KR20170028571A - Apparatus and method for detecting abnormal domestic annimal - Google Patents

Apparatus and method for detecting abnormal domestic annimal Download PDF

Info

Publication number
KR20170028571A
KR20170028571A KR1020150125358A KR20150125358A KR20170028571A KR 20170028571 A KR20170028571 A KR 20170028571A KR 1020150125358 A KR1020150125358 A KR 1020150125358A KR 20150125358 A KR20150125358 A KR 20150125358A KR 20170028571 A KR20170028571 A KR 20170028571A
Authority
KR
South Korea
Prior art keywords
group
groups
livestock
vectors
disease
Prior art date
Application number
KR1020150125358A
Other languages
Korean (ko)
Inventor
이웅섭
Original Assignee
경상대학교산학협력단
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 경상대학교산학협력단 filed Critical 경상대학교산학협력단
Priority to KR1020150125358A priority Critical patent/KR20170028571A/en
Publication of KR20170028571A publication Critical patent/KR20170028571A/en

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01K—ANIMAL HUSBANDRY; CARE OF BIRDS, FISHES, INSECTS; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
    • A01K29/00—Other apparatus for animal husbandry
    • A01K29/005—Monitoring or measuring activity, e.g. detecting heat or mating
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization

Abstract

An apparatus and method for determining whether a disease has occurred in a livestock that can determine whether or not a disease has occurred in the livestock. The apparatus for determining whether or not a disease has occurred in the livestock includes monitoring body information such as body temperature, momentum, voice and sound of the livestock, and classifying the animals into a plurality of groups using the monitored biological information. The device for determining whether a disease has occurred in the livestock determines whether or not the disease has occurred in the livestock depending on how the size of each group changes with time.

Description

[0001] APPARATUS AND METHOD FOR DETECTING ABNORMAL DOMESTIC ANNIMAL [0002]

The following embodiments are directed to an apparatus and a method for detecting unhealthy or abnormal livestock. More particularly, the present invention relates to an apparatus and a method for detecting unhealthy animals in an abnormal state, And more particularly,

The consumption of meat and dairy products is increasing due to the improvement of living standards. In addition, automation and unmanned livestock farming are expected due to shortage of farmers in rural areas and expected population sensitivity.

If we can identify the abnormal individuals in the premises early and prevent the spread of infectious diseases by separating them from other individuals, we can greatly reduce the labor force for the livestock industry. Moreover, It is expected to be a big help.

Recently, the smart housing industry has been attracting attention as it has attached sensor to livestock through the development of Internet of things, and monitoring and analyzing biometric information of livestock. However, there has been a great deal of research on how to analyze the collected biometric information.

The consumption of meat and dairy products is increasing due to the improvement of living standards. In addition, automation and unmanned livestock farming are expected due to shortage of farmers in rural areas and expected population sensitivity.

If we can identify the abnormal individuals in the premises early and prevent the spread of infectious diseases by separating them from other individuals, we can greatly reduce the labor force for the livestock industry. Moreover, It is expected to be a big help.

Recently, the smart housing industry has been attracting attention as it has attached sensor to livestock through the development of Internet of things, and monitoring and analyzing biometric information of livestock. However, there has been a great deal of research on how to analyze the collected biometric information.

According to an exemplary embodiment of the present invention, there is provided a monitoring apparatus comprising: a monitoring unit for monitoring biometric information from livestock; a vector generating unit that includes the monitored biometric information as an element and generates a vector corresponding to each animal; And a judging unit for judging whether or not the disease has occurred in the animals according to the size change of the groups.

Here, if the change in size of each group is equal to or greater than the first threshold value, the determination unit may determine that an infectious disease has occurred in the animals.

If the size of the smallest group among the groups is less than or equal to a second threshold value, the determination unit may determine that a disease has occurred in at least one of the animals.

In addition, the classifier may classify the vectors into a plurality of groups using Expectation Maximization.

Here, the classifying unit may classify the vectors into a plurality of groups by alternately repeating the following E and M steps.

