JP2021035355A - Health condition management system for livestock, wearable device for livestock, health condition management method for livestock, and program - Google Patents

Abstract

To provide a technique capable of accurately detecting a health condition of livestock to be managed with a simple system configuration.SOLUTION: A health condition management system for livestock according to the present invention comprises: detection means including a temperature sensor and an acceleration sensor, fitted to a livestock to be managed, and acquiring body surface temperature information of the livestock, posture information of the livestock, and activity amount information of the livestock; and diagnosis means for diagnosing a health condition of the livestock on the basis of the body surface temperature information of the livestock, the posture information of the livestock, and the activity amount information of the livestock.SELECTED DRAWING: Figure 2

Description

The present invention relates to a livestock health condition management system, a livestock wearable device, a livestock health condition management method, a program, and the like.

In recent years, while the number of livestock workers is decreasing and the population is aging, the number of livestock raised per household is increasing, and it is becoming difficult to devote sufficient observation time to each animal. In addition, intensifying competition in livestock product prices with overseas is expected in the future, and labor saving and cost reduction of livestock production are urgent issues. For this reason, there is an increasing need for technology that can constantly monitor the health status of many livestock.

On the other hand, for example, a wearable device worn on a cow and information acquired from a position information sensor provided in a controlled area allow the cow to eat, drink, walk, lie down, run, stand still, and estrus. A technique for managing which activity of rumination is being carried out has been proposed (Patent Document 1 and the like).

Japanese Unexamined Patent Publication No. 2018-7613

According to the method as described in the patent document, it is possible to manage the activity state of the livestock to be managed without relying on human hands. However, although it is possible to determine that a cow is in estrus by the technique described in the above document, it is not possible to accurately determine the state of health. In addition, a means for detecting the position of the cow to be managed is required, resulting in a large-scale system. Furthermore, there is a lot of information to be processed in order to determine which active state the cow is in, and the algorithm becomes complicated accordingly, which may result in a high-cost system.

In view of the above problems, it is an object of the present invention to provide a technique capable of accurately knowing the health condition of livestock to be managed by configuring a simple system.

In order to solve the above problems, the livestock health condition management system according to the present invention includes a temperature sensor and an acceleration sensor, and is attached to the livestock to be managed, and has body surface temperature information of the livestock and attitude information of the livestock. , And the detection means for acquiring the activity amount information of the livestock, and
A diagnostic means for diagnosing the health condition of the livestock based on the body surface temperature information of the livestock, the posture information of the livestock, and the activity amount information of the livestock.
Have.

In addition, "livestock body surface temperature information", "livestock posture information", and "livestock activity information" are general information on livestock body surface temperature, general information on livestock posture, and livestock activity amount, respectively. It means all the information about.

Further, as the detection means referred to here, for example, a wearable device worn on livestock can be used. According to the system having such a configuration, a wearable device equipped with a temperature sensor and an acceleration sensor, and a calculation attached to each livestock to calculate the body surface temperature, posture, and activity amount of the livestock from the output information of the sensor. A simple system configuration with a device makes it possible to accurately grasp the health condition of livestock.

Further, the posture information of the livestock includes information on whether the livestock is in a standing position or a lying position, the acceleration sensor is a three-dimensional acceleration sensor, and the detection means is the livestock. It is equipped with a posture index calculating means that is attached to the tail and calculates the roll angle information of the tail of the livestock from the output of the three-dimensional acceleration sensor, and the diagnostic means is such that the livestock is standing and lying down based on the roll angle information. It may be provided with a posture state determining means for determining which state of the position. The "standing position" indicates that the patient is in a standing position, and the "lying position" indicates that the patient is in a lying position.

Such a configuration of the detection means makes it possible to equip livestock with a wearable device in a non-invasive manner and with a low burden. Further, since the posture of the livestock can be determined to be either the standing position or the lying position based on the roll angle of the tail obtained from the three-dimensional acceleration sensor, a separate configuration for detecting the posture is omitted. Can be done.

In addition, the diagnostic means extracts the highest value within the latest first predetermined time of the body surface temperature of the livestock, and the highest value and the latest first predetermined time from which the highest value is extracted belong to. Calculation of body temperature index to calculate the difference value from the average value of the maximum value of the body surface temperature of the livestock in the same time zone as the time zone to which the most recent first predetermined time belongs on each day in the predetermined several days before the day. A third predetermined time of the posture index calculation unit and the activity amount of the livestock, which calculates the total value of the unit and the time when the livestock is standing or lying down within the second predetermined time. Based on the activity index calculation unit that calculates the total value of the above and the way of fluctuation of each calculated value, the health condition of the livestock corresponds to one of a plurality of preset statuses. It may be provided with a health condition determination unit for determining whether or not.

Alternatively, the diagnostic means extracts the highest value within the most recent first predetermined time of the body surface temperature of the livestock, and the highest value and the most recent first predetermined time from which the highest value is extracted belong to. Calculation of body temperature index to calculate the difference value from the average value of the maximum body surface temperature of the livestock in the same time zone as the time zone to which the most recent first predetermined time belongs on each day in the predetermined several days before the day. The total value of the part and the time during which the livestock is standing or lying down within the latest fourth predetermined time, and within the fourth predetermined time in the past from before the fourth predetermined time. The posture index calculation unit that calculates the ratio of the total value of the time that is the same posture as the posture selectively selected from the standing position or the lying position, and the activity of the livestock within the latest fifth predetermined time. The activity index calculation unit that calculates the ratio of the total value of the amount and the total value of the activity amount within the fifth predetermined time from before the fifth predetermined time to the past, and the variation of each calculated value. A health condition determination unit for determining whether the health condition of the livestock corresponds to any of a plurality of preset types of status may be provided based on the above method.

The first predetermined time, the second predetermined time, the third predetermined time, the fourth predetermined time, and the fifth predetermined time may be all or part of the same time. In addition, the "plurality of statuses" referred to here can include classifications of normal, estrus, parturition, fever, milk fever, and hoof fever. By determining the health condition of livestock by such processing, anyone can easily grasp the health condition of livestock.

Further, the activity amount information includes the number of posture changes of the livestock, and the diagnostic means uses a posture change number calculation unit that calculates the number of posture changes of the livestock based on the roll angle information. Further, the health condition determination unit uses information on how the number of times the posture of the livestock changes, and whether the health condition of the livestock corresponds to any of a plurality of preset statuses. May be used to determine.

Since it is possible to calculate the number of changes in the posture of the livestock (for example, changes in the standing position and the lying position) from the information on the roll angle of the tail of the livestock, it is more accurate to use such information together. It is possible to judge the health condition of livestock well.

Further, the health state determination unit may be a learned model learned by a machine learning method.

Further, the detecting means and the diagnostic means are separate devices each including a wireless communication means, and the detecting means transmits information including at least the output values of the temperature sensor and the acceleration sensor via the wireless communication means. It may be transmitted to the diagnostic means.

According to such a configuration, the detection means can be simplified, the burden on livestock can be further reduced, and the cost of individual wearable devices can be suppressed.

Further, the diagnostic means may further include an output means for outputting the diagnosis result of the health condition of the livestock. Further, the output means may include a wireless communication means for transmitting the diagnosis result of the health condition of the livestock to another device by wireless communication.

With such a configuration, for example, it is possible to have a mobile information terminal transmit a health condition diagnosis result, and it is possible to grasp the health condition of livestock in real time without being restricted by a place.

In addition, the livestock health information management method according to the present invention is
A detection step for acquiring body surface temperature information of a livestock to be managed, posture information of the livestock, and activity amount information of the livestock, and
It has a diagnostic step of diagnosing the health condition of the livestock based on the body surface temperature information of the livestock, the posture information of the livestock, and the activity amount information of the livestock.

Further, the posture information of the livestock includes the roll angle information of the tail of the livestock, and in the diagnosis step, the livestock is in either the standing position or the lying position based on the roll angle information of the tail of the livestock. It may further have a step of determining whether it is in a state.

The diagnostic step includes a body temperature index calculation step, a posture index calculation step, an activity amount index calculation step, and a health condition determination step. In the body temperature index calculation step, the latest body surface temperature of the livestock is obtained. The most recent first predetermined value of each day in the predetermined number of days before the day to which the highest value and the most recent first predetermined time to which the highest value was extracted are extracted. Difference value from the average value of the maximum body temperature of the livestock in the same time zone as the time zone to which the predetermined time belongs and the average value of the maximum body temperature of the same time zone in each day of the previous predetermined several days. In the posture index calculation step, the total value within the second predetermined time of the time when the livestock is standing or lying down is calculated, and in the activity amount index calculation step, the livestock The total value of the activity amount within the third predetermined time is calculated, and in the health condition determination step, the fluctuation of the value calculated in each step of the body temperature index calculation step, the posture index calculation step, and the activity index calculation step. Based on the above method, it may be determined whether the health condition of the livestock corresponds to any of a plurality of preset types of statuses.

Alternatively, the diagnostic step includes a body temperature index calculation step, a posture index calculation step, an activity amount index calculation step, and a health condition determination step, and in the body temperature index calculation step, the body surface temperature of the livestock. The highest value within the most recent first predetermined time is extracted, and the highest value and the most recent first of each day in a predetermined number of days before the day to which the latest first predetermined time to which the highest value is extracted belong. The average value of the maximum body surface temperature of the livestock in the same time zone as the time zone to which the predetermined time of 1 belongs, and the average value of the maximum body surface temperature of the same time zone of each day in the previous predetermined several days. The difference value is calculated, and in the posture index calculation step, the total value of the standing time or the lying time of the livestock within the latest fourth predetermined time and the time before the fourth predetermined time Further, the ratio of the total value of the time in which the posture is the same as the posture selectively selected from the standing or lying position within the fourth predetermined time in the past is calculated, and in the activity amount index calculation step, the above-mentioned The ratio of the total value of the activity amount of the livestock within the latest fifth predetermined time and the total value of the activity amount within the fifth predetermined time from before the fifth predetermined time to the past is calculated. In the health condition determination step, the health condition of the livestock is preset based on how the values calculated in each step of the body temperature index calculation step, the posture index calculation step, and the activity amount index calculation step fluctuate. It may be one that determines whether one of a plurality of types of statuses is applicable.

Further, the activity amount information includes the number of posture changes of the livestock, and the diagnosis step is a posture change number calculation step of calculating the number of posture changes of the livestock based on the roll angle information. Further, in the health state determination step, whether the health state of the livestock corresponds to any of a plurality of preset types of status by using the information on how the number of times the posture change of the livestock changes. May be used to determine.

Further, the present invention can be regarded as a program for causing the information processing apparatus to execute each of the above methods, and as a computer-readable recording medium in which such a program is recorded non-temporarily.

Moreover, the method of creating the trained model according to the present invention is
A method of creating a trained model for determining the health status of managed livestock.
A step of inputting the body surface temperature information of the livestock, the posture information of the livestock, and the activity amount information of the livestock over a predetermined period.
A step of outputting a status indicating the health condition of the livestock based on the combination of the input information is included.

