CN208095182U - A kind of agricultural, which is examined, beats integrated machine system - Google Patents
A kind of agricultural, which is examined, beats integrated machine system Download PDFInfo
- Publication number
- CN208095182U CN208095182U CN201820513325.3U CN201820513325U CN208095182U CN 208095182 U CN208095182 U CN 208095182U CN 201820513325 U CN201820513325 U CN 201820513325U CN 208095182 U CN208095182 U CN 208095182U
- Authority
- CN
- China
- Prior art keywords
- sprinkling
- data
- model
- grade
- plant
- 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
Landscapes
- Catching Or Destruction (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
It is examined the utility model discloses a kind of agricultural and beats integrated machine system, including unmanned plane body, monitoring system, sprinkling system, flight control system, flight control system includes data transmission interface, processor and data sink.The utility model will monitor system, flight control system and sprinkling system and integrate on unmanned plane body same again, can calculate fountain height in real time while monitoring, and accurately sprayed, can save pesticide or fertilizer, reduce environmental pollution.
Description
Technical field
The utility model relates to crops plant protection technology fields, examine more particularly, to a kind of agricultural and beat integrated machine system.
Background technique
Current relatively advanced agricultural plant protection method mainly divides two ways：
The first is the man-machine plant protection method being most widely used, and agriculture base personnel will go to scene to carry out monitoring tune on the spot first
It looks into, determines whether farmland has pest and disease damage, if the situations such as water shortage or fertilizer deficiency.It is then based on the guidance of agriculture base personnel, unmanned plane is planted
Protect fertilising or application that team recycles unmanned plane to carry out large area.However it is multiple, manually monitoring on the spot for large area is not only time-consuming
Laborious and investigation coverage is not high；And the amount of pesticide, fertilizer spray is often just set before aircraft flight, is one
Fixed amount, but whole block crop growth conditions are different, some local fertilizer deficiencys or has pest and disease damage, but some places are strong
Health, if whole block is all sprayed by an amount, the waste of pesticide and fertilizer and the pollution to environment certainly will be will cause.
Second is more advanced unmanned plane monitoring and unmanned plane plant protection method, and this method is first with unmanned monitoring machine
Field-grown situation is monitored, field-grown condition monitor data and monitoring figure are generated, to precisely detect in field eachly
The pest and disease damage of point, water content and fertilising situation.Plant protection team is further according to field-grown condition monitor data for different serious journeys
It is applied fertilizer or is administered in the place of degree.But the above method needs different types of unmanned plane to carry out asynchronous operation, causes
Information on time and space disconnects, it is likely that in plant protection drone operation, the position of pest has occurred that variation, very great Cheng
The accuracy of application is affected on degree.
Integrated machine system is beaten it would therefore be highly desirable to provide a kind of agricultural and examine, to promote the accuracy of fertilising or application.
Utility model content
It is examined in view of this, the present invention provides a kind of agricultural and beats integrated machine system, solved and apply fertilizer in the prior art
Or application accuracy reduce the technical issues of.
In order to solve the above-mentioned technical problem, it is examined the utility model proposes a kind of agricultural and beats integrated machine system and method.
The agricultural, which is examined, beats integrated machine system, including：
Unmanned plane body；
Monitoring system, the monitoring system include high-spectrum remote-sensing monitor, and the high-spectrum remote-sensing monitor is for obtaining
Take the high-spectral data of plant in farmland；
Sprinkling system, the sprinkling system include water pump, liquid storing barrel and spray head；
Flight control system, the flight control system include data transmission interface, processor and data sink,
In, the data transmission interface is respectively connected with the high-spectrum remote-sensing monitor and the processor, the data receiver
Device is connected with the processor.
Further, the monitoring system further includes：High definition imager, for obtaining the high definition figure of sample farmland plant
Picture.
Further, the processor further includes flight control modules.
Further, the data transmission interface is RS232 serial ports.
Compared with prior art, the utility model realizes following beneficial effect：
Monitoring system, flight control system and sprinkling system are integrated on same unmanned plane body by the utility model,
The fountain height of pesticide or agricultural fertilizer etc. can be calculated in real time while monitoring, and accurately be sprayed, can save pesticide
Or fertilizer, it reduces environmental pollution.
Detailed description of the invention
It is combined in the description and the attached drawing for constituting part of specification shows the embodiments of the present invention, and
And together with its explanation for explaining the principles of the present invention.