[Step E]

here,

Figure pat00001
Is the number of vectors included in the jth group in the tth calculation.
Figure pat00002
Is an event indicating whether i-th monitored biometric information is included in the j-th group,
Figure pat00003
Shows statistical characteristics such as mean, variance, and size of two groups.
Figure pat00004
Is the ratio of the number of vectors included in the jth group to the total number of vectors used in the classification.
Figure pat00005
Average
Figure pat00006
, And dispersion
Figure pat00007
In Gaussian distribution.

[Step M]

Figure pat00008

here,

Figure pat00009
Is the ratio of the number of vectors included in the jth group to the total number of vectors used in the classification.

Figure pat00010

Figure pat00011

here,

Figure pat00012
Is the mean value of the first group,
Figure pat00013
Is the mean value of the second group.

The biometric information may include at least one of body temperature, momentum, utterance, and water supply of each of the livestock.

According to yet another exemplary embodiment, there is provided a method for monitoring bio-information, comprising the steps of: monitoring biometric information from livestock; generating a vector corresponding to each of the livestock including the monitored biometric information as an element; Classifying the animals into groups, and determining whether disease has occurred in the animals according to the size change of the groups.

Here, the determining step may determine that a communicable disease has occurred in the animals if the size change of each group is equal to or greater than a first threshold value.

If the size of the smallest group among the groups is less than or equal to a second threshold value, it may be determined that the disease has occurred in at least one of the animals.

Also, the classifying step may classify the vectors into a plurality of groups using Expectation Maximization.

Here, the classifying step may alternately repeat the following E and M steps to classify the vectors into a plurality of groups.

[Step E]

here,

Figure pat00014
Is the number of vectors included in the jth group in the tth calculation.
Figure pat00015
Is an event indicating whether i-th monitored biometric information is included in the j-th group,
Figure pat00016
Shows statistical characteristics such as mean, variance, and size of two groups.
Figure pat00017
Is the ratio of the number of vectors included in the jth group to the total number of vectors used in the classification.
Figure pat00018
Average
Figure pat00019
, And dispersion
Figure pat00020
In Gaussian distribution.

[Step M]

Figure pat00021

here,

Figure pat00022
Is the ratio of the number of vectors included in the jth group to the total number of vectors used in the classification.

Figure pat00023

Figure pat00024

here,

Figure pat00025
Is the mean value of the first group,
Figure pat00026
Is the mean value of the second group.

The biometric information may include at least one of body temperature, momentum, utterance, and water supply of each of the livestock.

According to the embodiments described below, it is possible to quickly judge whether or not a disease occurrence in a livestock based on livestock biometric information.

According to the following examples, it can be judged whether an infectious disease has occurred in a group of livestock or only an individual has a disease.

1 is a view showing a method for judging whether or not a disease of a domestic animal has occurred according to an exemplary embodiment.
2 is a block diagram showing the structure of an apparatus for determining whether a disease has occurred in a livestock according to an exemplary embodiment.
3 is a diagram showing a concept of classifying vectors including biometric information of a livestock in a vector space according to an exemplary embodiment.
4 is a diagram illustrating a concept of classifying vectors including livestock biometric information on a vector space using an expectation value maximization algorithm.
FIG. 5 is a diagram showing a case where an abnormality occurs in some of the objects or all objects are normal.
6 to 8 are diagrams illustrating a concept of determining whether disease has occurred in livestock according to changes in size of groups.
FIG. 9 is a flowchart illustrating a method of determining whether a disease has occurred in a livestock according to another exemplary embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings.

Demand for livestock products is gradually increasing due to the development of culture and economy due to improvement of diet, national income, population increase, and improvement of education level. In order to cover the demand for increasing livestock products, cattle, pigs and other livestock are increasingly being raised in the farms.

If the livestock are raised collectively in the housing, the infectious disease may develop and the infection may spread to the livestock within the housing. Therefore, it is very important to determine whether the disease has occurred to the animals in the house or whether the invented disease is an epidemic.

1 is a view showing a method for judging whether or not a disease of a domestic animal has occurred according to an exemplary embodiment. According to an exemplary embodiment, a sensor may be installed in the housing and the installed sensors may be used to monitor the vital signs from the livestock 110, 120, 130 in the housing.

Here, the living body signal may include at least one of body temperature, momentum, utterance sound, and water supply amount of each of the livestock 110, 120, and 130.