In addition, the method of creating the trained model is as follows.
The body surface temperature information of the livestock, the posture information of the livestock, and the activity amount information of the livestock over the predetermined period are used as training data.
A status indicating the health condition of the livestock, which is preset according to the combination of the body surface temperature information of the livestock, the attitude information of the livestock, and the activity amount information of the livestock within the predetermined period. Using a training data set that uses information as correct data,
It may be one that trains the model to be trained.

Further, the trained model according to the present invention is
A trained model for determining the health status of managed livestock.
For input of the body surface temperature information of the livestock, the posture information of the livestock, and the activity amount information of the livestock over a predetermined period.
A status indicating the health condition of the livestock, which is preset according to the combination of the body surface temperature information of the livestock, the posture information of the livestock, and the activity amount information of the livestock during the predetermined period. It is characterized by outputting.

It should be noted that each of the above configurations and processes can be combined with each other to construct the present invention as long as no technical contradiction occurs.

According to the present invention, it is possible to accurately know the health condition of livestock to be managed by configuring a simple system.

FIG. 1 is a diagram showing an outline of a configuration example of a livestock health management system according to the first embodiment. FIG. 2 is a functional block diagram showing an outline of the functional configuration of the livestock health management system according to the first embodiment. FIG. 3 is a diagram showing the arrangement relationship between the acceleration detection axis of the acceleration sensor according to the first embodiment and the tail of a cow. FIG. 4 shows the relationship between the type of cattle health condition determined by the health condition determination unit according to the first embodiment, the change in the body surface temperature of the cow, the change in the lying (lying position) time, and the change in the amount of activity. It is a list showing an example. FIG. 5 is an explanatory diagram in which the experimental data of Experimental Example 1 is graphed. FIG. 6 is an explanatory diagram in which the experimental data of Experimental Example 2 is graphed. FIG. 7 is a flowchart showing the overall flow of the health state management process of the livestock health management system according to the first embodiment. FIG. 8 is a flowchart showing a routine of the health condition diagnosis process according to the first embodiment. FIG. 9 is a block diagram showing a functional configuration of a livestock health management system according to a modified example of the first embodiment. FIG. 10 is a flowchart showing a subsystem of the health condition diagnosis process according to the modified example of the first embodiment. FIG. 11 is an explanatory diagram in which the experimental data of Experimental Example 3 is graphed. FIG. 12 is an explanatory diagram in which the experimental data of Experimental Example 4 is graphed. FIG. 13 is an explanatory diagram in which the experimental data of Experimental Example 5 is graphed. FIG. 14 is a diagram showing an outline of a configuration example of a livestock health management system according to the second embodiment.

Hereinafter, specific embodiments of the present invention will be described with reference to the drawings.

<Embodiment 1>
First, a first example of the embodiment of the present invention will be described with reference to FIGS. 1 to 8. However, unless otherwise specified, the dimensions, materials, shapes, relative arrangements, and the like of the components described in this embodiment are not intended to limit the scope of the present invention to those.

(System configuration)
FIG. 1 is a diagram showing an outline of a livestock health management system 1 according to the present embodiment, and FIG. 2 is a block diagram showing a functional configuration of a livestock health management system 1. As shown in FIG. 1, the livestock health management system 1 has a wearable device 110 attached to the livestock, and an information processing terminal 220 that performs information communication by wireless communication with the wearable device 110. A desired known technique can be used for wireless communication, and for example, Wi-Fi (registered trademark), Bluetooth (registered trademark), ZigBee (registered trademark) and the like can be adopted. Further, communication may be performed by a public wireless communication network such as LTE (Long Term Evolution).

(Wearable device)
In this embodiment, the wearable device 110 is worn near the base of the cow's tail. Therefore, although not shown, the wearable device 110 has a configuration including a sensor module and a mounting belt portion. The sensor module is integrally or detachably arranged on the mounting belt portion, and the wearable device 110 is mounted by winding and fixing the mounting belt portion near the base of the cow's tail. Further, the wearable device 110 includes a battery and supplies electric power to each sensor, control unit, and the like. The battery can also be a rechargeable secondary battery.

Further, as shown in FIG. 2, the wearable device 110 includes functional units of a temperature sensor 111, an acceleration sensor 112, a storage unit 113, a communication unit 114, and a control unit 115. In this embodiment, the wearable device 110 corresponds to the detection means.

The temperature sensor 111 measures the body surface temperature of the cow to be controlled every predetermined time (for example, 2 minutes) and outputs the temperature. The acceleration sensor 112 is a so-called three-dimensional acceleration sensor capable of measuring acceleration in the three-axis directions of the X-axis, Y-axis, and Z-axis, and outputs acceleration in the three-axis direction of the cow's tail. FIG. 3 shows the relationship between the directions of the three axes and the tail of the cow in this embodiment. As shown in FIG. 3, in the present embodiment, the Y-axis will be described as being on the base of the tail of the cow and the extension line of the tail.

The storage unit 113 is a memory device such as a RAM (Random Access Memory) and a ROM (Read only memory), and is a program executed by the control unit 115 described later, timing information for measurement by each sensor, and a measured value by each sensor. , Etc. are stored.

The communication unit 114 includes a communication antenna and wireless communication processing means (neither shown) for performing information communication with the information processing terminal 120 by a wireless communication method. The wireless communication processing unit of the communication unit 114 converts the data transmitted from the control unit 115, which will be described later, into a wireless communication signal, and transmits the signal from the antenna. As a result, the wearable device 110 uses the communication unit 114 to acquire the body surface temperature information of the cow (for example, the body surface temperature every 2 minutes), the posture information (for example, the posture index described later), and the activity. The amount information (for example, the activity amount index described later) is transmitted to the information processing terminal 120.

The control unit 115 includes a CPU (Central Processing Unit) and controls each functional unit of the wearable device 110. In addition, each function module of the posture index calculation unit 1151 and the activity amount index calculation unit 1152 is provided.

The posture index calculation unit 1151 calculates the average value of the roll angles of the cow's tail for each predetermined time (for example, 2 minutes) as the posture index based on the output value obtained from the acceleration sensor 112. For example, this may be obtained from the average value of the X-axis / Z-axis acceleration of the cow's tail measured by the acceleration sensor 112 at predetermined time intervals. The posture index is included in the posture information transmitted to the information processing terminal 120.

The activity index calculation unit 1152 calculates the activity index of cattle based on the output value obtained from the acceleration sensor 112. Specifically, the standard deviation of the minimum value of the activity amount of cattle for a predetermined time (for example, 5 seconds) for a predetermined time (for example, 2 minutes) is calculated, and this is used as the activity amount index. The activity amount index is included in the activity amount information transmitted to the information processing terminal 120.

(Information processing terminal)
The information processing terminal 120 includes a control unit 121, a communication unit 122, a storage unit 123, an input unit 124, and a display unit 125. The information processing terminal 120 can be configured as a computer including a CPU, RAM, ROM, and the like, and a general-purpose information processing device such as a personal computer can be adopted. In this embodiment, the information processing terminal 120 corresponds to the diagnostic means.

The control unit 121 controls each part of the information processing terminal 120, and further includes each function module of a body temperature index calculation unit 1211, a posture index calculation unit 1212, an activity amount index calculation unit 1213, and a health condition determination unit 1214. .. Each function module will be described later.

The communication unit 122 includes a communication antenna and a wireless communication processing unit (neither shown) for performing information communication with the communication unit 114 of the wearable device 110 by a wireless communication method. As a result, the information processing terminal 120 acquires information from the wearable device 110 via the communication unit 122.

The storage unit 123 is composed of a large-capacity storage device such as a RAM, ROM, HDD (Hard Disk Drive) or SSD (Solid State Drive), and is a program executed by the control unit 121, information acquired from the wearable device 110, and a cow. Various information such as history information of the health condition of is stored.

The input unit 124 is various input means for receiving the user's operation on the information processing terminal 120, and a desired known input means such as a keyboard, a mouse, and a touch panel display (none of which are shown) can be adopted. ..

The display unit 125 is an output device that displays various information including the health condition of the cow to be managed diagnosed by the control unit 121, and for example, a liquid crystal display monitor or the like can be adopted.

(Function module for determining health condition)
Subsequently, the functional module included in the control unit 121 will be described. The body temperature index calculation unit 1211 calculates the body temperature index, and calculates the body temperature index based on the body temperature index. More specifically, from the body surface temperature information stored in the storage unit 123, the highest value of the cow's body surface temperature within the last hour is calculated (extracted) as a body temperature index, and the extracted body temperature index and the body temperature are obtained. The difference value between the average value of the maximum body surface temperature of the cow in the same time zone as the time zone of each day in the three days before the day to which the index is extracted belongs is calculated as the body temperature index. Specifically, for example, assuming that the current time is 14:00, the maximum value of the body surface temperature within 1 hour from 13:00 to 14:00 is extracted, and the maximum value is stored in the storage unit 123. Then, the average value of the maximum value of the body surface temperature between 13:00 and 14:00 in the past 3 days is obtained from the storage unit 123, and the average value and the body surface temperature between 13:00 and 14:00 on the day extracted earlier are obtained. Calculate the difference from the highest value.

The posture index calculation unit 1212 calculates the total value of the cow's lying down time within a predetermined time (for example, the past 24 hours) as a posture index based on the posture index of the cow stored in the storage unit 123. Specifically, for example, when the posture index of the tail of the cow (for example, the average value of the roll angle for 2 minutes) is less than -0.4 or more than +0.4, the posture of the cow is in the recumbent position, -0. If it is 4 or more and +0.4 or less, the posture of the cow is judged to be standing. Then, the total value of the lying down time of the cow within the predetermined time (for example, the past 24 hours) is calculated, and this is used as the posture index. That is, in the present embodiment, the posture index calculation unit also serves as the posture state determination means. The posture index may be the total value of the standing time within a predetermined time instead of the lying time.

Further, the activity amount index calculation unit 1213 uses the total value of the cattle activity amount index within a predetermined time (for example, the past 24 hours) as the activity amount index based on the cattle activity amount information stored in the storage unit 123. calculate.

The health condition determination unit 1214 determines whether the health condition of the cow corresponds to one of a plurality of preset statuses based on the calculated method of fluctuation of the body temperature index, the posture index, and the activity amount index. judge. Specifically, for example, when the body temperature index decreases, the posture index increases, and the activity amount index decreases, it is determined that delivery is approaching. Then, the result of the determination is output as a diagnosis result of the health condition of the cow. The health condition determination unit 1214 can employ artificial intelligence (learning model) that has been learned by machine learning such as deep learning.

FIG. 4 shows the type of cattle health condition determined by the health condition determination unit 1214 of the present embodiment, the change in the body surface temperature (body temperature index) of the cow, the change in the lying time (posture index), and the activity amount index. It is a list showing an example of the relationship with change. As shown in FIG. 4, in the present embodiment, the cow has "normal", "estrus", "parturition", "fever disease", "milk fever", and "hoof fever" depending on how each index is increased or decreased. Determine which is the case.

Here, when the health condition determination unit 1214 is used as artificial intelligence, each of the above indicators is used as training data over a predetermined period (for example, the past 3 days), and a combination of how each index fluctuates during the predetermined period is used. It is possible to use a trained model that has been trained using a training data set in which each of the above-mentioned health states that are set in advance is used as correct answer data.

