JPH04309823A - Measurement of dynamic weight of four-footed animal and device thereof - Google Patents

Abstract

PURPOSE:To measure a static weight value of a four-footed animal accurately by setting a sampling period more than twice as long as a time interval during which the other feet can land after one feet of two feet landing on the weighing surface closely with each other in terms of time have landed. CONSTITUTION:A four-footed animal is weighed by means of a weighing means S while walking on a weighing machine, and its data signals are sent to a measurement processing device H. A period, during which sets by two feet landing on the weighing surface of the weighing machine closely with each other in terms of time can land respectively, is shown, and smooth data can be obtained by smoothing its value into a condition capable of showing a collected maximal value by means of a lotus filter 5B. Its data is taken in successively as a sampling series, and walking periodicity is calculated according to its maximal value, and an effective data section is set from the sampling series, and stationary weight of the four-footed animal is read. In this case, a sampling period in the case of taking in the smooth data is set more than twice as long as a time interval during which the other feet can land after one feet of landing two feet have landed.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for dynamic weight measurement of quadrupedal animals. Specifically, while a quadruped animal is walking on a scale, weight value data is obtained by a weighing means, and pairs of two legs alternately land on the weighing surface of the scale in time. is expressed, and the weighing values are set so that the weighing values when two legs that land temporally close to the weighing surface of the weighing device land on the weighing surface collectively represent one maximum value. The value data is smoothed to obtain smoothed data, the smoothed data is sequentially taken in as a sampling series, the gait periodicity is calculated from the sampling series based on the maximum value, and the gait periodicity is calculated based on the gait periodicity. A method for dynamically measuring the weight of a quadruped animal that sets a valid data interval from a sampling series and reads a static weight value of the quadruped animal from the sampling series in the valid data interval; A weighing means for weighing weight and weighing value data obtained by the weighing means are expressed in a period in which pairs of two legs alternately land on the weighing surface of the weighing device temporally close to each other, and , a smoothing means for smoothing the weighing values when two legs that land temporally close to the weighing surface of the weighing device land on the weighing surface to collectively represent one maximum value; a sampling means for sequentially taking in the smoothed data smoothed by the smoothing means as a sampling series; a gait periodicity evaluation means for calculating the gait periodicity from the sampling series based on the maximum value; and a gait periodicity evaluation means based on the gait periodicity. effective data interval setting means for setting an effective data interval from the sampling series;
The present invention relates to a dynamic weight measuring device for a four-legged animal, comprising a calculation means for calculating a static weight value of the four-legged animal from a sampling series in the valid data interval, and an output means for outputting the calculated static weight value.

0002

[Prior Art] The above-mentioned dynamic weight measuring method and device for a quadrupedal animal are disclosed in, for example, Japanese Patent Application Laid-Open No. 1-280222, and the weight can be measured efficiently while the animal is walking. Furthermore, weight measurement can be performed without stressing the animal compared to forcing the animal to stop. Moreover, in order to remove high-frequency noise caused by vibrations generated when the quadruped animal walks on the weighing device, by smoothing the weighed value data obtained by the weighing means as described above, This allows for highly accurate measurements. In this smoothing,
Taking advantage of the fact that many quadrupedal animals such as cows have a habit of landing their two legs on the weighing surface close to each other in time, the periodicity in which each pair of four legs alternately lands is irregular. Obtain smoothed data by smoothing the weighing values when two legs that land close to the weighing surface land on the weighing surface to collectively represent a single maximum value, while ensuring that the weighing values do not become clear. . In other words, if it is possible to capture the period in which two of the four legs alternately land on the ground, it is possible to sequentially determine the walking periodicity, the effective data interval, and the static weight value. , the above-mentioned high-frequency noise is removed by smoothing the weighing values when two legs that land close to the weighing surface temporally land on the weighing surface to a state where they collectively represent a single maximum value. It is designed to do so. By the way, to smooth weight data, a low-pass filter is generally used, and in the case of cows, for example,
The period in which each set of two of the four legs alternately lands is 1.
It is around 5Hz, and the time interval or period from when one of the two legs lands temporally close to the weighing surface until the other lands is generally 3 to 10Hz.
For example, the cutoff frequency is set to 2Hz.
It will be set to .