Fig. 1 is that a kind of agricultural provided by the embodiment of the utility model examines the functional block diagram for beating integrated machine system；
Fig. 2 is that agricultural provided by the embodiment of the utility model examines the functional block diagram for making the processor of integrated machine system；
Fig. 3 is that another agricultural provided by the embodiment of the utility model examines the functional block diagram for beating integrated machine system；
Fig. 4 is that agricultural provided by the embodiment of the utility model examines dozen one sprinkling control method flow chart in real time.
Specific embodiment
The various exemplary embodiments of the utility model are described in detail now with reference to attached drawing.It should be noted that：Unless another
It illustrates outside, the component and the positioned opposite of step, numerical expression and numerical value otherwise illustrated in these embodiments is unlimited
The scope of the utility model processed.
Be to the description only actually of at least one exemplary embodiment below it is illustrative, never as to this is practical
Novel and its application or any restrictions used.
Technology, method and apparatus known to person of ordinary skill in the relevant may be not discussed in detail, but suitable
In the case of, the technology, method and apparatus should be considered as part of specification.
It is shown here and discuss all examples in, any occurrence should be construed as merely illustratively, without
It is as limitation.Therefore, other examples of exemplary embodiment can have different values.
It should be noted that：Similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, then in subsequent attached drawing does not need that it is further discussed.
Embodiment 1
Present embodiments provide a kind of agricultural and examine and beat integrated machine system, can using unmanned plane to crops in farmland into
Row monitoring, and be administered or apply fertilizer simultaneously, according to the agricultural land information that monitoring system real-time monitors, calculate required pesticide or fertilizer
The fountain height of material, so that fertilising and application are more accurate in real time, is reduced so that targetedly crops are administered or be applied fertilizer
The waste of pesticide or fertilizer, while reducing environmental pollution, labour has been liberated, has been improved work efficiency.Fig. 1 is that this is practical new
The agricultural that type embodiment 1 provides examines the functional block diagram for beating integrated machine system, and Fig. 2 is the agricultural that the utility model embodiment 1 provides
The functional block diagram for making the processor of integrated machine system is examined, such as Fig. 1 and Fig. 2, which includes：
Unmanned plane body (not shown), for carrying monitoring system 10 and sprinkling system 30, to agriculture in farmland
Crop is monitored and sprays fertilizer/pesticide.Plant protection drone in the prior art can be used in unmanned plane, only need to be by each module collection
At in existing plant protection drone, cost is saved.
Monitoring system 10, monitoring system 10 include high-spectrum remote-sensing monitor 101, and high-spectrum remote-sensing monitor 101 is integrated
On unmanned plane body, for obtaining the high-spectral data of plant in farmland, wherein high-spectral data includes the light of multiple wave bands
Modal data；Airborne hyperspectral imager, such as SOC710GX airborne hyperspectral imaging can be used in high-spectrum remote-sensing monitor 101
Spectrometer, small in size, light-weight, installation is simple, optical property and stability are high, spectral region covers 400-1000nm, is suitable for
Application is installed on unmanned plane and small aircraft.
Sprinkling system 30, sprinkling system 30 include water pump 302, liquid storing barrel 303 and spray head 301；Liquid storing barrel 303 is fixedly installed
On unmanned plane body, liquid storing barrel 303 storage pesticide or fertilizer.The inlet of water pump 302 protrudes into liquid storing barrel 303, water pump
302 liquid outlet connects spray head 301, when water pump 302 works can by liquid storing barrel 303 pesticide or fertilizer be transported to spray head
301 spray, and achieve the effect that automatic spraying, can save manpower, improve working efficiency.Wherein water pump 302 can use DC30
The water pump of series, DC40 series or DC50 series, the size of above-mentioned series water pump is light-weight between 3CM-5CM, convenient for peace
Dress.
Flight control system 20, flight control system 20 include data transmission interface 203, processor 201 and data receiver
Device 202, wherein data transmission interface 203 be RS232 serial ports, respectively with 201 phase of high-spectrum remote-sensing monitor 101 and processor
Connection, for high-spectral data to be transmitted to processor 201.
Data sink 202 is connected with processor 201, for receiving pretreatment condition information and being input to processor
201, wherein pretreatment condition information includes farmland plant kind information, farmland plant planting patterns information, farmland plant fertility
Period information and/or medicine fertilizer information to be sprayed.Data sink 202 receives external information by wired or wireless way,
Before unmanned machine operation, user needs the type and growth period according to current Tanaka crops, passes through data sink
202 by pretreatment condition information input to processor 201, and furthermore data sink 202 is additionally provided with hardware or software is formed
Mode switch button, switch between fertilizer for controlling sprinkling system 30 spraying insecticide and spray.