In addition, when livestock are bred under conditions where behavior such as tethering is restricted, the amount of activity does not increase so much even if they are in estrus, but in such a case. However, it is considered that the time when the posture is standing is extended. For this reason, changes in health status that may be overlooked only by changes in the amount of activity can be detected by combining with posture information. In addition, by making a judgment based on such a combination, it is possible to strictly set a threshold value for determining the health condition for each of the posture information and the activity amount information, so that the fluctuation of the health condition can be detected accurately. be able to. In addition, even at the time of calving, the body temperature index is based on the breeding mode of the livestock at the time of calving (whether the calving is completed or moved to the calving chamber) and the size of the livestock when it is transferred to the calving chamber. By using the three indicators of, posture index, and activity amount index, it is possible to flexibly respond to various breeding forms.

The following shows the experimental data when the wearable device 110 was attached to the tail of a cow and the experiment was conducted.

(Experimental Example 1)
FIG. 5 is a graph showing the body temperature index, the posture index, and the activity amount index before and after the estrus of the cow in chronological order. From FIG. 5, it can be seen that the cow is in estrus at the timing when the conditions that the body temperature index suddenly increases, the posture index decreases, and the activity amount index increases are met.

(Experimental Example 2)
FIG. 6 is a graph showing the body temperature index, the posture index, and the activity amount index before and after calving of the cow in chronological order. From FIG. 6, it can be seen that the cow delivers at the timing when the conditions that the body temperature index rises, then drops sharply, then rises slightly, the posture index sharply decreases, and the activity index rises are met.

From each of the above experimental examples, it can be seen that there is a clear correlation between the amount of fluctuation (how it fluctuates) between each index in important events for health management such as estrus and delivery.

(Processing flow)
Next, based on FIGS. 7 and 8, the flow of processing for managing the health condition of cattle by the livestock health management system 1 in the present embodiment will be described. FIG. 7 is a flowchart showing the overall flow of the health management process by the livestock health management system 1. As shown in FIG. 7, first, the temperature sensor 111 of the wearable device 110 detects the body surface temperature of the cow to be managed at predetermined time intervals (step S101). Further, the acceleration sensor 112 detects the acceleration of the X-axis, Y-axis, and Z-axis of the cow's tail (step S102). Further, based on the value detected in step S102, the posture index calculation unit 1151 calculates the average value of the roll angle of the tail of the cow at predetermined time intervals (step S103), and the activity index calculation unit 1152 calculates the activity of the cow. The quantity index is calculated (step S104). Then, the body surface temperature information, the attitude information, and the activity amount information thus obtained are transmitted to the information processing terminal 120 via the communication unit 130 (step S105), and are transmitted by the control unit 121 of the information processing terminal 120. The cow health condition diagnosis process is carried out (step S106). This completes the routine once, but the routine is repeatedly executed at any time. This makes it possible to grasp and manage the health condition of cattle in real time.

In the present embodiment, the processes from step S101 to step S104 correspond to the detection step. The processes from step S101 to step S104 do not necessarily have to be executed in such an order, and the order of processes from step S101 and steps S102 to S104 may be interchanged. Further, the processing order of step S103 and step S104 may be reversed.

Subsequently, the subsystem of the health condition diagnosis process in step S106 will be described. FIG. 8 is a flowchart showing the flow of the cattle health condition diagnosis process executed by the information processing terminal 120. As shown in FIG. 8, first, the body temperature index calculation unit 1211 extracts the maximum value of the body surface temperature within a predetermined time based on the body surface temperature information stored in the storage unit 123 (step S111). The body temperature index is calculated based on this (step S112). Next, the posture index calculation unit 1212 determines the standing / lying position of the cow at predetermined time intervals (step S113), and the posture index is calculated based on this (step S114). Further, the activity amount index calculation unit 1213 calculates the activity amount index (step S115). Subsequently, the health condition determination unit 1214 executes the health condition determination process (step S116), outputs the determination result to the display unit 125 (step S117), and the routine ends once. Since the body temperature index, the posture index, the activity amount index, and the health condition determination process have already been described, detailed description thereof will be omitted here.

In the present embodiment, the processes from step S111 to step S116 correspond to the diagnostic step. Further, steps S111 and S112 correspond to the body temperature index calculation step, steps S113 and S114 correspond to the posture index calculation step, step S115 corresponds to the activity amount index calculation step, and step S116 corresponds to the health condition determination step. The order of processing of the body temperature index calculation step, the posture index calculation step, and the activity amount index calculation step may be interchanged.

According to the livestock health management system 1 having the above configuration, if the wearable device 110 having a simple configuration is attached to the livestock, the health of the livestock to which the module is attached can be obtained even if the livestock is away from the livestock. You can know the state. In addition, since the health condition of livestock is automatically diagnosed, even a person with poor biological and medical knowledge about livestock can manage the health of livestock.

<Modification example 1>
In the above embodiment, the activity index is the total value of the activity index of the cattle within the predetermined time, and the posture index is the total value of the lying time of the livestock within the predetermined time. Not limited.

For example, the activity amount index is the total value of the activity amount index of livestock within the past predetermined time (for example, 24 hours) and the same predetermined time in the past from the predetermined time before (for example, 24 hours before 4).
It may be the ratio (activity amount ratio) with the total activity amount index value (between 8 hours ago).

Further, the activity amount may be used as the number of tail behaviors, and the activity amount index may be obtained according to the degree of tail raising. Specifically, the average acceleration of the Y-axis within the past predetermined time (for example, 6 hours) and the Y-axis within the same predetermined time in the past (for example, between 6 hours and 12 hours) from before the predetermined time. It may be the ratio (Y-axis ratio) with the average acceleration of. The activity amount ratio and the Y-axis ratio may not be used alternately as the activity amount index, but may be used in combination.

In addition, the posture index is the total standing time of the livestock within the past predetermined time (for example, 24 hours) and the same predetermined time in the past from the predetermined time (for example, between 24 hours and 48 hours). It may be the ratio (standing time ratio) to the total standing time value of. It is natural that the lying time can be used instead of the standing time.

<Modification 2>
In addition, the information used by the health condition determination unit 1214 to determine the health condition of livestock is not limited to the body temperature index, the posture index, and the activity amount index, and the health condition is determined using other information. Is also possible. In the following, the livestock health management system 11 which is such a modification will be described with reference to FIGS. 9 and 10. FIG. 9 is a block diagram showing a functional configuration of the livestock health management system 11 according to the present modification, and FIG. 10 is a flowchart showing a subsystem of the health condition diagnosis process according to the present modification.

As shown in FIG. 9, the livestock health management system 11 according to the present modification has a configuration in which the information processing terminal 1200 includes a posture change frequency calculation unit 1215. Since the other points are the same as those of the health management system 1 of the first embodiment, the same configurations and processes as those of the health management system 1 of the first embodiment are designated by the same reference numerals, and the description thereof will be omitted.

The posture change number calculation unit 1215 counts the number of posture changes (for example, changes in standing and lying positions) of the livestock based on the roll angle information of the tail of the livestock obtained from the acceleration sensor 112, and counts the number of times the posture changes (for example, changes in standing and lying positions) in the past predetermined time ( The ratio (number of posture changes) between the number of posture changes of a livestock within (for example, 24 hours) and the number of posture changes within the same predetermined time (for example, between 24 hours and 48 hours) before the predetermined time. It is a functional part that calculates the ratio).

As shown in FIG. 10, in the subroutine of the health condition determination process according to the present modification, the posture change frequency ratio is calculated after the body temperature index, the posture index, and the activity amount index are calculated in the processes of steps S111 to S115. It is calculated (S121). Then, the health condition determination unit 1214 performs a health condition determination process (S116) using the posture change frequency ratio in addition to the body temperature index, the posture index, and the activity amount index, and outputs the diagnosis result (S117).

The following shows the experimental data when the wearable device 110 was attached to the tails of a plurality of livestock and the experiment was conducted.

(Experimental Example 3)
In FIG. 11, the number of samples is 25, and the body temperature index, the activity amount ratio (activity amount index), the standing time ratio (posture index), and the recumbent frequency ratio (that is, the posture change frequency ratio) before and after estrus of the cow are timed. It is a graph displayed in series. From FIG. 11, it can be seen that the cow is in estrus at the timing when the conditions that the body temperature index, the activity amount ratio, and the standing time ratio suddenly increase, and the standing frequency ratio decreases and then suddenly increase.

(Experimental Example 4)
FIG. 12 is a graph showing the body temperature index, the activity amount ratio, the standing time ratio, the lying frequency ratio, and the Y-axis ratio (activity amount index) before and after delivery of the cow as the number of samples 30 in chronological order. From FIG. 12, it can be seen that the cows deliver at the timing when the conditions that the body temperature index decreases, the activity ratio and the standing time ratio rapidly increase, and the Y-axis ratio suddenly decreases are met.

(Experimental Example 5)
FIG. 13 is a graph showing the body temperature index, the activity amount ratio, the lying time ratio (posture index), and the posture change frequency ratio before and after delivery of the pig as the number of samples 9 in chronological order. From FIG. 13, the pigs can deliver at the timing when the conditions are met that the body temperature index rises after a gradual downward trend, the activity ratio and the posture change frequency ratio rise sharply, and the lying time ratio falls. Understand.

Even when the activity ratio, standing (or lying down) time ratio, posture change frequency ratio, and Y-axis ratio are used as in each of the above experimental examples, these fluctuations include estrus and delivery. It can be seen that there is a clear correlation depending on the events that are important for health management.

<Embodiment 2>
Next, another embodiment of the present invention will be described with reference to FIG. FIG. 12 is a schematic view showing a configuration example of the livestock health condition management system 2 according to the present embodiment. As shown in FIG. 12, the livestock health condition management system 2 includes wearable devices 210a, 210b, 210c, 210d ... Attached to a plurality of cows, an information processing terminal 220, and a mobile information terminal 230. , These are interconnected by network N. For the network N, for example, WAN (Wide Area Network), which is a world-wide public communication network such as the Internet, and other public communication networks such as LTE may be adopted. Also, network N
May include a wireless communication network such as Wi-Fi, and may include a signal relay device if necessary.

The mobile information terminal 230 may be made of a desired known technology, and for example, a notebook PC, a tablet terminal, a smartphone, or the like can be adopted. Since the configurations of the wearable devices 210a, 210b, 210c, 210d ..., And the information processing terminal 220 are the same as those of the first embodiment, detailed description thereof will be omitted. The communication unit of the wearable device 210 and the information processing terminal 220 connects to the network N to perform information communication.

In the livestock health condition management system 2 of the present embodiment, each of the plurality of wearable devices has body surface temperature information and posture information together with an identification signal capable of identifying each wearable device on the information processing terminal 220 via the network N. , Send activity information. Then, the information processing terminal 220 executes a health condition diagnosis process for each wearable device 210 identified by the identification information.