According to one aspect, the disease occurrence occurrence determination device 140 may analyze the monitored bio-signals to determine whether disease has occurred in the domestic animals or whether the disease is a communicable disease.

2 is a block diagram showing the structure of an apparatus for determining whether a disease has occurred in a livestock according to an exemplary embodiment. According to an exemplary embodiment, the apparatus 200 for determining whether or not a disease exists in a domestic animal includes a monitoring unit 210, a vector generating unit 220, a classifying unit 230, and a determining unit 240.

The monitoring unit 210 monitors biometric information sensed from cattle.

The vector generation unit 220 generates a vector corresponding to each animal using the monitored biometric information. For example, the vector generation unit 220 may generate a vector corresponding to a specific entity so as to include biometric information monitored from a specific entity as an element.

The classifying unit 230 classifies the generated vectors into a plurality of groups on a vector space.

The concept of classifying the generated vectors into a plurality of groups on the vector space will be described in detail with reference to FIG.

3 is a diagram showing a concept of classifying vectors including biometric information of a livestock in a vector space according to an exemplary embodiment. The abscissa of the graph shown in Fig. 3 is the body temperature in the biometric information of the livestock, and the ordinate of the graph is the momentum in the biometric information of the livestock. In FIG. 3, for convenience of description, a two-dimensional vector space is taken as an example. However, the vector space of the type of the monitored biometric information can be enlarged to N dimensions.

In FIG. 3, the vectors 311, 312, and 313 may be classified into one group 310, and the vectors 321 and 322 may be classified into another group 320. Since the momentum and body temperature of normal individuals not infected with the disease are often within a certain range, normal individuals are highly likely to be classified into specific groups. Also, since the momentum and body temperature of diseased individuals often deviate from a certain range, individuals who are infected with the disease are more likely to be classified into different groups.

In FIG. 3, vectors 311, 312, and 313 are vectors corresponding to normal individuals, and vectors 321 and 322 are vectors corresponding to an abnormal entity.

4 is a diagram illustrating a concept of classifying vectors including livestock biometric information on a vector space using an expectation value maximization algorithm.

The classifier 230 shown in FIG. 2 can classify the vectors into a plurality of groups using an expectation maximization algorithm (Expectation Maximization). Here, the expected value maximization algorithm is a classification method that can be used even when the statistical characteristics of the vectors to be classified are unknown. The expectation maximization algorithm can classify the vectors into a plurality of groups by iteratively calculating the parameters maximizing the maximum likelihood.

The expectation maximization algorithm iteratively performs the E-step (E-step) to predict the conditional probability and the M-step to predict the parameters of the Gaussian distribution using the conditional probability.

Here, the step E may be performed according to the following equation (1).

[Equation 1]

Figure pat00027

here,

Figure pat00028
Is the conditional probability that the i-th monitored biometric information is included in the j-th group. The suffix t at the top indicates the number of calculations.
Figure pat00029
Is an event indicating whether i-th monitored biometric information is included in the j-th group,
Figure pat00030
Shows statistical characteristics such as mean, variance, and size of two groups.
Figure pat00031
Is the ratio of the number of vectors included in the jth group to the total number of vectors used in the classification.
Figure pat00032
Average
Figure pat00033
, And dispersion
Figure pat00034
In Gaussian PDF (Probability Density Function)
Figure pat00035
Is the probability that the

Also, step M may be performed as shown in the following equations (2) to (4).

&Quot; (2) "

Figure pat00036

here,

Figure pat00037
Is the ratio of the number of vectors included in the jth group to the total number of vectors used in the classification.

&Quot; (3) "

Figure pat00038

here,

Figure pat00039
Is the mean value of the first group.

&Quot; (4) "

Figure pat00040

here,

Figure pat00041
Is the mean value of the second group.

Therefore, some parameters are initialized and used in the first step, and some errors may occur in the initial classification due to the initialized parameters.

4 (a) shows vectors 410 corresponding to a normal entity and vectors 420 corresponding to an ideal entity are arranged on a vector space.

4B shows that a part of the vectors corresponding to the normal entity and a part of the vectors corresponding to the ideal entity are classified into the same group 430 and 440 according to the erroneous setting of the initial value of the expected value maximization algorithm .