Specifically, human-computer interaction interface is provided on data sink 202 in a kind of specifically embodiment, it is man-machine
Control button is provided on interactive interface, which includes crop varieties select button and plant growth period button, is made
Article kind select button may include wheat, corn, rice, cotton etc., and user waits for the variety of crops of operation according to Tanaka, lead to
Cross the selection that crop varieties select button carries out variety of crops.Plant growth period button is for dividing locating for crops not
Same growth phase, such as the growth period of certain crops is divided into the first stage to the 5th stage according to sequencing, that is, have
There are five growth period button, user selects growth period button according to the growth period of current crops.
Processor 201 includes：
Data preprocessing module 2011 presets the corresponding relationship of pretreatment condition information and sensitive band, due to bloom
The high-spectral data that spectrum remote sensing monitoring instrument 101 monitors has the spectroscopic data of thousands of a wave bands, and data processing amount is larger, and on
In the spectroscopic data of thousand wave bands, according to the difference of variety of crops and growth period, corresponding sensitive band is also different, and
To variety of crops and growth period that Mr. Yu determines, the spectroscopic data of sensitive band with respect to its all band spectroscopic data,
Resolution highest, it is more preferable to best embody crop growthing state validity, therefore, first by experiment and it is empirically determined go out difference
Pretreatment condition information and sensitive band corresponding relationship, actual monitoring be administered or apply fertilizer during, data prediction
Module 2011 filters out that corresponding with pretreatment condition information is received several are quick from the spectroscopic data of thousands of a wave bands
The spectroscopic data for feeling wave band, obtains pretreated high-spectral data, to greatly reduce the workload of system, improves
The processing speed of spectroscopic data.For example, wheat best embodies wheat growth state in the spectroscopic data when being in a certain period
Spectroscopic data be m and n-th of wave band spectroscopic data, user in import wheat on control button and which in stage,
Data sink 202 is connecing wheat and the Information input processing device 201 of its growth period, data preprocessing module 2011
After receiving the collected high-spectrum remote sensing data of high-spectrum remote-sensing monitor 101, from wheat is wherein filtered out in m and n-th
The corresponding high-spectral data of wave band drops to the greatest extent under the premise of the growth conditions of crop can be accurately identified by guaranteeing
The workload of low processor 201 accelerates data processing speed.
Class computing module 2012 is sprayed, presets neural network sprinkling model, it includes crop that neural network, which sprays model,
Pest model and crop fertilizer model, wherein the input vector of crop disease and insect model is high-spectral data, and output vector is
The sprinkling grade of pesticide；The input vector of crop fertilizer model is high-spectral data, and output vector is the sprinkling grade of fertilizer.
When pretreated high-spectral data is input to sprinkling class computing module 2012 by data preprocessing module 2011
Afterwards, neural network sprinkling model obtains corresponding sprinkling grade.For example, the pest and disease damage degree of crop is divided into one to five, totally five
A grade, each grade respectively correspond a sprinkling grade, and corresponding fountain height increases each grade step by step from low to high,
After pretreated spectroscopic data inputs neural network model, neural network model recognizes the pest and disease damage degree when preceding crop
Seriously, output sprinkling grade is five poles, that is, indicates that the dose for needing to spray when preceding crop is maximum dose.
Fountain height computing module 2013, presets fountain height computation model, and the dependent variable of fountain height computation model is sprinkling
Amount, independent variable are sprinkling grade, and fountain height computing module 2013 is used to be calculated according to sprinkling grade and fountain height computation model
To the fountain height, wherein fountain height is the volume of the pesticide that spray head 301 sprays in the unit time or fertilizer, can specifically be led to
Cross the power control fountain height of control water pump 302, the i.e. voltage of control water pump 302 work；It can also be by controlling water pump
302 working time control fountain height.
Specifically, fountain height computation model is as follows：
Q=Max*Y, wherein when 1<=X<When=4, Y=0.25* (X-1), as X=5, Y=1, wherein Q is described
Fountain height, X are the sprinkling grade, and Y is sprinkling ratio.
For example, high-spectrum remote-sensing monitor 101 collects the EO-1 hyperion letter in the region when unmanned plane flies to a certain region
Data are ceased, processor 201 identifies that the sprinkling grade in the region is 3 after receiving hyperspectral information data, and (i.e. when X=3) passes through
Dose conversion formula obtains Y=0.25*2=0.5, Q=Max*0.5=0.5Max, i.e. the practical dose that adds in the region is maximum
The half of dose；It when unmanned plane flies to subsequent region, and monitors that the sprinkling grade of subsequent region is 5 (i.e. when X=5), passes through
Dose conversion formula obtains Q=Max, i.e., the dose that region actual needs adds is maximum dose Max, to realize to difference
The effect that the Crop Root in region is administered according to its disease grade specific aim, reduces the waste of pesticide.