The information processing terminal 220 transmits the diagnosis result to the mobile information terminal 230 based on a predetermined rule. For example, it may be set to transmit the diagnosis result together with the identification information (that is, in a manner that can identify the cow) only when there is a cow whose health condition is diagnosed as other than "normal". .. Alternatively, the desired cow health information can be sent periodically, or arbitrary information (for example, a list of managed cow health status) can be sent according to the user's operation.

According to the livestock health condition management system 2 having the above configuration, the user can acquire information on the health condition of cattle in real time on the mobile information terminal 230 wherever he / she is. In addition, since a large number of cows can be automatically managed together, human costs can be significantly reduced. Further, since the configuration of each wearable device 210 is simple, the cost of the device can be suppressed.

<Others>
The description of each of the above examples merely illustrates the present invention, and the present invention is not limited to the above specific form. The present invention can be variously modified and combined within the scope of its technical idea.

For example, in the second embodiment, a configuration may include a plurality of mobile information terminals, or conversely, a plurality of wearable devices directly communicate information with an information processing terminal without including a network and a mobile information terminal in the configuration. It may be configured. Further, the information processing terminal may be a cloud server or the like, and the business operator may centrally manage the information processing terminal.

Further, the calculation of each index / index and the determination processing described above may be executed by either the wearable device or the information processing terminal. For example, in each of the above embodiments, the body temperature index is calculated by the information processing terminal, but this may be calculated by the wearable device. Further, regarding the information on the posture, the wearable device may perform the standing / lying position determination process of the cow based on the posture index, and the determination result may be sent to the information processing terminal. Alternatively, the posture index and the activity index may also be calculated by the information processing terminal. Further, it is also possible to complete all the health condition diagnosis processing performed by the information processing terminal with a wearable device and transmit only the diagnosis result to an external device.

Further, in the above embodiment, the posture index is calculated by obtaining the roll angle of the tail of the cow from the X-axis / Z-axis acceleration of the three-dimensional acceleration sensor, but it may be obtained by another method. For example, a uniaxial or biaxial accelerometer may be used to determine the posture state of the cow from the frequency of tail movement.

In the above embodiment, cattle are used as livestock to be managed, but the livestock to be managed is not limited to cattle, and can be applied to other animals such as horses, sheep, goats, and pigs. Also, the classification of health status is not limited to the ones given in the above example.

Further, although the artificial intelligence in which machine learning is performed as the health condition determination unit is illustrated in the above embodiment, the present invention is not limited to this, and a so-called expert system may be used. Further, the determination process is not limited to artificial intelligence, and the determination process may be performed according to a threshold value according to a predetermined program.

1, 2 ... Livestock health management system 110, 210 ... Wearable device 120, 220, 1200 ... Information processing terminal 115, 121 ... Control unit 230 ... Mobile information terminal

Claims (19)

A detection means provided with a temperature sensor and an acceleration sensor, which is attached to a livestock to be managed and acquires body surface temperature information of the livestock, posture information of the livestock, and activity amount information of the livestock.
A diagnostic means for diagnosing the health condition of the livestock based on the body surface temperature information of the livestock, the posture information of the livestock, and the activity amount information of the livestock.
Livestock health management system. The posture information of the livestock includes information on whether the livestock is in a standing position or a lying position.
The acceleration sensor is a three-dimensional acceleration sensor.
The detection means is attached to the tail of the livestock and includes a posture index calculating means for calculating roll angle information of the tail of the livestock from the output of the three-dimensional acceleration sensor.
The diagnostic means includes a posture state determining means for determining at least whether the livestock is in a standing position or a lying position based on the roll angle information.
The livestock health condition management system according to claim 1, wherein the livestock health condition management system is characterized in that. The diagnostic means
The highest value within the most recent first predetermined time of the body surface temperature of the livestock is extracted, and each day in the predetermined several days before the day to which the highest value and the most recent first predetermined time from which the highest value is extracted belong. The body temperature index calculation unit that calculates the difference between the average value of the maximum body surface temperature of the livestock in the same time zone as the time zone to which the most recent first predetermined time belongs, and
A posture index calculation unit that calculates the total value within a second predetermined time of the time when the livestock is standing or lying down.
An activity amount index calculation unit that calculates the total value of the activity amount of the livestock within a third predetermined time, and
A health condition determination unit for determining whether the health condition of the livestock corresponds to any of a plurality of preset types of status based on the calculated fluctuation of each value is provided.
2. The livestock health condition management system according to claim 2. The diagnostic means
The highest value within the most recent first predetermined time of the body surface temperature of the livestock is extracted, and each day in the predetermined several days before the day to which the highest value and the most recent first predetermined time from which the highest value is extracted belong. The body temperature index calculation unit that calculates the difference between the average value of the maximum body surface temperature of the livestock in the same time zone as the time zone to which the most recent first predetermined time belongs, and
The total value of the time during which the livestock is standing or lying down within the most recent fourth predetermined time, and the standing within the fourth predetermined time from before the fourth predetermined time or further in the past. A posture index calculation unit that calculates the ratio of the total value of time that is the same posture as the posture selectively selected from the lying position, and the posture index calculation unit.
The ratio of the total value of the activity amount of the livestock within the latest fifth predetermined time and the total value of the activity amount within the fifth predetermined time from before the fifth predetermined time to the past is calculated. Activity index calculation department and
A health condition determination unit for determining whether the health condition of the livestock corresponds to any of a plurality of preset types of status based on the calculated fluctuation of each value is provided.
2. The livestock health condition management system according to claim 2. The activity amount information includes the number of times the posture of the livestock changes.
The diagnostic means further includes a posture change number calculation unit that calculates the number of posture changes of the livestock based on the roll angle information.
The health condition determination unit also uses information on how the number of changes in the posture of the livestock changes, and determines whether the health condition of the livestock corresponds to any of a plurality of preset statuses. ,
The livestock health condition management system according to claim 3 or 4, characterized in that. The health condition determination unit is a trained model learned by a machine learning method.
The livestock health condition management system according to any one of claims 3 to 5, wherein the livestock health condition management system is characterized in that. The detection means and the diagnostic means are separate devices each having a wireless communication means.
The detection means transmits information including at least the output values of the temperature sensor and the acceleration sensor to the diagnostic means via the wireless communication means.
The livestock health condition management system according to any one of claims 1 to 5, wherein the livestock health condition management system is characterized in that. The diagnostic means further includes an output unit that outputs a diagnosis result of the health condition of the livestock.
The livestock health condition management system according to any one of claims 1 to 7, wherein the livestock health condition management system is characterized in that. The output unit includes a wireless communication means for transmitting the diagnosis result of the health condition of the livestock to another device by wireless communication.
8. The livestock health condition management system according to claim 8. A livestock wearable device mounted on the tail of a domestic animal, comprising a temperature sensor, an acceleration sensor, and a wireless communication means, which constitutes the detection means according to at least one of claims 1 to 9. A detection step for acquiring body surface temperature information of a livestock to be managed, posture information of the livestock, and activity amount information of the livestock, and
A diagnostic step for diagnosing the health condition of the livestock based on the body surface temperature information of the livestock, the posture information of the livestock, and the activity amount information of the livestock.
How to manage the health of livestock with. The attitude information of the livestock includes roll angle information of the tail of the livestock.
The diagnostic step further comprises a step of determining whether the livestock is in at least a standing or lying position based on the roll angle information of the tail of the livestock.
The method for managing the health condition of livestock according to claim 11. The diagnostic step includes a body temperature index calculation step, a posture index calculation step, an activity amount index calculation step, and a health condition determination step.
In the body temperature index calculation step, the highest value within the most recent first predetermined time of the body surface temperature of the livestock is extracted, and the highest value and the most recent first predetermined time from which the highest value is extracted belong to. The average value of the maximum body surface temperature of the livestock in the same time zone as the time zone to which the most recent first predetermined time belongs, the same time of each day in the preceding predetermined several days. Calculate the difference value from the average value of the highest value of the body surface temperature of the band,
In the posture index calculation step, the total value of the time when the livestock is standing or the time when the livestock is lying down within the second predetermined time is calculated.
In the activity amount index calculation step, the total value of the activity amount of the livestock within the third predetermined time is calculated.
In the health state determination step, the health state of the livestock is preset based on how the values calculated in each step of the body temperature index calculation step, the posture index calculation step, and the activity amount index calculation step fluctuate. Determine if any of the multiple statuses apply,
The method for managing the health condition of livestock according to claim 12. The diagnostic step includes a body temperature index calculation step, a posture index calculation step, an activity amount index calculation step, and a health condition determination step.
In the body temperature index calculation step, the highest value within the most recent first predetermined time of the body surface temperature of the livestock is extracted, and the highest value and the most recent first predetermined time from which the highest value is extracted belong to. The average value of the maximum body surface temperature of the livestock in the same time zone as the time zone to which the most recent first predetermined time belongs, the same time of each day in the preceding predetermined several days. Calculate the difference value from the average value of the highest value of the body surface temperature of the band,
In the posture index calculation step, the total value of the standing time or the lying time of the livestock within the most recent fourth predetermined time and the fourth predetermined time from before the fourth predetermined time to the past. Calculate the ratio of the total value of the time that is the same posture as the posture selectively selected from the standing or lying position within the predetermined time.
In the activity amount index calculation step, the total value of the activity amount of the livestock within the latest fifth predetermined time and the total activity amount within the fifth predetermined time from before the fifth predetermined time to the past. Calculate the ratio of the value to
In the health state determination step, the health state of the livestock is preset based on how the values calculated in each step of the body temperature index calculation step, the posture index calculation step, and the activity amount index calculation step fluctuate. Determine if any of the multiple statuses apply,
The method for managing the health condition of livestock according to claim 12. The activity amount information includes the number of times the posture of the livestock changes.
The diagnostic step further includes a posture change number calculation step for calculating the number of posture changes of the livestock based on the roll angle information.
In the health condition determination step, it is determined whether or not the health condition of the livestock corresponds to any of a plurality of preset types of statuses, using information on how the number of times the posture changes of the livestock changes. ,
The method for managing the health condition of livestock according to claim 13 or 14. A program for causing an information processing apparatus to execute each step according to any one of claims 11 to 15. A method of creating a trained model for determining the health status of managed livestock.
A step of inputting the body surface temperature information of the livestock, the posture information of the livestock, and the activity amount information of the livestock over a predetermined period.
A method of creating a trained model including a step of outputting a status indicating the health condition of the livestock based on a combination of the above-mentioned information input. The body surface temperature information of the livestock, the posture information of the livestock, and the activity amount information of the livestock over the predetermined period are used as training data.
A status indicating the health condition of the livestock, which is preset according to the combination of the body surface temperature information of the livestock, the attitude information of the livestock, and the activity amount information of the livestock within the predetermined period. Using a training data set that uses information as correct data,
Train the model to be trained,
The trained model creation method according to claim 17, wherein the trained model is created. A trained model for determining the health status of managed livestock.
For input of the body surface temperature information of the livestock, the posture information of the livestock, and the activity amount information of the livestock over a predetermined period.
A status indicating the health condition of the livestock, which is preset according to the combination of the body surface temperature information of the livestock, the posture information of the livestock, and the activity amount information of the livestock during the predetermined period. Output,
Trained model.

Images