The expectation maximization algorithm can classify the vectors normally by repeating steps E and M, even if there are some errors in the initialized parameters.

4C shows that the vectors 450 corresponding to the normal entity and the vectors 460 corresponding to the ideal entity are normally classified as the expected value maximization algorithm is repeatedly performed.

The determination unit 240 may determine whether disease has occurred in the animals according to the results of classification of the vectors as shown in FIG.

5 to 8, a description will be given of a configuration in which the judging unit judges whether disease has occurred in the animals.

FIG. 5 is a diagram showing a case where an abnormality occurs in some of the objects or all objects are normal.

Even in the case of using the maximum likelihood algorithm, it is impossible to successfully classify the abnormal entity and the normal entity in all cases.

5 (a) shows a case where the abnormal entities 520 are located in a space separated from one another in the vector space. In order to classify the abnormal entities 520 into the same group, some of the normal entities 510 are included in the same group as the abnormal entities 520. Therefore, it is difficult to judge whether or not a disease has occurred, because some individuals are simultaneously included in a group of normal individuals 510 and a group of abnormal individuals 520.

FIG. 5 (b) is a diagram showing a case where all the entities are normal. Since the expectation maximization algorithm classifies the vectors in the vector space mechanically into a plurality of groups, normal individuals are classified into different groups (530, 540).

Therefore, it is not possible to accurately determine whether or not a disease has occurred in livestock with only the expectation value maximization algorithm.

According to the exemplary embodiment, the determination unit 240 of FIG. 2 may determine whether the disease has occurred to the animals according to the size change of the groups in which the vectors are classified.

FIG. 6 (a) is a diagram showing the distribution of vectors when no disease has occurred in the animals.

If the disease does not occur in the animals, the statistical properties of all individuals can be assumed to be the same. Thus, entities can be divided into groups of similar size.

FIG. 6B shows the size change of the classified groups, where the horizontal axis represents time and the vertical axis represents the ratio of the number of individuals included in a specific group among the total population. If the disease does not occur in the livestock, the ratio of the number of individuals in a particular group among the total population converges at a value of around 0.5.

If the sizes of the classified groups are not changed according to time, they are kept constant, and when the sizes are similar to each other, the determination unit 240 can determine that all the individuals are normal individuals in which no disease has occurred. According to one aspect, when the variation of the size of all the classified groups is below the second threshold and the ratio of the number of the individuals included in the specific group among the entire population is about 0.5, the determination unit 240 determines It can be judged that it is a normal individual that has not occurred.

FIG. 7 (a) is a diagram showing the distribution of vectors in a case where diseases that are not infectious diseases occur in some livestock. If some livestock are not infectious diseases, normal and abnormal individuals may be divided into different groups.

FIG. 7B shows the size change of the classified groups, where the horizontal axis represents time and the vertical axis represents the ratio of the number of individuals included in a specific group among the total population. Unless the outbreak is an infectious disease, the ratio of the number of individuals in a particular group among the total population does not change significantly over time. In addition, since the number of abnormal individuals is generally smaller than the number of normal individuals, the ratio of the number of individuals included in a specific group is much smaller than 0.5 and converges at a value larger than 0.

If the sizes of the classified groups are not changed according to the time but remain constant but they are different from each other, the judging unit 240 can judge that some diseases have occurred in some of the objects rather than infectious diseases. According to one aspect, when the change amount of the size of all the classified groups is equal to or less than the second threshold value and the ratio of the number of the individuals included in the specific group among the whole population numbers is around 0 or 1, the determination unit 240 determines It can be judged that an individual has a disease other than an infectious disease.

8 (a) is a diagram showing the distribution of the initial vectors in which an infectious disease occurs. If some livestock are not infectious diseases, normal and abnormal individuals may be divided into different groups.

FIG. 8 (b) is a diagram showing the distribution of vectors after the infectious disease occurs. Over time, the disease was spread to individuals who were normal.

FIG. 8 (c) shows the size change of the classified groups, with the horizontal axis representing time and the vertical axis representing the ratio of the number of individuals included in a particular group among the total population. In the case of an epidemic, the proportion of abnormal individuals increases over time.

If the sizes of the classified groups change with time, the determination unit 240 may determine that a communicable disease has occurred in some of the objects. According to one aspect, if the amount of change in the size of any one of the grouped groups is equal to or greater than the first threshold, the determination unit 240 can determine that a disease other than a contagious disease has occurred in some of the individuals.

The operation of the apparatus for determining whether a disease has occurred in the animals described in Figs. 2 to 8 can be summarized by the following algorithm.

Step 1:

Figure pat00042
Initialize

here,

Figure pat00043
Is an average value of each group to be classified,
Figure pat00044
Is the variance of each group to be classified. Also,
Figure pat00045
(The number of vectors included in the group) of each group to be classified.

Step 2:

Figure pat00046
,
Figure pat00047
Initialize

here,

Figure pat00048
An epidemic counter,
Figure pat00049
Is a non-communicable counter. According to one side,
Figure pat00050
,
Figure pat00051
Can be initialized to '0'.

Step 3:

Figure pat00052
,
Figure pat00053
With counter threshold

Figure pat00054
Wow
Figure pat00055
To the counter threshold
Figure pat00056
. if
Figure pat00057
Wow
Figure pat00058
Everyone has a counter threshold
Figure pat00059
The step 4 and the following steps are performed.

Step 4: Monitor biometric information from livestock.

Step 5: Average value of each group previously calculated

Figure pat00060
To
Figure pat00061
, And the variance of each group previously calculated
Figure pat00062
To
Figure pat00063
And the size of each group previously calculated
Figure pat00064
To
Figure pat00065
.

Step 6: The monitored biometric information is included as an element, a vector corresponding to each animal is generated, and the generated vectors are classified into a plurality of groups in a vector space using an expectation value maximization algorithm.

Depending on the newly categorized group,

Figure pat00066
Lt; / RTI >

Step 7: For each group, it is determined whether the size change amount of the group is between the first threshold value and the second threshold value. (5) " (5) "

&Quot; (5) "

Figure pat00067

here,

Figure pat00068
Is a first threshold value,
Figure pat00069
Is a second threshold value. If the size change amount of the group is between the first threshold value and the second threshold value, the counter of the group is incremented by '1'.

Step 8: The smallest value among the sizes of the groups is compared with the third threshold value. This can be expressed by the following equation (6).

&Quot; (6) "

Figure pat00070

here,

Figure pat00071
Is a third threshold value. If the smallest value among the sizes of the groups is smaller than or equal to the third threshold value, the counter of the group is incremented by '1'.

Thereafter, the flow advances to step 4 to compare the incremented counter value with the counter threshold value, and judge whether the disease has occurred in the animals. For example, an epidemic counter

Figure pat00072
≪ / RTI >
Figure pat00073
If the value is larger, it can be judged that an infectious disease has occurred between the livestock. Or non-communicable counter
Figure pat00074
≪ / RTI >
Figure pat00075
If the value is larger, it can be judged that a disease other than an infectious disease has occurred among the livestock.

FIG. 9 is a flowchart illustrating a method of determining whether a disease has occurred in a livestock according to another exemplary embodiment.

In step 910, the device for determining whether a disease has occurred in the livestock monitors biological information from the livestock. Here, the biometric information may include at least one of body temperature, momentum, utterance sound, and water supply amount of each livestock.

In step 920, the apparatus for determining whether a disease has occurred in a livestock includes monitored biometric information as an element, and generates a vector corresponding to each livestock. For example, a device for determining whether a disease has occurred in a livestock may generate a vector corresponding to a specific entity so as to include biometric information monitored from a specific entity as an element.

In step 930, the apparatus for determining whether a disease has occurred in a livestock classifies the generated vectors into a plurality of groups on a vector space. According to one aspect, the apparatus for determining whether a disease has occurred in a livestock can classify vectors into a plurality of groups using an expectation value maximization algorithm. According to another aspect, the apparatus for determining whether or not a disease has occurred in a livestock may be divided into a plurality of groups by alternately repeating the steps E and M described in Equations (1) to (4).

In step 940, the device for determining whether a disease has occurred in the livestock can determine whether the disease has occurred in the livestock according to the size change of the classified groups. For example, when the size change of each group is equal to or greater than the first threshold value, the device for determining whether a disease occurs in a livestock can determine that an infectious disease has occurred in the livestock. According to another embodiment, the apparatus for determining whether a disease has occurred in a livestock may determine that disease occurs in at least one of the livestock if the size of the smallest group is less than or equal to a second threshold value.

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

110, 120: Normal object
130: Object over
140: Apparatus for determining whether a disease has occurred in a livestock

Claims (13)

A monitoring unit for monitoring biometric information from the animals;
A vector generating unit that includes the monitored biometric information as an element and generates a vector corresponding to each animal;
A classifier for classifying the vectors into a plurality of groups in a vector space;
A judgment unit for judging whether disease has occurred in the animals according to the size change of the groups,
Wherein the animal is an animal. The method according to claim 1,
Wherein the determination unit determines that an infectious disease has occurred in the animals if the size change of each group is equal to or greater than a first threshold value. The method according to claim 1,
Wherein the determination unit determines that a disease has occurred in at least one of the livestock if the size of the smallest group is less than a second threshold value among the groups. The method according to claim 1,
Wherein the classifying unit classifies the vector into a plurality of groups using an expectation maximization algorithm. The method according to claim 1,
Wherein the classifying unit repeatedly repeats the following steps E and M to classify the vector into a plurality of groups.

[Step E]

Figure pat00076


here,
Figure pat00077
Is the number of vectors included in the jth group in the tth calculation.
Figure pat00078
Is an event indicating whether i-th monitored biometric information is included in the j-th group,
Figure pat00079
Shows statistical characteristics such as mean, variance, and size of two groups.
Figure pat00080
Is the ratio of the number of vectors included in the jth group to the total number of vectors used in the classification.
Figure pat00081
Average
Figure pat00082
, And dispersion
Figure pat00083
In Gaussian distribution.

[Step M]

Figure pat00084


here,
Figure pat00085
Is the ratio of the number of vectors included in the jth group to the total number of vectors used in the classification.

Figure pat00086

Figure pat00087


here,
Figure pat00088
Is the mean value of the first group,
Figure pat00089
Is the mean value of the second group.
The method according to claim 1,
Wherein the biometric information includes at least one of a body temperature, a momentum, a voice, and a water supply amount of each of the livestock. Monitoring biometric information from the animals;
Generating a vector corresponding to each of the livestock by including the monitored biometric information as an element;
Classifying the vectors into a plurality of groups on a vector space;
Determining whether disease has occurred in the animals according to the size change of the groups
The method comprising the steps of: 8. The method of claim 7,
Wherein the determining step determines that the infectious disease occurs in the livestock if the size change of each group is equal to or greater than the first threshold value. 8. The method of claim 7,
Wherein the determining step determines that a disease has occurred in at least one of the livestock if the size of the smallest group is less than a second threshold value among the groups. 8. The method of claim 7,
Wherein the classifying step classifies the vectors into a plurality of groups using an expectation maximization algorithm. 8. The method of claim 7,
Wherein the classifying step comprises alternately repeating steps E and M below to classify the vector into a plurality of groups.

[Step E]

Figure pat00090


here,
Figure pat00091
Is the number of vectors included in the jth group in the tth calculation.
Figure pat00092
Is an event indicating whether i-th monitored biometric information is included in the j-th group,
Figure pat00093
Shows statistical characteristics such as mean, variance, and size of two groups.
Figure pat00094
Is the ratio of the number of vectors included in the jth group to the total number of vectors used in the classification.
Figure pat00095
Average
Figure pat00096
, And dispersion
Figure pat00097
In Gaussian distribution.

[Step M]

Figure pat00098


here,
Figure pat00099
Is the ratio of the number of vectors included in the jth group to the total number of vectors used in the classification.

Figure pat00100

Figure pat00101


here,
Figure pat00102
Is the mean value of the first group,
Figure pat00103
Is the mean value of the second group. 8. The method of claim 7,
Wherein the biometric information includes at least one of a body temperature, a momentum, a voice, and a water supply of each of the livestock. A computer-readable recording medium having recorded thereon a program for executing the method according to any one of claims 7 to 12.
KR1020150125358A 2015-09-04 2015-09-04 Apparatus and method for detecting abnormal domestic annimal KR20170028571A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1020150125358A KR20170028571A (en) 2015-09-04 2015-09-04 Apparatus and method for detecting abnormal domestic annimal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1020150125358A KR20170028571A (en) 2015-09-04 2015-09-04 Apparatus and method for detecting abnormal domestic annimal

Related Child Applications (1)

Application Number Title Priority Date Filing Date
KR1020170176394A Division KR101906456B1 (en) 2017-12-20 2017-12-20 Apparatus and method for detecting abnormal state using artificial intelligence

Publications (1)

Publication Number Publication Date
KR20170028571A true KR20170028571A (en) 2017-03-14

Family

ID=58460038

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020150125358A KR20170028571A (en) 2015-09-04 2015-09-04 Apparatus and method for detecting abnormal domestic annimal

Country Status (1)

Country Link
KR (1) KR20170028571A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20190143701A (en) * 2018-06-21 2019-12-31 엘지이노텍 주식회사 Apparatus and method for detecting abnormal object and imaging device comprising the same
KR20190143518A (en) * 2018-06-07 2019-12-31 엘지이노텍 주식회사 Apparatus and method for determining abnormal object
KR20210151556A (en) * 2020-06-05 2021-12-14 한국전자기술연구원 Method for Detecting Livestock Abnormal Signs in a Farm by Monitoring Activity Energy based on the Sound of Livestock Behavior

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20190143518A (en) * 2018-06-07 2019-12-31 엘지이노텍 주식회사 Apparatus and method for determining abnormal object
KR20190143701A (en) * 2018-06-21 2019-12-31 엘지이노텍 주식회사 Apparatus and method for detecting abnormal object and imaging device comprising the same
KR20210151556A (en) * 2020-06-05 2021-12-14 한국전자기술연구원 Method for Detecting Livestock Abnormal Signs in a Farm by Monitoring Activity Energy based on the Sound of Livestock Behavior

Similar Documents

Publication Publication Date Title
US10332254B2 (en) Computer Aided Diagnosis (CAD) apparatus and method
JP6941123B2 (en) Cell annotation method and annotation system using adaptive additional learning
Thida et al. Laplacian eigenmap with temporal constraints for local abnormality detection in crowded scenes
US10607147B2 (en) Estimating a number of occupants in a region
US9824296B2 (en) Event detection apparatus and event detection method
CN105122270B (en) The method and system of people is counted using depth transducer
KR101607934B1 (en) The method for monitoring communicable disease and system using the method, recording medium for performing the method
JP4970531B2 (en) A method for automatically characterizing the behavior of one or more objects.
EP2889805A2 (en) Method and system for emotion and behavior recognition
KR101906456B1 (en) Apparatus and method for detecting abnormal state using artificial intelligence
JP5889019B2 (en) Label adding apparatus, label adding method and program
TWI567660B (en) Multi-class object classifying method and system
Beyan et al. Detection of abnormal fish trajectories using a clustering based hierarchical classifier
Arablouei et al. Animal behavior classification via deep learning on embedded systems
Haque et al. Semi supervised adaptive framework for classifying evolving data stream
KR102506029B1 (en) Apparatus and method for monitoring growing progress of livestock individual based on image
KR102582506B1 (en) Monitoring system for pig house disease management
KR20170028571A (en) Apparatus and method for detecting abnormal domestic annimal
JP6950504B2 (en) Abnormal candidate extraction program, abnormal candidate extraction method and abnormal candidate extraction device
CN116597377A (en) Intelligent monitoring management method and system for cattle breeding
Ozdemir et al. Comparison of deep learning techniques for classification of the insects in order level with mobile software application
US20230385699A1 (en) Data boundary deriving system and method
US20160155068A1 (en) Information processing apparatus, information processing method, and recording medium for classifying input data
Colonna et al. A distributed approach for classifying anuran species based on their calls
CN115049889A (en) Storage medium and inference method

Legal Events

Date Code Title Description
A201 Request for examination
E902 Notification of reason for refusal
E601 Decision to refuse application
AMND Amendment
A107 Divisional application of patent