Flight control modules, flight control modules are for controlling the unmanned plane body according to predetermined altitude and predetermined flight
Speed flies at a constant speed.
Control module 2014 is sprayed, is the controlling terminal of sprinkling system 30, for controlling sprinkling system 30 according to fountain height
It is sprayed.
High-spectrum remote-sensing monitoring is integrated on same unmanned plane body by the system with medicine calculation and sprinkling, passes through height
Spectroscopic data determines fountain height, and compared to directly fountain height is determined by machine vision picture, data processing amount is small, real-time
Height sees that the effect of operation when seeing can be applied in real time according to the crop growth situation real-time monitored conducive to realization side overlap
Medicine or fertilising improve the accuracy of application or fertilising, reduce human cost, improve work efficiency, meanwhile, it is sprayed determining
When grade, the screening of sensitive band has been carried out to high-spectral data, has further decreased data processing amount, has promoted real-time.
Embodiment 2
The present embodiment on the basis of embodiment 1, provides a kind of preferred agricultural and examines and beat integrated machine system, related place
It can refer to the description of above-described embodiment 1, specifically, Fig. 3 is that another agricultural provided by the embodiment of the utility model examines dozen one
The functional block diagram of machine system, as shown in Figure 1, Figure 2 and Figure 3, which includes：
Unmanned plane body (not shown), for carrying monitoring system 10 and sprinkling system 30, to agriculture in farmland
Crop is monitored and sprays fertilizer/pesticide.Plant protection drone in the prior art can be used in unmanned plane, only need to be by each module collection
At in existing plant protection drone, cost is saved.
Monitoring system 10, monitoring system 10 include high-spectrum remote-sensing monitor 101, and high-spectrum remote-sensing monitor 101 is integrated
On unmanned plane body, for obtaining the high-spectral data of plant in farmland, wherein high-spectral data includes the light of multiple wave bands
Modal data.Monitoring system 10 further includes high definition imager 102, and for obtaining the high-definition image of sample farmland plant, every vertical frame dimension is clear
The growing state of farmland plant in characterization image predetermined unit area.
Picture recognition module 40, picture recognition module 40 identifies high-definition image using support vector machines, is planted
Object growth information, wherein plant growth information include lesion area ratio, dead leaf rate, plant leaf shape, plant leaf color,
Plant tassel shape, spot pattern, scab color, plant lodging information and/or plant grain distribution rule, wherein scab face
Ratio shared by scab, dead leaf rate refer to the list of high-definition image characterization in unit area of the product than referring to high-definition image characterization
Ratio shared by dead leaf in plane product.Support vector machines (SVM) algorithm can the lesion area ratio of crop in automatic detection image,
The plant growths information such as dead leaf rate, leaf tassel shape, after determining plant growth information, the empirically determined high definition out
The corresponding sample of image sprays grade.
Model construction module 50, model construction module 50 is used to sample high-spectral data be input, with sample sprinkling etc.
Grade is output, is trained to neural network model, and neural network sprinkling model is obtained.Specifically, in building neural network mould
When the training sample of type, unmanned plane is by high-spectrum remote-sensing monitor 101 and high definition imager 102 simultaneously in sample farmland
The growing state of crops is monitored, a large amount of high-spectral data of acquisition as sample high-spectral data, meanwhile, acquisition
Largely high-definition image corresponding with sample high-spectral data.
Picture recognition module 40 identifies high-definition image, after the growth information for recognizing crop, according to previous big
Amount pest and disease damage tests the big data and agronomy expert appraisal obtained, the growth letter based on the crop that every high-definition image is reacted
Breath, obtains sprinkling grade corresponding to every high-definition image, wherein sprinkling grade includes pesticide spraying grade and fertilizer spray
Sprinkling grade classification is five grades (1-5 grade more high required dose or fertilizer amount is higher) in the present embodiment by grade,
To after high-definition image and the corresponding relationship of sprinkling grade, using with sample high-spectral data corresponding to high-definition image as training set
In input vector sample training, building are carried out to neural network model to spray grade as the output vector in training set
Neural network sprays model out.It in actual job, works without high definition imager 102, need to only acquire high-spectral data
Model, which is sprayed, by neural network accurately and rapidly calculates sprinkling grade.
