CN107027650B - A kind of boar abnormal state detection method and device based on PSO-SVM - Google Patents
A kind of boar abnormal state detection method and device based on PSO-SVM Download PDFInfo
- Publication number
- CN107027650B CN107027650B CN201710170147.9A CN201710170147A CN107027650B CN 107027650 B CN107027650 B CN 107027650B CN 201710170147 A CN201710170147 A CN 201710170147A CN 107027650 B CN107027650 B CN 107027650B
- Authority
- CN
- China
- Prior art keywords
- boar
- condition
- individual information
- body temperature
- static
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Abstract
The present invention discloses a kind of boar abnormal state detection method and device based on PSO-SVM, doubtful sick pig can be effectively detected out, and accurately and timely identify boar estrus behavior, reduces artificial observation cost, improves the benefit of raiser.This method comprises: S1, acquisition boar individual information, wherein the individual information includes 3-axis acceleration, body temperature and feeding information；S2, by the individual information carry out feature extraction, construction feature matrix；S3, based on the boar knowledge base pre-established, using the eigenmatrix, judge whether boar sick using rule-based method；S4, identification classification is carried out to boar estrus behavior by the boar estrus behavior identification model based on support vector machines that eigenmatrix input is pre-created, obtain whether heat and heat probability two output, wherein, support vector machines parameter is optimized by particle swarm algorithm when the boar estrus behavior identification model is created.
Description
Technical field
The present invention relates to the processing of livestock and poultry cultivation internet of things data and analysis fields, and in particular to a kind of based on PSO-SVM's
Boar abnormal state detection method and device.
Background technique
In recent years, boar breeding technology is widely popularized with China, the large-scale degree for raising boar is constantly mentioning
It is high.Boar health status is the guarantee of pig production, and the country still monitors livestock by the way of artificial observation mostly at present
Behavior does so and not only takes a substantial amount of time and energy, and the data subjectivity that artificial observation arrives is strong, is unfavorable for essence
Really, stablize, continuously record；Timely and accurately identification the normal heat of sow be obtain sow optimum period breeding time and improve its by
The key problem in technology of tire rate, but in actual production, the means identification oestrus of sow that current most of kind of pig farm is only manually observed
Behavior, many raisers are since carelessness has neglected the heat performance of sow and is difficult to grasp in due course scientific breeding opportunity of sow etc.
Factor, cause breeding time bad or conception rate reduce, greatly affected the normal breeding potential of sow, there are heavy workload,
High labor cost and the problem of inefficiency, the economic benefit of raiser is had an important influence on.Therefore, real-time detection boar is supported
During growing the health status of individual pig and timely and accurately identification oestrus of sow behavior grind as livestock and poultry cultivation field is important
Study carefully content.
Performance is usually the following when oestrus of sow: 1. body temperature is relatively risen usually；2. activity increased significantly；③
It walks about more frequently, mounting behavior；4. appetite slightly subtracts, type of mostly chewing carefully and swallow slowly, Feeding time is relatively extended usually.Wherein live
Momentum and Temperature changing are the most obvious, also mostly carry out the research of oestrus of sow detection in current research according to the two features.
It is showed when boar is sick usual are as follows: 1. body temperature increases；2. lassitude, sleepingly drowsiness；3. appetite stimulator or stopping；
4. diarrhea.Studying at present more is to carry out disease detection for physiological performances such as respiratory disease, diarrhea.
It is less for the sick research detected simultaneously with heat abnormality in boar breeding production at present, and needle
The research with oestrous detection sick to boar is identified according to a certain physiological characteristic when boar is sick or heat, than
If the variation individually according to activity or body temperature is identified and ignores other correlation factors, greatly influence abnormality inspection
The accuracy rate of survey.
Summary of the invention
In view of this, the problem to be solved in the present invention: a kind of boar abnormal state detection method and dress based on PSO-SVM
It sets.
On the one hand, the embodiment of the present invention proposes a kind of boar abnormal state detection method based on PSO-SVM, comprising:
S1, acquisition boar individual information, wherein the individual information includes 3-axis acceleration, body temperature and feeding information；
S2, by the individual information carry out feature extraction, construction feature matrix；
S3, it is planted using the eigenmatrix using the judgement of rule-based method based on the boar knowledge base pre-established
Whether pig is sick；
S4, mould is identified by the boar estrus behavior based on support vector machines that eigenmatrix input is pre-created
Type carries out identification classification to boar estrus behavior, obtain whether heat and heat probability two output, wherein in the boar
Estrus behavior identification model optimizes support vector machines parameter by particle swarm algorithm when creating.
On the other hand, the embodiment of the present invention proposes a kind of boar abnormal state detecting apparatus based on PSO-SVM, comprising:
Acquisition unit, for acquiring boar individual information, wherein the individual information include 3-axis acceleration, body temperature and
Feeding information；
Construction unit, for by carrying out feature extraction, construction feature matrix to the individual information；
First recognition unit, based on the boar knowledge base pre-established, using the eigenmatrix, use is rule-based
Method judges whether boar is sick；
Second recognition unit, for by the way that the eigenmatrix is inputted the boar based on support vector machines being pre-created
Estrus behavior identification model carries out identification classification to boar estrus behavior, obtain whether heat and heat probability two output,
Wherein, support vector machines parameter is optimized by particle swarm algorithm when the boar estrus behavior identification model is created.
Boar abnormal state detection method and device provided in an embodiment of the present invention based on PSO-SVM, comprehensively considers kind
The individual correlation factor such as activity, body temperature, changes in diet of pig, constructs boar knowledge base, is judged by rule-based method
It is whether sick；Method based on support vector machines carries out heat identification, and by particle swarm algorithm to support vector cassification
Device parameter optimizes, and the accuracy rate of boar estrus behavior identification is improved, to improve the conception rate of boar.
Detailed description of the invention
Fig. 1 is that the present invention is based on the flow diagrams of one embodiment of boar abnormal state detection method of PSO-SVM；
Fig. 2 is that the present invention is based on the structural schematic diagrams of one embodiment of boar abnormal state detecting apparatus of PSO-SVM.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical solution in the embodiment of the present invention is explicitly described, it is clear that described embodiment is the present invention
A part of the embodiment, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not having
Every other embodiment obtained under the premise of creative work is made, shall fall within the protection scope of the present invention.
Referring to Fig. 1, the present embodiment discloses a kind of boar abnormal state detection method based on PSO-SVM, comprising:
Individual information when S1, acquisition boar activity, wherein the individual information includes 3-axis acceleration, body temperature and adopts
Eat information；
S2, by the individual information carry out feature extraction, construction feature matrix；
S3, it is planted using the eigenmatrix using the judgement of rule-based method based on the boar knowledge base pre-established
Whether pig is sick；
S4, mould is identified by the boar estrus behavior based on support vector machines that eigenmatrix input is pre-created
Type carries out identification classification to boar estrus behavior, obtain whether heat and heat probability two output, wherein in the boar
Estrus behavior identification model optimizes support vector machines parameter by particle swarm algorithm when creating.
Boar abnormal state detection method provided in an embodiment of the present invention based on PSO-SVM, comprehensively considers the work of boar
The individual correlation factor such as momentum, body temperature, changes in diet, constructs boar knowledge base, judges whether to give birth to by rule-based method
Disease；Method based on support vector machines carries out heat identification, and by particle swarm algorithm to support vector machine classifier parameter
It optimizes, the accuracy rate of boar estrus behavior identification is improved, to improve the conception rate of boar.
In order to achieve the above object, technical solution of the present invention proposes a kind of boar abnormality inspection based on PSO-SVM
Survey method, this method acquire the individual informations such as 3-axis acceleration, body temperature, the dietary amount of boar by awareness apparatus first, and
Carry out the operation such as data consistency detection；Secondly the pretreatment operations such as feature extraction, normalization are carried out, 6 dimensional feature matrixes are constructed；
Then boar knowledge base is established, judges whether boar is sick using rule-based method；Finally creation is based on support vector machines
Boar estrus behavior identification model, wherein being optimized by particle swarm algorithm to support vector machines parameter；To boar heat
Behavior carries out identification classification, includes whether heat and heat probability two outputs.
This method includes four steps:
Individual information acquisition.Kind is acquired by ADXL345 3-axis acceleration sensor, infrared radiation thermometer and self-feeder
The individual informations such as 3-axis acceleration, body temperature, the feeding of pig.Information gathering process are related to acquisition node, routing node, centromere
Point, RS485 and RS232 switching node and PC terminal (information centre).Specific collection process are as follows:
1) ADXL345 3-axis acceleration sensor acquisition node is worn on boar neck, acquires certain moment boar X, Y, Z
3-axis acceleration data；
2) infrared radiation thermometer is combined with self-feeder to acquisition body temperature and feeding information, when boar searches for food automatically into
In feeding station, door of standing is closed, and guarantees that an only boar searches for food every time, ear tag card reader passes through electron ear tage identification record
The time in station is entered and left to boar, blanking device is able to record the feed intake of the boar, and infrared radiation thermometer collects kind
Body temperature when pig searches for food.Wherein the acquisition node of 3-axis acceleration sensor and routing node and central node formed ZigBee without
Gauze network transfers data to central node through routing node, and infrared radiation thermometer, ear tag card reader and self-feeder are adopted
Collect node and central node is transferred data to through routing node by cable；
3) central node is protected by data such as acceleration, body temperature, feedings by RS485 and RS232 wire transmission to PC terminal
It deposits to server.
Eigenmatrix building.Activity, body temperature, changes in diet are more apparent when boar heat, according to this three parameter buildings
Boar estrus behavior eigenmatrix.According to static, mounting, each accounting example of walking, unit time mean body temperature, list in training sample
The eigenmatrix of 6 position time intake time, unit feed intake input vector sign building M*6 dimensions, M is the acquisition of boar individual information
Number of days.Specifically: 1) it can be static, mounting, walking three classes by the crawler behavior rough segmentation of boar according to different standard and purpose.
ADXL345 3-axis acceleration sensor acquires X, Y, Z 3-axis acceleration data, sets training set number of samples as N, corresponds to number of days
For M, the 3-axis acceleration value that t moment acquires is respectivelyTag along sort isThen constituted by 3 axle accelerations
Acceleration signature vector (input vector) X of one N*3 dimensionv, tag along sort (it is static: -1, mounting: 0, walking: 1) for N*1 tie up
Label vector (output vector) Yv。
Data normalization processing is carried out to the feature vector of training set, after being optimized by the training of PSO-SVM algorithm
Boar crawler behavior disaggregated model, using boar crawler behavior disaggregated model to the 3-axis acceleration feature vector (normalizing of boar
After change) handled, to obtain boar crawler behavior classification results, i.e., in the unit time, boar is static, mounting, walking time
Number.
Wherein support vector machines (Support Vector Machine, SVM) is a kind of realization structural risk minimization think of
The method thought, SVM structure exactly like three layer perceptron.Wherein input layer is not make any processing fortune to store input data
It calculates；Middle layer is selected kernel function K (x, xi), i=1,2,3 ..., l by the study to sample set；The last layer is exactly structure
Make classification function.Whole process be equivalent in feature space construct an optimal hyperlane so that between two class samples away from
From maximization.
Given training sample set
T={ (x1,y1),(x2,y2),...(xl,yl)} (1)
In formula, xi--- n-dimensional vector, xi∈Rn, yi--- corresponding output variable, yi∈ R,
The largest interval model classified using approximate non-linear is introduced kernel function the data of luv space and is mapped to higher-dimension
In space, optimization problem is indicated are as follows:
C in formula --- penalty factor, αi--- the Lagrange multiplier of introducing, l --- sample set number, K (xi,
xj) --- kernel function uses Radial basis kernel function as the kernel function of SVM herein.
Particle swarm algorithm (Particle Swarm Optimization, PSO) is by the cooperation between individual in population
A kind of Swarm Intelligent Algorithm of optimal solution is found with information sharing.
Define 1 each optimization problem potential solution can be imagined as D dimension search space on a point, we term it
" particle " (Particle), all particles have the adaptive value (Fitness Value) determined by objective function, each
Particle determines direction and distance that they circle in the air there are one speed, position, speed and the index expression of fitness value three grain
Subcharacter；
It defines the fitness value optimal location being calculated in the undergone position of 2 individuals and is known as individual extreme value；Refer to population
In all particle search to fitness optimal location be known as group's extreme value.
PSO algorithm initializes a group particle first in solution space, and each particle represents extremal optimization problem
One potential optimal solution；Then particle moves in solution space, by tracking individual extreme value and group's extreme value more new individual position,
As soon as the every update time position of particle, calculates a fitness value, and by comparing the fitness value of new particle and individual extreme value,
The fitness value of group's extreme value more new individual extreme value and group's extreme value place；It is empty in solution that particles just follow current optimal particle
Between middle search, by successive ignition, final evolve obtains optimum individual extreme value and group's extreme value place and fitness value.