By the way, when the smoothed smoothed data is sequentially taken in as a sampling series, in view of the fact that it is smoothed as described above, the sampling period is set alternately between two sets of four legs. It is conceivable to set the period twice as long as the landing period.

0004

[Problem to be Solved by the Invention] As mentioned above, even if smoothing is performed by using a low-pass filter, etc., the frequency components caused by the two legs landing temporally close to the weighing surface cannot be completely removed by the smoothing. It may not be possible to remove it. Therefore, when sequentially importing smoothed data as a sampling series, if the sampling period is set to twice the period in which two of the four legs alternately land, an error may occur in determining the maximum value. There was a risk that proper measurements could not be taken. In other words, if the sampling period is set to twice the period in which two of the four legs alternately land, the frequency components originally caused by the two legs landing close to the weighing surface in time will be removed. In some cases, it may not be possible to properly determine the local maximum value, which may result in the inability to perform proper measurements. The present invention has been made in view of the above-mentioned circumstances, and its purpose is to solve the problem even when the frequency components caused by two legs landing temporally close to the weighing surface cannot be completely removed by smoothing. , the point is to enable proper measurement.

[0005]

[Means for Solving the Problems] A method for dynamically measuring the weight of a quadrupedal animal according to the present invention involves obtaining weighed value data by a weighing means while the quadrupedal animal is walking on a weighing device, and measuring the weighing surface of the weighing device. When a period in which pairs of two legs that land close to each other in time alternately lands, and two legs that land close to the weighing surface of the weighing device in time land on the weighing surface of the weighing device. The weighed value data is smoothed to a state in which the weighed values collectively represent one local maximum value to obtain smoothed data, and the smoothed data is sequentially imported as a sampling series, and based on the local maximum value from the sampling series. Calculating its walking periodicity, setting a valid data section from the sampling series based on the walking periodicity, and reading a static weight value of the quadruped animal from the sampling series in the valid data section, The feature is that the sampling period when taking in the smoothed data as the sampling series is such that one of the two legs that lands temporally close to the weighing surface of the weighing device lands before the other lands. The feature is that the time interval is more than twice as long as before. The weight measuring device for a quadrupedal animal according to the present invention includes a weighing means for weighing a quadruped animal walking on a scale, and a weighing value data obtained by the weighing means on a weighing surface of the scale. A cycle in which pairs of two legs that land close to each other in time alternately lands, and the weighing amount when two legs that land close to the weighing surface of the weighing device in time land on the weighing surface. a smoothing means for smoothing the values into a state where the values are grouped together to represent one maximum value; a sampling means for sequentially taking in the smoothed data smoothed by the smoothing means as a sampling series; gait periodicity evaluation means for calculating the gait periodicity based on the gait periodicity; effective data interval setting means for setting an effective data interval from the sampling sequence based on the gait periodicity; calculation means for calculating a static weight value of the animal;
and an output means for outputting the calculated static weight value, the characteristic configuration of which is that the sampling period of the sampling means is a bipod that lands temporally close to the weighing surface of the weighing device. The point is that the time interval from when one of them lands to when the other one lands is set to be more than twice as long.

[0006]

[Operation] According to the method for dynamically measuring the weight of a quadruped animal according to the present invention, the sampling period in which smoothed data is sequentially taken in as a sampling series can be changed by one of the two legs landing temporally close to the weighing surface. If the frequency component caused by two legs landing temporally close to the weighing surface cannot be completely removed by smoothing because the time interval from one landing to the other landing is more than twice the time interval between one landing and the other landing. Even in
It becomes possible to appropriately determine the maximum value due to the frequency component and appropriately determine the static weight value. The dynamic weighing device for quadrupedal animals according to the present invention specifies a device for implementing the above method, and its operation is similar to that of the above method.