Flight control system 20, flight control system 20 include data transmission interface 203, processor 201 and data receiver
Device 202, wherein data transmission interface 203 can be S232 serial ports, respectively with high-spectrum remote-sensing monitor 101 and processor
201 are connected, and for high-spectral data to be transmitted to processor 201, data transmission stability is good and speed is fast.
Data sink 202 is connected with processor 201, for receiving pretreatment condition information and being input to processor
201, wherein pretreatment condition information includes farmland plant kind information, farmland plant planting patterns information, farmland plant fertility
Period information and/or medicine fertilizer information to be sprayed.
Processor 201 includes：
Data preprocessing module 2011 presets the corresponding relationship of pretreatment condition information and sensitive band, due to bloom
The high-spectral data that spectrum remote sensing monitoring instrument 101 monitors has the spectroscopic data of thousands of a wave bands, and data processing amount is larger, and on
In the spectroscopic data of thousand wave bands, according to the difference of variety of crops and growth period, corresponding sensitive band is also different, and
To variety of crops and growth period that Mr. Yu determines, the spectroscopic data of the spectroscopic data of sensitive band with respect to its all band
Resolution highest best embodies crop growthing state, and validity is more preferable, therefore, passes through experiment and empirically determined difference out first
Pretreatment condition information and sensitive band corresponding relationship, actual monitoring be administered or apply fertilizer during, data prediction
Module 2011 filters out that corresponding with pretreatment condition information is received several are quick from the spectroscopic data of thousands of a wave bands
The spectroscopic data for feeling wave band, obtains pretreated high-spectral data, to greatly reduce the workload of system, improves
The processing speed of spectroscopic data.For example, wheat best embodies wheat growth state in the spectroscopic data when being in a certain period
Spectroscopic data be m and n-th of wave band spectroscopic data, user in import wheat on control button and which in stage,
Data sink 202 is connecing wheat and the Information input processing device 201 of its growth period, data preprocessing module 2011
It receives after the collected high-spectrum remote sensing data of high-spectrum remote-sensing monitor 101 from wheat is wherein filtered out in m and n-th of wave
The corresponding high-spectral data of section reduces to the greatest extent under the premise of the growth conditions of crop can be accurately identified by guaranteeing
The workload of processor 201 accelerates data processing speed.
Class computing module 2012 is sprayed, presets neural network sprinkling model, it includes crop that neural network, which sprays model,
Pest model and crop fertilizer model, wherein the input vector of crop disease and insect model is high-spectral data, and output vector is
The sprinkling grade of pesticide；The input vector of crop fertilizer model is high-spectral data, and output vector is the sprinkling grade of fertilizer.When
After pretreated high-spectral data is input to sprinkling class computing module 2012 by data preprocessing module 2011, neural network
Model is sprayed it can be concluded that spraying grade accordingly.
Fountain height computing module 2013, presets fountain height computation model, and the dependent variable of fountain height computation model is sprinkling
Amount, independent variable are sprinkling grade, and fountain height computing module 2013 is used to be calculated according to sprinkling grade and fountain height computation model
To the fountain height, wherein fountain height is the amount of liquid that spray head 301 sprays in the unit time, specifically can be by controlling water pump
The voltage that 302 power control fountain height, i.e. control water pump 302 work；When can also be by the work of control water pump 302
Between control fountain height.
Sprinkling system 30, sprinkling system 30 include water pump 302, liquid storing barrel 303 and spray head 301；Liquid storing barrel 303 is fixedly installed
On unmanned plane body, liquid storing barrel 303 storage pesticide or fertilizer, and liquid storing barrel 303 can only store pesticide and fertilizer every time
One of.The inlet of water pump 302 protrudes into explosive barrel, and the liquid outlet of water pump 302 connects spray head 301, when water pump 302 works
Can by liquid storing barrel 303 pesticide or fertilizer be transported to spray head 301 and spray, achieve the effect that automatic spraying, people can be saved
Power improves working efficiency.
It is examined using the agricultural that the embodiment provides and beats integrated machine system, when establishing neural network sprinkling model, based on height
The plant growth information of clear image reaction divides sprinkling grade, on the one hand neural network sprinkling model directly sets up EO-1 hyperion number
According to the relationship with sprinkling grade, guarantee that the speed calculated fountain height meets the requirement of side overlap operation, on the other hand, mind
It is the relationship of high-spectral data and high-definition image through what network sprinkling model substantially reacted, so that the sprinkling grade determined
Accuracy is higher.
Embodiment 3
The present embodiment, which proposes a kind of agricultural, to be examined and beats one sprinkling control method in real time, realizes that plant protection drone is raw to crop
The effect of the monitoring of long message progress synchronous with spraying operation, specifically, Fig. 4 is that agricultural provided by the embodiment of the utility model is examined
One sprinkling control method flow chart in real time is beaten, as shown in Figure 4 (the agricultural that can be provided referring concurrently to Fig. 1 to Fig. 3), the embodiment
Examining dozen one, sprinkling control method includes the following steps in real time：
S101：Obtain the high-definition image of sample farmland plant；
Specifically, being acquired using image of the high definition imager 102 to farmland plant, the fixation of high definition imager 102 is set
It sets on unmanned plane body.
S102：The high-spectral data for obtaining plant in the sample farmland, obtains sample high-spectral data；
Specifically, while carrying out Image Acquisition using high definition imager 102, using high-spectrum remote-sensing monitor 101
The high-spectral data of plant in farmland is obtained, airborne hyperspectral imager can be used in high-spectrum remote-sensing monitor 101, such as
SOC710GX airborne hyperspectral imaging spectrometer, it is small in size, light-weight, installation is simple, optical property and stability are high, spectrum
Range covers 400-1000nm, is suitable for installing application on unmanned plane and small aircraft.
S103：The high-definition image is identified using support vector machines, obtains plant growth information；
Wherein, the plant growth information include lesion area ratio, dead leaf rate, plant leaf shape, plant leaf color,
Plant tassel shape, spot pattern, scab color, plant lodging information and/or plant grain distribution rule.
S104：The corresponding sample sprinkling grade of the high-definition image is determined according to the plant growth information；
Specifically, the identification of the remote sensing big data and agronomy expert obtained according to previous a large amount of pest and disease damages experiment, root
Show that sample corresponding to high-definition image sprays grade according to plant growth information.Wherein sprinkling grade include pesticide spraying grade with
And fertilizer spray grade, it by sprinkling grade classification is five grades (1-5 grade more high required doses or fertilizer amount in the present embodiment
It is higher).
S105：It is input with the sample high-spectral data, is output with sample sprinkling grade, to neural network mould
Type is trained, and obtains the neural network sprinkling model；
Obtain high-definition image and spray grade corresponding relationship after, with sample EO-1 hyperion number corresponding to high-definition image
Neural network model is carried out to spray grade as the output vector in training set according to as the input vector in training set
Sample training guarantees that unmanned plane can be fast while flight to improve the recognition speed and calculating speed of processor 201
Speed comes out rating calculation is sprayed, and realizes that monitoring carries out simultaneously with sprinkling.
S106：Obtain the high-spectral data of plant in farmland；
Wherein, the high-spectral data includes the spectroscopic data of multiple wave bands, optionally, the spectrum including 1024 wave bands
Data.
S107：Receive pretreatment condition information；
Wherein, the pretreatment condition information includes farmland plant kind information, farmland plant planting patterns information, farmland
Plant growthdevelopmental stage information and/or medicine fertilizer information to be sprayed, in this step, data sink 202 pass through wired or wireless
Mode receives external information, before unmanned machine operation, when user needs the type and growth according to current Tanaka crops
Phase, by data sink 202 by pretreatment condition information input to processor 201.
Furthermore data sink 202 is also additionally provided with the mode switch button that hardware or software are formed, for controlling spray
It spills system 30 and switches between fertilizer spraying insecticide and spray.
Specifically, human-computer interaction interface is provided on data sink 202 in a kind of specifically embodiment, it is man-machine
Control button is provided on interactive interface, which includes crop varieties select button and plant growth period button, is made
Article kind select button may include wheat, corn, rice, cotton etc., variety of crops choosing of the user according to Tanaka with operation
Select corresponding crop choice button.Plant growth period button is used to divide different stages of growth locating for crops, such as
The growth period of certain crops is divided into first stage to the 5th stage according to sequencing, i.e., tool is pressed there are five growth period
Button, user select plant growth period button according to the growth period of current crops.
S108：According to the corresponding relationship of preset the pretreatment condition information and sensitive band, from the multiple wave band
Spectroscopic data in select the spectroscopic data of sensitive band corresponding with the pretreatment condition information is received, obtain pre-
Treated high-spectral data；
Since the high-spectral data that high-spectrum remote-sensing monitor 101 monitors has the spectroscopic data of thousands of a wave bands, number
It is larger according to treating capacity, and in the spectroscopic data of thousands of a wave bands, it is corresponding according to the difference of variety of crops and growth period
Sensitive band is also different, and to the variety of crops and growth period that Mr. Yu determines, the spectroscopic data of sensitive band is with respect to it
The resolution highest of the spectroscopic data of all band, it is more preferable to best embody crop growthing state validity, therefore, passes through experiment first
With it is empirically determined go out different pretreatment condition information and sensitive band corresponding relationship, in the mistake that actual monitoring is administered or applies fertilizer
Cheng Zhong, data preprocessing module 2011 filter out from the spectroscopic data of thousands of a wave bands and receive pretreatment condition information phase
The spectroscopic data of several corresponding sensitive bands obtains pretreated high-spectral data, to greatly reduce system
Workload, improve the processing speed of spectroscopic data.
S109：The pretreated high-spectral data is inputted into preset neural network and sprays model, to be sprayed
Grade；
Wherein, the input vector of the neural network sprinkling model is the high-spectral data, and output vector is the spray
Spill grade, wherein neural network sprinkling model includes crop disease and insect model and crop fertilizer model, crop disease and insect model
Input vector high-spectral data, output vector are the sprinkling grade of pesticide；The input vector high-spectral data of crop fertilizer model,
Output vector is the sprinkling grade of fertilizer.After the corresponding relationship of the growth information and sprinkling grade that obtain crop, with crop
The corresponding relationship of growth information and sprinkling grade is input with sample high-spectral data as training set, and sprinkling grade is output
A large amount of sample training is carried out, to improve the recognition speed and calculating speed of processor 201.Model is sprayed in neural network
After building, unmanned plane can carry out plant protection operation, and when operation works without high definition imager 102, need to only acquire EO-1 hyperion
Data can accurately and rapidly calculate sprinkling grade.
S110：Fountain height is calculated according to the sprinkling grade and fountain height computation model；
Wherein, the dependent variable of the fountain height computation model is the fountain height, and independent variable is the sprinkling grade, described
Fountain height is the volume of the pesticide that spray head 301 sprays in the unit time or fertilizer；Specifically, fountain height computation model is as follows：
Q=Max*Y, wherein when 1<=X<When=4, Y=0.25* (X-1), as X=5, Y=1, wherein Q is described
Fountain height, X are the sprinkling grade, and Y is sprinkling ratio.
S111：Sprinkling system 30 is controlled according to the fountain height to be sprayed.
The power control fountain height of control water pump 302, the i.e. voltage of control water pump 302 work can specifically be passed through；
The working time control fountain height of control water pump 302 can also be passed through.
This method carries out mind using the corresponding relationship of a large amount of sample high-spectral data combination high-definition image and sprinkling grade
Through network training, obtain neural network sprinkling model, can be improved flight control system to the processing speed of high-spectral data and
Accuracy directlys adopt high-spectrum remote-sensing monitoring without using high definition imager after neural network sprays model construction
Instrument acquires the high-spectral data of farmland plant, and neural network is inputted after pre-processing to collected high-spectral data and sprays mould
Type can immediately arrive at sprinkling grade, accurately be sprayed to control sprinkling system, be different from the prior art in height
The method that clear image is analyzed and calculates degree of disease, method provided by the utility model greatly reduce data processing amount,
Improve processing speed and accuracy, it is ensured that unmanned plane calculates sprinkling grade in real time while flight monitoring and goes forward side by side
The synchronous sprinkling of row.
Although being described in detail by some specific embodiments of the example to the utility model, this field
It is to be understood by the skilled artisans that example above merely to be illustrated, rather than in order to limit the scope of the utility model.This
Field it is to be understood by the skilled artisans that can not depart from the scope of the utility model and spirit in the case where, to above embodiments
It modifies.The scope of the utility model is defined by the following claims.
Claims (4)
1. a kind of agricultural, which is examined, beats integrated machine system, which is characterized in that including：
Unmanned plane body；
Monitoring system, the monitoring system include high-spectrum remote-sensing monitor, and the high-spectrum remote-sensing monitor is for obtaining agriculture
The high-spectral data of plant in field；
Sprinkling system, the sprinkling system include water pump, liquid storing barrel and spray head；
Flight control system, the flight control system include data transmission interface, processor and data sink, wherein
The data transmission interface is respectively connected with the high-spectrum remote-sensing monitor and the processor, the data sink
It is connected with the processor.
2. agricultural according to claim 1, which is examined, beats integrated machine system, which is characterized in that
The monitoring system further includes：High definition imager, for obtaining the high-definition image of sample farmland plant.
3. agricultural according to claim 1, which is examined, beats integrated machine system, which is characterized in that
The processor further includes flight control modules.
4. agricultural according to claim 1, which is examined, beats integrated machine system, which is characterized in that
The data transmission interface is RS232 serial ports.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201820513325.3U CN208095182U (en) | 2018-04-11 | 2018-04-11 | A kind of agricultural, which is examined, beats integrated machine system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201820513325.3U CN208095182U (en) | 2018-04-11 | 2018-04-11 | A kind of agricultural, which is examined, beats integrated machine system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN208095182U true CN208095182U (en) | 2018-11-16 |
Family
ID=64120184
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201820513325.3U Active CN208095182U (en) | 2018-04-11 | 2018-04-11 | A kind of agricultural, which is examined, beats integrated machine system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN208095182U (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022257139A1 (en) * | 2021-06-11 | 2022-12-15 | 深圳市大疆创新科技有限公司 | Plant state determination method, terminal, and computer-readable storage medium |
| CN117115687A (en) * | 2023-08-02 | 2023-11-24 | 江苏商贸职业学院 | Unmanned aerial vehicle accurate fertilization method and system based on artificial intelligence technology |
-
2018
- 2018-04-11 CN CN201820513325.3U patent/CN208095182U/en active Active
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022257139A1 (en) * | 2021-06-11 | 2022-12-15 | 深圳市大疆创新科技有限公司 | Plant state determination method, terminal, and computer-readable storage medium |
| CN117115687A (en) * | 2023-08-02 | 2023-11-24 | 江苏商贸职业学院 | Unmanned aerial vehicle accurate fertilization method and system based on artificial intelligence technology |
| CN117115687B (en) * | 2023-08-02 | 2024-04-09 | 江苏商贸职业学院 | Unmanned aerial vehicle accurate fertilization method and system based on artificial intelligence technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108684282A (en) | A kind of agricultural, which is examined, beats integrated machine system and sprinkling control method | |
| CN108693119B (en) | Intelligent pest and disease damage investigation and printing system based on unmanned aerial vehicle hyperspectral remote sensing | |
| Chen et al. | An AIoT based smart agricultural system for pests detection | |
| CN111582055A (en) | Aerial pesticide application route generation method and system for unmanned aerial vehicle | |
| Arakeri et al. | Computer vision based robotic weed control system for precision agriculture | |
| CA2740503C (en) | Variable rate sprayer system and method of variably applying agrochemicals | |
| CN105173085B (en) | Unmanned plane variable farm chemical applying automatic control system and method | |
| CN109470299A (en) | A kind of plant growth information monitoring system and method based on Internet of Things | |
| CN108157343A (en) | A kind of accurate variable farm chemical applying system in short anvil apple orchard based on unmanned aerial vehicle remote sensing | |
| CN208095182U (en) | A kind of agricultural, which is examined, beats integrated machine system | |
| CN109613022A (en) | A kind of method, apparatus and system of low latitude high-spectrum remote-sensing detection Citrus Huanglongbing pathogen | |
| CN109557003B (en) | Pesticide deposition amount detection method and device and data acquisition combination device | |
| Črtomir et al. | Application of Neural Networks and Image Visualization for Early Forecast of Apple Yield. | |
| Hong et al. | Detection system of smart sprayers: Status, challenges, and perspectives | |
| TW202107972A (en) | Unmanned aerial system and liquid spraying method with artificial intelligence image processing technology capable of accurately controlling irrigation and spraying timing for water/liquid fertilizer/medicine solution required for crop growth | |
| CN109446987A (en) | Method based on PCA and PNN algorithm detection rice pest grade | |
| CN113724210B (en) | Crop growth condition identification method and system | |
| CN109977901A (en) | A kind of corn weeds in field recognition methods based on unmanned aerial vehicle remote sensing | |
| Rizk et al. | Robotized early plant health monitoring system | |
| Zhang et al. | Opportunities of UAVs in orchard management | |
| CN108037081A (en) | A kind of vinifera maturity monitoring method and system | |
| CN106896077B (en) | The detection method of transgenic corns glyphosate tolerant phenotype based on imaging-PAM | |
| CN109960972A (en) | A kind of farm-forestry crop recognition methods based on middle high-resolution timing remotely-sensed data | |
| CN114954948A (en) | Intelligent pesticide applying system for plant protection unmanned aerial vehicle and control method | |
| CN112640870B (en) | Plant protection unmanned aerial vehicle-based pest control system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of utility model: A kind of agricultural inspection and fighting machine system Effective date of registration: 20201210 Granted publication date: 20181116 Pledgee: Guangzhou Baidu venture capital artificial intelligence equity investment partnership (L.P.) Pledgor: BEIJING MAIFEI TECHNOLOGY Co.,Ltd. Registration number: Y2020990001432 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |