wearing is detected.

  For example, the movement of the second cow (other cow) in the vicinity of the back (cross part) of the cow by the infrared sensor 1-3 attached to the back, for example, riding behavior that is a sign of estrus development is detected. In more detail, conditions such as “the position of the back of the cow that can detect the riding behavior of other cows”, “the position where the individual's nose attached to the device cannot be touched”, “the position where the installation is as stable as possible” By mounting at a position that satisfies the condition, the movement of other cows in the vicinity of the wearing cow cross is sensed, and riding behavior that is a sign of estrus development is detected. Then, the detection result is notified wirelessly to a device (2) 3 installed in a ranch management center that manages a ranch including a remote information processing device, for example. By this radio, the estrus of cattle can be known at the ranch management center.

  FIG. 5 is a flowchart showing the flow of the riding behavior detection process in the device (2) 3. These processes may be performed by, for example, the calculation unit 3-6, the action pattern determination unit 3-7, and the like, or may be performed by other processing units. Here, a case where detection is performed by the calculation unit 3-6 and the behavior pattern determination unit 3-7 will be described as an example.

  As shown in FIG. 5, in the device (1) 1, the presence of the second cow based on the riding behavior of the second cow by the infrared sensor 1-3 is detected, and the communication unit 1 of the device (1) 1 is detected. When the data is sent from -9, the process in the device (2) 3 is started (step S11), and the communication unit 3-1 detects the detection data and the detection timing of the cow based on the infrared rays sent from the device (1) 1. And the time in the timer 1-6 indicating (step S12). The reception of data may be started at the reception timing or may be continuously performed.

  Here, in step S13, whether the continuous reception period from the start to the end of data reception by the infrared sensor 1-3 is a threshold value for distinguishing between cow riding and avoidance, for example longer than 3 seconds? Determine whether or not. If it is longer than 3 seconds (Yes in step S13), the process proceeds to step S14, and the detected data is determined to be a riding allowable action. The threshold value can be adjusted as appropriate.

Next, in step S16, the determination results of steps S14 and S15 are output. In step S17, it is determined whether or not to end the process, and either the process ends (step S18) or the process continues (step S12). Proceed to
Then, the detection result is wirelessly notified to a remote device (2) 3 installed in, for example, a ranch management center that manages the ranch. By this radio, the estrus of cattle can be known at the ranch management center.

  As described above, in the behavior determination method according to the present embodiment, the riding behavior of a cow can be detected using infrared data. In addition, it is possible to determine whether the behavior is a riding allowance behavior or a riding rejection behavior.

FIG. 14 is a diagram showing the time data (horizontal axis) based on the timer 1-7 and the detection result (vertical axis) of the riding tolerance behavior by the infrared sensor 1-3, and the system A according to the present embodiment. It is a figure which shows the example of a recording of the utterance time of a riding tolerance action. This is data obtained by outputting the result of step S14 in FIG. As shown in FIG. 14, by using the system according to the present embodiment, regardless of whether it is within the working hours or outside the working hours, the work is engaged like a white plot that cannot be found in the past. You will also be able to find out-of-hours riding tolerance behavior. For example, in the output step of the determination result in step S16, data as shown in FIG. 14 is sequentially recorded, and when it is outside the working hours, for example, an emergency center (ranch management center) or a person in charge is notified. Can go.
Therefore, it can be discovered with high accuracy without missing the estrus time of the cow.

  FIG. 15 is a diagram illustrating an example of the detection rate of riding behavior in the system according to the present embodiment. In this figure, the estrus cattle equipped with the terminal device 1 having the pyroelectric infrared sensor 1-3 are released to the paddock together with the other 10 cows. At night, there are three cows and three cows. When comparing the riding behavior monitoring by the infrared sensor 1-3 and the monitoring by video recording, the detection rates are both 100%, and even in the case of detection by the infrared sensor, it has high detection accuracy as in the case of video monitoring. Take it.

(Third embodiment)
Next, an action discrimination technique according to the third embodiment of the present invention will be described.
In the present embodiment, cattle riding permission is utilized by using the behavior discriminating method according to the first embodiment and the discriminating method between cattle riding permission behavior and riding rejection behavior according to the second embodiment. The action and the no-ride-no-behavior action are easily determined.

  FIG. 13 is a diagram corresponding to FIG. 12, and in addition to FIG. 12, using the method for discriminating whether or not to allow riding according to the second embodiment, cattle riding tolerance and avoidance It is a plot of behavioral data. From this figure, cattle riding tolerance behavior is expressed as follows.

  The data on the cow's riding tolerance behavior and repelling behavior are, for example, the angular velocity (X direction) between the time when the reception of infrared data is started and the time when the reception of infrared data is finished in step S12 of FIG. Using the average values of acceleration (Y direction), angular velocity (Y direction), and acceleration (X direction), the relationship of (Expression 1) and (Expression 2) is obtained.

  From FIG. 13, the riding tolerance behavior is distributed in x> = 4.5 (movement area) and y> 2.0 (with movement or movement area), and the riding avoidance action is 1 <x <4. 5 (with operation) and y> 2.0 (with operation or moving range).

It is possible to determine whether the obtained value is the cow riding tolerance determination area or the riding avoidance determination area in FIG. 13 and to know whether the behavior is cow riding tolerance or riding avoidance or otherwise. This determination processing may be performed, for example, when the infrared data in step S12 in FIG. 5 is received.
According to the present embodiment, it is possible to easily know whether a cow is allowed to ride, avoid riding, or otherwise, using an animal behavior discrimination method.

(Fourth embodiment)
Next, an action discrimination technique according to the fourth embodiment of the present invention will be described.
Estrous cows have lower lying and standing times and longer travel times than non-estrous cows. Therefore, the start of estrus can be predicted by grasping the expression rate of behavior per unit time by using the behavior pattern discrimination method according to the first embodiment of the present invention.

  From step S5 in FIG. 4, as shown in FIG. 6, in step S21, it is determined whether or not the lying / standing time has decreased. If yes, the process proceeds to step S22, and the travel time has been extended. Determine. If “Yes” in step S22, the process proceeds to step S23 to predict the start of estrus, and output that effect. If No, the process returns to step S21.

  As described above, according to the present embodiment, it is possible to detect the estrus of cattle based on the lying / standing time and the moving time by using the behavior determination method of the first embodiment or the like. it can.

  According to the present embodiment, by using the acceleration sensor, the angular velocity sensor, and the present determination method, it is possible to accurately determine “reliable recumbent”, “recumbent or stationary standing”, “with motion”, and “movement”. be able to. In addition, it is possible to predict the start of estrus by grasping the expression rate of behavior per unit time.

  Among the estrous behaviors, the “ride riding allowed behavior” that is an index of the start of estrus can be determined as “ride riding avoidance behavior” or “no riding failure”. By applying the formulas (1) and (2) to animals other than cattle, it is possible to discriminate behavior.

  That is, although this embodiment has been described by taking a cow as an example, the behavior pattern can be determined for other animals in the same manner. For example, horses only have different parameter values, and parameters may be obtained by the same method.

  The processing and control can be performed by software processing by CPU (Central Processing Unit) or GPU (Graphics Processing Unit), ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate) hardware hardware.

  In the above-described embodiment, the configuration and the like illustrated in the accompanying drawings are not limited to these, and can be changed as appropriate within the scope of the effects of the present invention. In addition, various modifications can be made without departing from the scope of the object of the present invention.

Each component of the present invention can be arbitrarily selected, and an invention having a selected configuration is also included in the present invention.
In addition, a program for realizing the functions described in the present embodiment is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into a computer system and executed to execute processing of each unit. May be performed. The “computer system” here includes an OS and hardware such as peripheral devices.
Further, the “computer system” includes a homepage providing environment (or display environment) if a WWW system is used.

  The “computer-readable recording medium” refers to a storage device such as a flexible medium, a magneto-optical disk, a portable medium such as a ROM and a CD-ROM, and a hard disk incorporated in a computer system. Furthermore, the “computer-readable recording medium” dynamically holds a program for a short time like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line. In this case, a volatile memory in a computer system serving as a server or a client in that case, and a program that holds a program for a certain period of time are also included. The program may be a program for realizing a part of the above-described functions, or may be a program that can realize the above-described functions in combination with a program already recorded in a computer system. At least a part of the functions may be realized by hardware such as an integrated circuit.

  The present invention can be used for an action discrimination device.

A ... system, 1 ... device (1), 1-1 ... acceleration sensor, 1-2 ... angular velocity sensor, 1-3 ... infrared sensor, 1-4 ... storage unit, 1-5 ... control unit, 1-6 ... Timer, 1-7... Arithmetic unit, 1-8... Action determination unit, 1-9 .. communication unit, 1-10... Display unit (output), 3 .. device (2), 3-1. 2 ... display unit, 5 ... device (3), 5-1 ... communication unit, 5-2 ... display unit.

Claims (11)

  When the riding behavior of the second limb walking animal different from the first limb walking animal is detected by the infrared sensor attached to the back of the first limb walking animal, the detection result and the infrared sensor Based on the detection timing, it is determined whether or not a reception period of the detection data exceeds a predetermined threshold value, and when the detection data exceeds the threshold value, it is determined that the behavior is allowed for riding. Norigami action detection device.   2. The multiplication according to claim 1, wherein it is determined whether a reception period of the data exceeds a predetermined threshold value, and when the data reception period does not exceed the threshold value, it is determined as a riding avoidance action.駕 Behavior detection device. further,
An acceleration acquisition unit that acquires acceleration in the horizontal direction, an angular velocity acquisition unit that acquires the angular velocity in the horizontal direction, and an action pattern determination unit that determines the behavior pattern of the limb walking animal,
The behavior pattern determination unit includes a first value indicating a magnitude of an animal's longitudinal motion obtained based on the horizontal acceleration and the horizontal angular velocity, the horizontal acceleration, and the horizontal direction. 3. The riding behavior detection according to claim 1, wherein the behavior pattern is discriminated based on a second value indicating the magnitude of the movement of the animal in the front-rear and left-right directions obtained based on the angular velocity of the animal. apparatus. The first value is given by | angular velocity (X direction) | / | acceleration (Y direction) |
The second value is obtained by | an angular velocity (X direction) | / | acceleration (Y direction) | × | angular velocity (Y direction) | / | acceleration (X direction) | The riding behavior detection device according to claim 3.
However, the X direction is the animal side direction, and the Y direction is the animal front direction. The behavior pattern determination unit
Based on the positive correlation obtained when the logarithm of the first value and the logarithm of the second value obtained at the time of the behavior of the animal are taken on a two-dimensional xy graph, the behavior of the animal The riding behavior detection apparatus according to claim 4, wherein the pattern is discriminated. The behavior pattern determination unit
The logarithm of the first value and the logarithm of the second value obtained at the time of the behavior of the animal are in the range of xy depending on the behavior pattern obtained when taking the two-dimensional xy graph. 5. The riding behavior detection device according to claim 4, wherein the behavior pattern of the animal is determined based on the behavior pattern.   A terminal device attached to the back of a first limb walking animal, comprising an infrared sensor, a timer, and a communication unit, and a second limb walking different from the first limb walking animal by the infrared sensor. A terminal device that, when an animal's riding behavior is detected, includes a result of the detection and a time of detection timing by the timer as a set of data and transmits the data to a remote information processing device by the communication unit; Norigami behavior detection device. When the riding behavior of the second limb walking animal different from the first limb walking animal is detected by the infrared sensor attached to the back of the first limb walking animal, the detection result and the infrared sensor Determining whether a reception period of the detection data exceeds a predetermined threshold based on detection timing; and
A method of detecting a riding behavior when the threshold value is exceeded.   9. The method according to claim 8, further comprising the step of determining whether or not a reception period of the data exceeds a predetermined threshold value, and determining that the behavior is to avoid riding when the reception period does not exceed the threshold value. The described riding behavior detection method.   A program for causing a computer to execute the riding behavior detection method according to claim 8 or 9.   A computer-readable recording medium for recording the program according to claim 10.

Images