[0007]

Therefore, according to the method and apparatus for dynamically measuring the weight of a quadruped animal according to the present invention, it is possible to accurately measure the maximum value of the frequency component caused by the biped that lands temporally close to the weighing surface. It is now possible to capture and measure static weight values with even greater accuracy.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A case in which the present invention is applied to measuring the weight of a cow as a quadrupedal animal will be described below with reference to the drawings. As shown in FIG. 1, a weighing means S having a weighing device 1 as a main part is provided in a walking path where cattle are accustomed to walking, such as between a livestock barn and a pasture. The weight of the cow is measured while walking on it. The walking path is equipped with fences 2 on both sides, and these fences 2
The shelf portions 2a located on both sides of the weighing device 1 prevent the cow from moving laterally outward on the weighing device 1 so as to cause the cow to walk along a straight line or a substantially straight path. It is designed to function as a regulating device. The distance between the left and right shelf portions 2a is made slightly larger than the width of the cow, so that the above-mentioned regulating action can be performed accurately.

[0009] The length of the walking path of the cow on the weighing device 1 is:
The length is determined so that the cow can walk at least a few steps (for example, 3 to 6 steps) with its entire weight on the scale.

[0010] Furthermore, the inlet and outlet of the weighing device 1 are open. The weighing device 1 includes a weighing platform 1a and four load cells 1b (see FIG. 2) arranged to receive and support the four corners of the weighing platform 1a, and the output signal of the load cells 1b is , is output to the measurement processing section H as a weighed value. Incidentally, the output signals of the four load cells 1b are added and output to the measurement processing section H.

Furthermore, a rubber mat 1c is laid on the top surface of the weighing table 1a as a layer of elastic material for shock absorption,
It is designed to reduce the impact when the cow puts its foot down. Furthermore, a tag 3 is hung from the neck of each cow to transmit a radio signal, etc. to separate each cow, and a tag 3 is hung on the side of the weighing device 1.
A tag recognition means 4 consisting of an antenna or the like that receives information transmitted from the tag recognition means 4 is provided, and the information received by the tag recognition means 4 is also input to the measurement processing section H. That is, the measurement processing unit H identifies one cow among the plurality of cows, and measures the weight of the walking cow along with the identifying information.

As shown in FIG. 2, the measurement processing section H:
It is composed of an input means 5 for receiving an electric signal indicating a weighed value sent from the weighing means S, a central control device 6 for determining a static weight value, and an output means 7 for outputting the determined static weight value. ing. The input means 5 includes an amplifier 5A for amplifying an analog electrical signal as weighed value data from the weighing means S, a low-pass filter 5B as a smoothing means for smoothing the output of the amplifier, and a predetermined filter for transmitting the smoothed signal. It consists of an A/D converter 5C as a sampling means for quantizing with sampling time.

[0013] The smoothing of the low-pass filter 5B is performed in such a manner that the weighing values when two legs that land temporally close to the weighing surface of the weighing device 1 land on the weighing surface are collectively expressed as one maximum value. It is done so that Specifically, in the case of a cow, the period in which two of the four legs alternately land on the ground is around 1.5 Hz, and the two legs land on the weighing surface temporally close to each other. The cutoff frequency is set to 2Hz because the time interval or cycle from when one of the legs lands to when the other hits the ground is generally around 3 to 10Hz.

The sampling period of the A/D converter 5C is the time interval from when one of the two legs lands temporally close to the weighing surface of the weighing device 1 until the other one lands. It is set to more than double. Specifically, the time interval from when one of the two legs lands temporally close to the weighing surface until the other lands is generally 5
It is around HZ, but since fluctuations of 3 to 10 HZ are expected, it is set to 20 HZ or higher.

The central control device 6 includes a memory 6A as a storage means for storing time-series data of sampled weight values, and a gait periodicity evaluation means 6B for calculating the gait periodicity from the time-series data. valid data interval setting means 6C for setting a valid data interval from the time series based on the walking periodicity; and calculation means 6D for calculating a static weight value of the running animal from the time series data in the valid data interval. It consists of a microcomputer as the main part. The output means 7 is
In this embodiment, a CRT 7A and a printer 7B are used.

Next, the procedure for determining the static weight of a cow walking on the scale 1 will be explained. FIG. 3 is a graph of the amplifier output of an analog signal indicating the weight value sent from the load cell 1b as the cow walks on the weighing device 1. In this graph, the area indicated by a is the state in which only the front legs of the walking animal are placed on the weighing platform 1a, and the area indicated by b is
The entire weight of the cow is on the scale 1, and the area indicated by c is the state where only the cow's hind legs are on the scale 1a. Further, in the figure, each of the squared portions D corresponds to the weighing value when the two legs that land temporally close to the weighing surface land on the weighing surface.

FIG. 4 shows a signal obtained by applying the low-pass filter 5B to the output of this amplifier. Then, the signal after being applied with the low-pass filter 5B is sampled and quantized.

When the quantized weighing value data (hereinafter simply referred to as weighing value data) becomes larger than a set value K which is slightly larger than the weight of the weighing platform 1a, the cow is placed on the weighing platform 1a. Assuming that the cow has started riding, the weighing value data is sequentially stored in the memory 6A as a sampling series, and the storage is such that when the weighing value data becomes smaller than the set value K, the cow starts to ride on the weighing platform 1a.
It ends as if it had come down from.

Next, the gait periodicity evaluation means 6B determines the maximum value Wmax from the stored sampling data, and from this determines a data group within a predetermined range, in this embodiment, 90% or more of the maximum value. A data group having a value is regarded as a weighing value data group in a state where the entire weight of the cow is placed on the weighing platform 1a. Incidentally, the value obtained by multiplying the maximum value by 0.9 becomes the threshold value for setting the valid data section. In this embodiment, the value is multiplied by 0.9, but a number greater than 0.8 and less than 1 can be used as the value to be multiplied.

In the case shown in FIG. 4, a portion Dx of the weight value when the two legs land temporally close to the weighing surface lands on the weighing surface, even after applying the low-pass filter 5B. They are not smoothed into a state that collectively represents a single maximum value. In this example, the part of Dx that has not been smoothed into a state that collectively represents one maximum value is not the part that represents the maximum value, so the sampling period does not have a negative effect on measurement accuracy, but if If the part of Dx that has not been smoothed to collectively represent a single maximum value shows the maximum value, if the sampling period is not appropriate, the maximum value will not be able to be determined as desired, and the measurement accuracy will deteriorate. There is a risk that it will decline.

Next, the valid data interval setting means 6C sets the range of sampling data equal to or greater than the threshold value as the valid data interval T.

The sampling data within the valid data section thus set is averaged by the calculation means 6D, and the static weight value of the cow running on the weighing platform 1a is calculated. By this measurement, an animal weighing, for example, 1000 kg can be detected with accuracy within an error of 1 kg. In this average calculation, preferably a weighted average calculation method using an appropriate window function, such as a Hamming window, is used, thereby suppressing inconveniences associated with data extraction.

Other examples will be listed below. In the embodiments described so far, the sampling data included in the set valid data interval is simply averaged, but more effective smoothing is performed for time series data with large fluctuations. For this purpose, a moving average calculation method can be adopted. In this case, for example, a moving average value is calculated using a window function such as a Hamming window as needed over the valid data section, using a multiple of the obtained walking cycle as the averaging time. The moving average values thus obtained are evaluated, and the average or maximum value is taken, for example, as the static weight of the walking animal walking on the weighing device 1.

[0024] Instead of immediately considering the weight calculated by the above-mentioned averaging operation as the weight of the walking animal, the weight is multiplied by a correction coefficient determined in advance through experiments, etc., or
It is also possible to further improve the measurement accuracy by adding or subtracting correction values determined in advance through experiments or the like.

In carrying out the present invention, as in the embodiments described so far, all weighing value data from the time when the walking animal starts getting on the weighing platform 1a until the time when it gets off is stored, and then the weight is calculated. Instead of calculating, weight calculation may be performed in parallel with sampling of weighed value data. To explain an example, at the same time as the start of storing weighing value data, processing for determining local maximum values is executed, and when three or more local maximum values are determined, at each point in time when the local maximum values are calculated, the weighing value up to that point is The maximum value of the data is determined, a threshold value is determined from the maximum value in the same manner as in the above embodiment, and the valid data is set from the data above the threshold value to calculate the weight. . If the next maximum value does not appear for a predetermined period of time or more after the last maximum value among the maximum values obtained up to now appears, then at the time when the last maximum value mentioned above appears, The determined weight is displayed as the weight of the walking animal. In this way, even if the walking animal stops unnecessarily on the weighing platform 1a,
It is possible to display the weight, which is a great practical advantage.

Further, in the embodiments described so far, a cow was described as a four-legged animal as the object to be measured, but the present invention can also be applied to other four-legged animals.

[0027] Although reference numerals are written in the claims section for convenience of reference to the drawings, the present invention is not limited to the structure shown in the accompanying drawings.

[Brief explanation of drawings]

[Fig. 1] External view showing the principle configuration of the dynamic weight measurement method for walking animals and its device according to the present invention.

[Figure 2] Block diagram of dynamic weight measurement device

[Figure 3] Diagram showing weighing value analog signal

[Fig. 4] A diagram showing a signal after applying a low-pass filter to the signal according to Fig. 3.

[Explanation of symbols]

1. Weighing device 5B Smoothing means 5C Sampling means 6B Gait periodicity evaluation means 6C Valid data section setting means 6D Arithmetic means 7 Output means S Weighing means

Claims (2)

[Claims] 1. Obtain weighing value data by a weighing means (S) while a quadruped animal walks on a scale (1), and lands temporally close to the weighing surface of the scale (1). A period in which pairs of two legs alternately land on the ground is displayed, and the weighing values when the two legs that land temporally close to the weighing surface of the weighing device (1) land on the weighing surface are summarized. The weighing value data is smoothed to obtain smoothed data so as to be in a state representing one local maximum value, the smoothed data is sequentially taken in as a sampling series, the walking periodicity is calculated from the sampling series, and the Setting a valid data section from the sampling series based on the walking periodicity,
A dynamic weight measurement method for a quadrupedal animal that reads a static weight value of the quadrupedal animal from the sampling series in the effective data interval, the method comprising: adjusting the sampling period when capturing the smoothed data as the sampling series to the weighing device; (1) A method for measuring the dynamic weight of a quadrupedal animal in which the time interval between one of the two legs that lands temporally close to the weighing surface until the other one lands is twice or more. 2. Weighing means (S) for weighing the weight of a quadruped animal walking on a weighing device (1), and the weighing means (S)
The weighing value data obtained by a smoothing means (5B) for smoothing the weighing values when a biped that lands temporally close to the weighing surface lands on the weighing surface to collectively represent one maximum value; and the smoothing means (5
Sampling means (5C) for sequentially capturing the smoothed data smoothed by B) as a sampling series;
gait periodicity evaluation means (6B) for calculating the gait periodicity from the sampling series; valid data interval setting means (6C) for setting a valid data section from the sampling series based on the gait periodicity; A dynamic weight measurement device for a quadruped animal, comprising a calculation means (6D) for calculating a static weight value of the quadruped animal from a sampling series in a data interval, and an output means (7) for outputting the calculated static weight value. The smoothing means (5B) comprises:
The sampling means (5C) is configured to smooth the weighing value data, and the sampling period of the sampling means (5C) is such that one of the two legs that lands temporally close to the weighing surface of the weighing device (1) lands. A dynamic weight measurement device for quadrupeds that is set to a time interval that is at least twice the time interval from when the other side touches the ground.