In iterative process each time, particle updates speed and the position of itself by individual extreme value and group's extreme value, more
New formula is as follows:
In formula,--- the speed of i-th of particle at the kth iteration；
--- the position of i-th of particle at the kth iteration；
c1, c2--- related coefficient variable, value are respectively 1.5,1.7；
r1, r2--- random number；
W --- the inertia weight factor represents the ability of search speed before particle is inherited；
--- position when fitness is optimal in all paths of i-th of particle, i.e., individual extreme value；
--- position when fitness is optimal in the population in kth time iteration, i.e. group's extreme value.
When creating support vector cassification model, there are two the selections of parameter to play vital work to classifying quality
With: first is that penalty factor, i.e., to the tolerance of error, C is higher, and illustrating, which more can't stand, error occurs, is easy over-fitting, and C is got over
It is small, it is easy poor fitting, C is excessive or too small, and generalization ability is deteriorated；Second is that the parameter γ in kernel function, impliedly determines data
Distribution after being mapped to new feature space, γ is bigger, and supporting vector is fewer, and γ value is smaller, and supporting vector is more, and population is calculated
Method is updated according to current group optimal solution search always by creation random population, and continuous iteration is found until meeting condition
Optimal parameter C and γ.
According to above-mentioned boar crawler behavior classification results, as unit of day, according to boar within one day static, mounting,
The number of walking calculates separately static, mounting, each accounting example of walking, sets static, mounting, walking proportion difference in certain day
ForThe daily mean body temperature of boar can be calculated in the body temperature detected according to infrared radiation thermometerAccording to certainly
The boar feeding information of dynamic feeder record can obtain intake timeFeed intakeComprehensive boar crawler behavior, body temperature,
The eigenmatrix of parameter of searching for food building M*6 dimensional feature.
Static, mounting, each accounting example of walking, unit time mean body temperature, unit time intake time, unit according to each day
The eigenmatrix X of 6 input vector sign building M*6 dimensions of feed intakeo, tag along sort (whether heat, be: 0, it is no: 1) for M*1 tie up
Label vector Yo。
Disease detection.Judge whether boar is sick using rule-based method, first creation boar knowledge base；Then needle
The boar individual information acquired in real time is compared with rule base, is judged whether sick.Specifically:
1) normal condition in boar knowledge base is obtained by the means such as books and electronic material, questionnaire survey, expert interviewing
Under individual physiological characteristic information, including information such as body temperature, activity, diet under boar normal condition under different cultivars.
2) initial data got such as is chosen, is classification associated at the processing；Boar knowledge data base is designed, each product are arranged
The information, including body temperature, weight, intake time, feed intake, activity etc. of the boar of kind in normal state.
3) a disease detection rule base is generated, disease detection of the invention focuses on " differentiation of disease ", that is, judges certain
Whether boar is sick, therefore in rule base mainly based on threshold decision；If boar under normal circumstances (in health status and
Estrus), NpsIndicate daily quiescent time accounting, NThIndicate shell temperature peak, NTlIndicate shell temperature minimum,
NfnIndicate average daily feed intake；There are following three conditions:
If meeting any of the above condition, judge the boar for doubtful sick pig；Otherwise, it is determined that for healthy boar.
Heat identification.Creation and training boar heat identification model, carry out heat identification to boar.Specifically: 1) to M*6
The eigenmatrix of dimension is normalized, and by particle swarm algorithm optimum choice support vector machines parameter, creation is based on PSO-
The boar heat identification model of SVM, wherein PSO-SVM algorithm has a detailed description in eigenmatrix building.
2) estrus behavior classification is carried out to test set according to estrus behavior identification model, setting output model is bivector
Y{Yifo, Yopro, wherein YifoIndicate whether heat (be: 0, it is no: 1),Indicate heat probability.
Referring to Fig. 2, the present embodiment discloses a kind of boar abnormal state detecting apparatus based on PSO-SVM, comprising:
Acquisition unit 1, for acquiring the individual information of boar, wherein the individual information includes 3-axis acceleration, body temperature
With feeding information；
The acquisition unit is specifically used for passing through ADXL345 3-axis acceleration sensor, infrared radiation thermometer and automatic feeding
Device acquires corresponding boar individual information.
Construction unit 2, for by carrying out feature extraction, construction feature matrix to the individual information；
The construction unit, is specifically used for:
The 3-axis acceleration is divided into static, mounting and walking three classes by particle swarm algorithm and support vector machines, with
It is unit, and static, mounting, walking the number according to boar within one day calculates separately static, mounting, each accounting example of walking,
And calculate the daily mean body temperature of boar, average Feeding time and average feed intake；
Integrate static boar, mounting, each accounting example of walking, daily mean body temperature, average Feeding time and feed intake
Construction feature matrix.
First recognition unit 3, for based on the boar knowledge base pre-established, using the eigenmatrix, using being based on
The method of rule judges whether boar is sick；
First recognition unit, is specifically used for:
Judge whether the individual information of boar meets one in first condition, second condition, third condition and fourth condition
A condition, if the individual information of boar meets an item in the first condition, second condition, third condition and fourth condition
Part, it is determined that boar is sick, otherwise, it is determined that boar is not sick, wherein the first condition is that the static proportion of boar is big
In boar, daily quiescent time accounting, the second condition are that the mean body temperature of boar is less than boar normal condition under normal circumstances
Lower shell temperature minimum, the third condition are that the mean body temperature of boar is greater than boar shell temperature highest under normal circumstances
Value, the fourth condition are that average feed intake is less than boar average daily feed intake under normal circumstances.
Second recognition unit 4, for by the way that the eigenmatrix is inputted the kind based on support vector machines being pre-created
Pig estrus behavior identification model carries out identification classification to boar estrus behavior, obtain whether heat and heat probability two it is defeated
Out, wherein excellent to the progress of support vector machines parameter by particle swarm algorithm in boar estrus behavior identification model creation
Change.
Boar abnormal state detecting apparatus provided in an embodiment of the present invention based on PSO-SVM, comprehensively considers the work of boar
The individual correlation factor such as momentum, body temperature, changes in diet, constructs boar knowledge base, judges whether to give birth to by rule-based method
Disease；Method based on support vector machines carries out heat identification, and by particle swarm algorithm to support vector machine classifier parameter
It optimizes, the accuracy rate of boar estrus behavior identification is improved, to improve the conception rate of boar.
The present invention has the advantage that
1, two kinds of abnormalities of sick and heat of boar are detected simultaneously, comprehensively considers activity, the body of boar
The variation of three temperature, diet aspects improves abnormal state detection accuracy rate as feature extraction index；The data of every group of measurement are same
Shi Jinhang disease detection and heat identification, so that data improve raiser's productivity effect using maximizing in boar breeding process.
2, according to consultant expert, check that the means such as books electronic bits of data obtain the information such as boar growth standard, and create
Boar knowledge base when carrying out disease detection, is carried out according to differentiation of disease rule in boar knowledge base and the boar individual data items of measurement
It compares, to judge whether sick, algorithm comparison is easy to accomplish, and for detecting that whether sick boar effect be preferable.
3, the identification of boar estrus behavior is carried out by double classification algorithm.In characteristic extraction procedure, using supporting vector
Machine classifier is classified (including static, mounting, walking three classes) to the 3-axis acceleration data of boar；In creation heat identification
When model, as unit of day, by daily static, mounting, walking proportion, mean body temperature, intake time, feed intake as 6
Dimensional feature matrix reuses support vector machine classifier and classifies to boar heat state, wherein passing through particle swarm algorithm
Support Vector Machines Optimized determines support vector machines parameter.By the sorting algorithm twice based on population and support vector machines, mention
High boar abnormal state detection accuracy.
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.Term " on ", "lower" etc. refer to
The orientation or positional relationship shown is to be based on the orientation or positional relationship shown in the drawings, and is merely for convenience of the description present invention and simplifies
Description, rather than the device or element of indication or suggestion meaning must have a particular orientation, constructed and grasped with specific orientation
Make, therefore is not considered as limiting the invention.Unless otherwise clearly defined and limited, term " installation ", " connected ",
" connection " shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or be integrally connected；It can be
Mechanical connection, is also possible to be electrically connected；It can be directly connected, two can also be can be indirectly connected through an intermediary
Connection inside element.For the ordinary skill in the art, above-mentioned term can be understood at this as the case may be
Concrete meaning in invention.
In specification of the invention, numerous specific details are set forth.Although it is understood that the embodiment of the present invention can
To practice without these specific details.In some instances, well known method, structure and skill is not been shown in detail
Art, so as not to obscure the understanding of this specification.Similarly, it should be understood that disclose in order to simplify the present invention and helps to understand respectively
One or more of a inventive aspect, in the above description of the exemplary embodiment of the present invention, each spy of the invention
Sign is grouped together into a single embodiment, figure, or description thereof sometimes.However, should not be by the method solution of the disclosure
Release is in reflect an intention that i.e. the claimed invention requires more than feature expressly recited in each claim
More features.More precisely, as the following claims reflect, inventive aspect is less than single reality disclosed above
Apply all features of example.Therefore, it then follows thus claims of specific embodiment are expressly incorporated in the specific embodiment,
It is wherein each that the claims themselves are regarded as separate embodiments of the invention.It should be noted that in the absence of conflict, this
The feature in embodiment and embodiment in application can be combined with each other.The invention is not limited to any single aspect,
It is not limited to any single embodiment, is also not limited to any combination and/or displacement of these aspects and/or embodiment.And
And can be used alone each aspect and/or embodiment of the invention or with other one or more aspects and/or its implementation
Example is used in combination.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement；And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme should all cover within the scope of the claims and the description of the invention.
Claims (8)
1. a kind of boar abnormal state detection method based on PSO-SVM characterized by comprising
S1, acquisition boar individual information, wherein the individual information includes 3-axis acceleration, body temperature and feeding information；
S2, by the individual information carry out feature extraction, construction feature matrix；
S3, based on the boar knowledge base pre-established, using the eigenmatrix, judge that boar is using rule-based method
It is no sick；
S4, the boar estrus behavior identification model pair based on support vector machines by the way that eigenmatrix input to be pre-created
Boar estrus behavior carries out identification classification, obtain whether heat and heat probability two output, wherein in the boar heat
Support vector machines parameter is optimized by particle swarm algorithm when Activity recognition model creation.
2. the method according to claim 1, wherein the individual information collecting work in the S1 is by information collection
System complete, the information acquisition system by acquisition node, routing node, central node, RS485 and RS232 switching node and
PC terminal composition acquires corresponding boar by ADXL345 3-axis acceleration sensor, infrared radiation thermometer and self-feeder
Body information, acquisition node and routing node and central node form ZigBee wireless network, through routing node by individual information number
According to central node is sent to, individual information data are passed through RS485 and RS232 switching node wire transmission to PC end by central node
End is saved to server.
3. according to the method described in claim 2, it is characterized in that, the S2, comprising:
The 3-axis acceleration is divided into static, mounting and walking three classes by particle swarm algorithm and support vector machines, is with day
Unit, static, mounting, walking the number according to boar within one day calculate separately static, mounting, each accounting example of walking, and count
Calculate the daily mean body temperature of boar, average Feeding time and average feed intake；
Integrate static boar, mounting, each accounting example of walking, daily mean body temperature, average Feeding time and average feed intake
Construction feature matrix.
4. according to the method described in claim 3, it is characterized in that, the S3, comprising:
Judge whether the individual information of boar meets an item in first condition, second condition, third condition and fourth condition
Part, if the individual information of boar meets a condition in the first condition, second condition, third condition and fourth condition,
Then determine that boar is sick, otherwise, it is determined that boar is not sick, wherein the first condition is that the static proportion of boar is greater than
Daily quiescent time accounting, the second condition are that the mean body temperature of boar is less than boar under normal circumstances to boar under normal circumstances
Shell temperature minimum, the third condition are that the mean body temperature of boar is greater than boar shell temperature peak under normal circumstances,
The fourth condition is that average feed intake is less than boar average daily feed intake under normal circumstances.
5. a kind of boar abnormal state detecting apparatus based on PSO-SVM characterized by comprising
Acquisition unit, for acquiring boar individual information, wherein the individual information includes 3-axis acceleration, body temperature and feeding
Information；
Construction unit, for by carrying out feature extraction, construction feature matrix to the individual information；
First recognition unit, for based on the boar knowledge base pre-established, using the eigenmatrix, use to be rule-based
Method judges whether boar is sick；
Second recognition unit, for by the way that the eigenmatrix is inputted the boar heat based on support vector machines being pre-created
Activity recognition model carries out identification classification to boar estrus behavior, obtain whether heat and heat probability two output, wherein
Support vector machines parameter is optimized by particle swarm algorithm in boar estrus behavior identification model creation.
6. device according to claim 5, which is characterized in that the acquisition unit is specifically used for passing through tri- axis of ADXL345
Acceleration transducer, infrared radiation thermometer and self-feeder acquire corresponding boar individual information.
7. device according to claim 6, which is characterized in that the construction unit is specifically used for:
The 3-axis acceleration is divided into static, mounting and walking three classes by particle swarm algorithm and support vector machines, is with day
Unit, static, mounting, walking the number according to boar within one day calculate separately static, mounting, each accounting example of walking, and count
Calculate the daily mean body temperature of boar, average Feeding time and average feed intake；
Integrate static boar, mounting, each accounting example of walking, daily mean body temperature, average Feeding time and average feed intake
Construction feature matrix.
8. device according to claim 7, which is characterized in that first recognition unit is specifically used for:
Judge whether the individual information of boar meets an item in first condition, second condition, third condition and fourth condition
Part, if the individual information of boar meets a condition in the first condition, second condition, third condition and fourth condition,
Then determine that boar is sick, otherwise, it is determined that boar is not sick, wherein the first condition is that the static proportion of boar is greater than
Daily quiescent time accounting, the second condition are that the mean body temperature of boar is less than boar under normal circumstances to boar under normal circumstances
Shell temperature minimum, the third condition are that the mean body temperature of boar is greater than boar shell temperature peak under normal circumstances,
The fourth condition is that average feed intake is less than boar average daily feed intake under normal circumstances.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710170147.9A CN107027650B (en) | 2017-03-21 | 2017-03-21 | A kind of boar abnormal state detection method and device based on PSO-SVM |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710170147.9A CN107027650B (en) | 2017-03-21 | 2017-03-21 | A kind of boar abnormal state detection method and device based on PSO-SVM |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107027650A CN107027650A (en) | 2017-08-11 |
| CN107027650B true CN107027650B (en) | 2019-10-22 |
Family
ID=59533704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710170147.9A Active CN107027650B (en) | 2017-03-21 | 2017-03-21 | A kind of boar abnormal state detection method and device based on PSO-SVM |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107027650B (en) |
Families Citing this family (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107752987A (en) * | 2017-09-18 | 2018-03-06 | 浙江利尔达物联网技术有限公司 | A kind of animal health situation automatic analysis method and system |
| CN107771706B (en) * | 2017-09-18 | 2020-07-31 | 浙江利尔达物芯科技有限公司 | Sow oestrus detection method and system |
| CN107692980A (en) * | 2017-09-18 | 2018-02-16 | 浙江利尔达物联网技术有限公司 | A kind of automatic method for monitoring and analyzing of animal health situation and system |
| CN107711576B (en) * | 2017-09-18 | 2020-09-22 | 浙江利尔达物芯科技有限公司 | Sow oestrus identification method and system |
| CN108354594A (en) * | 2017-12-28 | 2018-08-03 | 杭州攻壳科技有限公司 | A kind of hog condition detection method and device based on wearable device |
| CN108207700B (en) * | 2018-01-15 | 2020-12-15 | 内蒙古大学 | Cow information monitoring method and system |
| CN108717523B (en) * | 2018-04-26 | 2020-07-31 | 华南农业大学 | Sow oestrus behavior detection method based on machine vision |
| CN108494807A (en) * | 2018-05-29 | 2018-09-04 | 广西电网有限责任公司 | Next-generation key message infrastructure network intruding detection system based on cloud computing |
| CN109258508B (en) * | 2018-09-26 | 2020-11-10 | 深圳市倍适沃智能设备有限公司 | Sow oestrus analysis method and device, terminal and computer readable storage medium |
| CN110363247B (en) * | 2019-07-19 | 2021-11-02 | 太原理工大学 | Intelligent monitoring method and system for abnormal diet of live pigs |
| CN110476879B (en) * | 2019-08-26 | 2022-02-22 | 重庆邮电大学 | Milk cow behavior classification and judgment method and device based on multi-label chain type ecological environment |
| CN112970620B (en) * | 2019-12-17 | 2023-07-21 | 中移(成都)信息通信科技有限公司 | Method, apparatus, system, device and medium for detecting estrus status |
| CN111685060B (en) * | 2020-06-10 | 2022-02-08 | 彭东乔 | Method for recognizing oestrus behavior of ruminant based on artificial intelligence |
| CN111914685B (en) * | 2020-07-14 | 2024-04-09 | 北京小龙潜行科技有限公司 | Sow oestrus detection method and device, electronic equipment and storage medium |
| CN112348238A (en) * | 2020-10-27 | 2021-02-09 | 浙江师范大学 | Egg yield prediction PSO-SVM regression model based on principal component analysis |
| CN112293295B (en) * | 2020-10-28 | 2022-03-18 | 内蒙古农业大学 | Milk cow early lameness recognition method and device |
| CN114041426A (en) * | 2020-12-31 | 2022-02-15 | 重庆市六九畜牧科技股份有限公司 | Backup sow management pigsty |
| CN114883010A (en) * | 2022-04-26 | 2022-08-09 | 深圳市中融数字科技有限公司 | Livestock survival state judging method and device, storage medium and terminal equipment |
| CN116034904B (en) * | 2023-03-31 | 2023-06-13 | 华南农业大学 | Pig health monitoring system and method based on track type inspection robot |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2504706A1 (en) * | 2002-11-04 | 2004-05-21 | Nycomed Danmark Aps | Coating of a particulate material with an organic solvent-based coating composition |
| CN105075886B (en) * | 2015-07-27 | 2018-01-02 | 河南科技大学 | A kind of milk cow automatic feeding device and corresponding intelligent feeding systems |
| CN105104291B (en) * | 2015-07-27 | 2017-11-21 | 河南科技大学 | A kind of milk cow motion state method of discrimination and corresponding intelligent feeding method |
| CN105654141A (en) * | 2016-01-06 | 2016-06-08 | 江苏大学 | Isomap and SVM algorithm-based overlooked herded pig individual recognition method |
| CN106199305B (en) * | 2016-07-01 | 2018-12-28 | 太原理工大学 | Underground coal mine power supply system dry-type transformer insulation health state evaluation method |
-
2017
- 2017-03-21 CN CN201710170147.9A patent/CN107027650B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN107027650A (en) | 2017-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107027650B (en) | A kind of boar abnormal state detection method and device based on PSO-SVM | |
| CN105091938B (en) | Livestock birds health condition monitoring method and system | |
| Pavlovic et al. | Classification of cattle behaviours using neck-mounted accelerometer-equipped collars and convolutional neural networks | |
| CN110476879B (en) | Milk cow behavior classification and judgment method and device based on multi-label chain type ecological environment | |
| Butuner et al. | Classification of deep image features of lentil varieties with machine learning techniques | |
| Tian et al. | Real-time behavioral recognition in dairy cows based on geomagnetism and acceleration information | |
| Lu et al. | An automatic splitting method for the adhesive piglets’ gray scale image based on the ellipse shape feature | |
| Raina et al. | A study on various techniques for plant leaf disease detection using leaf image | |
| Chouhan et al. | A deep learning approach for the classification of diseased plant leaf images | |
| Nandhini et al. | Automatic detection of leaf disease using CNN algorithm | |
| Meshram et al. | Border-Square net: a robust multi-grade fruit classification in IoT smart agriculture using feature extraction based Deep Maxout network | |
| Devi et al. | Inception Adaptive Gradient L2 Regularized Learning Rate CNN for Strawberry Leaf disease Detection | |
| Bonik et al. | A Convolutional Neural Network Based Potato Leaf Diseases Detection Using Sequential Model | |
| CN113420709A (en) | Cattle face feature extraction model training method and system and cattle insurance method and system | |
| CN116982572A (en) | Milk cow feeding behavior monitoring method | |
| CN116912025A (en) | Livestock breeding information comprehensive management method and system based on cloud edge cooperation | |
| Bijanzadeh et al. | Determining the most important features contributing to wheat grain yield using supervised feature selection model | |
| CN109543761B (en) | Method and device for classifying plant suitable habitat | |
| Malathi et al. | A review on rice crop disease classification using computational approach | |
| CN111950598A (en) | Method for individually classifying swimming crab groups based on K-Means algorithm and application | |
| Septiarini et al. | Oil Palm Leaf Disease Detection on Natural Background Using Convolutional Neural Networks | |
| Nasiri et al. | Estimating the Feeding Time of Individual Broilers via Convolutional Neural Network and Image Processing | |
| Lu et al. | Lettuce Plant Trace-Element-Deficiency Symptom Identification via Machine Vision Methods | |
| Praba et al. | Smart fish farming recommendation system using k-means algorithm | |
| CN115019162A (en) | Silkworm detection method based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |