CN109166110A - A kind of intelligence vegetables nursery breeding automatic load system and packing method - Google Patents

A kind of intelligence vegetables nursery breeding automatic load system and packing method Download PDF

Info

Publication number
CN109166110A
CN109166110A CN201810927040.9A CN201810927040A CN109166110A CN 109166110 A CN109166110 A CN 109166110A CN 201810927040 A CN201810927040 A CN 201810927040A CN 109166110 A CN109166110 A CN 109166110A
Authority
CN
China
Prior art keywords
data
particle
seed
vegetables nursery
hole tray
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.)
Pending
Application number
CN201810927040.9A
Other languages
Chinese (zh)
Inventor
张其安
严从生
俞飞飞
方凌
王艳
王明霞
江海坤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Institute of Soil and Fertilizer of Anhui Academy of Agricultural Sciences
Original Assignee
Institute of Soil and Fertilizer of Anhui Academy of Agricultural Sciences
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Institute of Soil and Fertilizer of Anhui Academy of Agricultural Sciences filed Critical Institute of Soil and Fertilizer of Anhui Academy of Agricultural Sciences
Priority to CN201810927040.9A priority Critical patent/CN109166110A/en
Publication of CN109166110A publication Critical patent/CN109166110A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01C—PLANTING; SOWING; FERTILISING
    • A01C7/00—Sowing
    • A01C7/002—Dibble seeders
    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01C—PLANTING; SOWING; FERTILISING
    • A01C7/00—Sowing
    • A01C7/20—Parts of seeders for conducting and depositing seed
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/12—Edge-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/149—Segmentation; Edge detection involving deformable models, e.g. active contour models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10024—Color image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20024—Filtering details

Landscapes

  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Soil Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Environmental Sciences (AREA)
  • Software Systems (AREA)
  • Quality & Reliability (AREA)
  • Image Processing (AREA)

Abstract

The invention belongs to vegetables nursery breeding automatic device technical fields, disclose a kind of intelligent vegetables nursery breeding automatic load system and packing method, it is built-in in a computer to control program, according to the density of required sowing, adjusts seed and fall into the time interval in hole tray；And control transverse bar and longitudinal rod and slided in electric track, it is orderly and registration in seed filling hole tray；And the camera by installing in hole tray acquires the missing image information of seed in hole tray, and feeds back to computer, closure of the control gear button to circular valve.Apparatus of the present invention pass through under the action of controlling program, transverse bar and longitudinal rod are slided in electric track, orderly and accurately in seed filling hole tray, the course of work of the automatical and efficient rate of the process loaded instead of artificial vegetables nursery breeding, the survival rate of seed is also improved simultaneously, large-scale promotion is suitble to use.

Description

A kind of intelligence vegetables nursery breeding automatic load system and packing method
Technical field
The invention belongs to vegetables nursery breeding automatic device technical field more particularly to a kind of intelligent vegetables nursery breedings certainly Dynamic loading system and packing method.
Background technique
Currently, the prior art commonly used in the trade is such that
Currently, whether flowers or vegetables, hole plate seedling growth are the most fundamental changes of modern horticultural, it is quick and big Batch production provides guarantees, and hole tray has become an important utensil in industrialization seeling industry technique, and hole tray out Generation is also required to the filling of seed, and is at present largely manual operation to the filling of hole tray, and manual operation is inaccurate, efficiency Also not high enough, therefore, it is necessary to a kind of helps of intelligent vegetables nursery breeding automatic load system, make filling process more efficiently quasi- Really.
In conclusion problem of the existing technology is:
Manually by the hole tray in seed implantation high density hole, there are low efficiency, artificial fixtures greatly, to cause to seed destructiveness The low problem of seed viability, while it also being be easy to cause pollution, cause seed germ contamination occur, therefore vegetables nursery breeding is certainly Dynamic loading system is indispensable, but nowadays system is not smart enough, is not able to satisfy the demand of people, the convenience brought It is not obvious enough, so needing the convenience of intelligence not obvious enough, so needing a kind of intelligent vegetables nursery breeding automatic load system It is convenient to give people.
It is different and different with distance during time interval prediction, to precision of prediction there are large effect, So prediction model can be further perfect.
Conventional color image contour extraction method is when extracting object boundary vulnerable to the interference of initial profile point and convergence rate Slowly, big so as to cause the color image profile noise of extraction, influence the effect of image segmentation.
Summary of the invention
In view of the problems of the existing technology, the present invention provides a kind of intelligent vegetables nursery breeding automatic load systems.
The invention is realized in this way a kind of intelligence vegetables nursery breeding automatic Loading method, comprising:
It is built-in in a computer to control program, according to the density of required sowing, adjusts seed and fall into the time in hole tray Interval；And control transverse bar and longitudinal rod and slided in electric track, it is orderly and registration in seed filling hole tray；
Computer adjustment seed is fallen into the time interval in hole tray, and control program includes: selection initial data, open Time interval data collection acquire data, shortage of data inspection is carried out to the data that are collected into and is supplemented complete；It is complete to supplementing Data calculate its rank correlation coefficient according to copula function, suitable input data is selected according to the size of rank correlation coefficient Time；
Data are finally entered according to what the size relation of grey Absolute data relating extent determined model to selected time interval； Modeling analysis is carried out to the data that finally enter of selection, the shape of Nonlinear state space model is established according to support vector regression State equation and measurement equation, and estimation prediction is carried out by state of the Unscented kalman filtering to model；Selection according to setting Standard optimizes and updates to the scale parameter in Unscented kalman filtering, obtains prediction result；
And the camera by installing in hole tray acquires the missing image information of seed in hole tray, and feeds back to computer, Gear button is controlled to the closure of circular valve；Camera acquires in hole tray in the missing image information of seed, using based on grain The color image contour extraction method of son filtering obtains the missing image information of seed in hole tray；With Sobel operator or it is based on face The predicted value of colour space clustering method calculating objective contour:
Image object profile is considered as to the collection { dl being made of N number of unit line segmenti}1,2 ..., N, for i=1,2 ..., N；In C0 In find and dliCorresponding position, according to C0The tangent line of middle corresponding position is as dliSampled reference value, generate primary Collection；Constantly assemble to known optimum solution direction according to state transition model guidance particle, avoids standard particle filtering The method of middle degeneration realizes particle state transfer, and calculates the corresponding profile point set of each particle；According to the observation mould of foundation Type calculates particle weights；The parameter dl obtained with the weighted average calculation current iteration of particle collectionj (i)=(kj (i), bj (i))；
IfWherein, ε=0.5 is taken, then is obtained with the weighted average of particle collection As dliThe estimation of parameter.
Further, initial data is selected, data is acquired in disclosed time interval data collection, the data being collected into is carried out Shortage of data is examined and is supplemented complete；Specifically: the data for the same period for selecting each time interval annual；Total selection 22 groups, every group of 192 data, then to all collected data carry out missing check for missing data more than 30 when Between directly delete, it is remaining to be filled using averaging method, that is, assuming that aiFor missing data, then the data of the position are filled into
If aiWhen the data of front and back 12 also have missing, recursion of drawing back forward, until obtaining the data of front and back 12；If aiFor First data, then at this timeSimilarly, aiWhen for the last one data,If aiBefore in data set When 12 and rear 12 data, still selection and the identical fill method of head and the tail.
Further, the scale parameter in Unscented kalman filtering is optimized and is updated, obtain prediction result, comprising:
The feasible set λ ∈ [0,12] of a specified scale parameter λ, update method is as follows:
(2) initial λ is selected；
Take it to beWherein λmax=12, λmin=0；
(2) when updating, every time in original λjOn the basis of be added a random value ej, the value meet normal state variation, the phase Hope to be 0, variance very little；
κj+=κj+ej ej~N (0, Pe j), enable j=0,1,
(3) by above-mentioned λj+And λjIt substitutes into Unscented kalman filtering respectively and carries out prediction calculating；
(4) the prediction error for calculating the two takes the small person of error to enter and updates in next step；
If
Then take λj+1=λj+
Otherwise λj+1=λj
(5) (2)-(4) step is recycled, until predicting that error reaches the threshold value of setting or update times reach established standards When, it updates and stops, obtaining optimal scale parameter λ.
Further, state transition model avoids the method degenerated in standard particle filtering from including:
It is located at and calculates i-th of unit line segment l of image outlineiParameterWhen, particle collection when iteration to s walks For
Wherein Θs (i)For straight line parameter set, Ws (i)For particle weights, Es (i)It is calculated for Snake energy function model Profile validity estimate, be the foundation of granular Weights Computing；Based on conditions above, local optimum particle are as follows:
Global optimum's particle are as follows:
Then PSO state transition model is as follows:
Wherein rk, rk1, rk2, rb, rb1, rb2Equal Normal Distribution, and
Further, the method for establishing observation model includes:
If the estimates of parameters of estimated the 1st to the (i-1)-th obtained unit line segment are as follows:
Corresponding point sequence of an outline are as follows:
Calculating i-th of unit line segment l of image object profileiParameter when, iteration to s walk when particle collection are as follows:
Thus particle collection obtains image object profile point and is
Local Snake energy value is calculated as follows:
Global Snake energy value is calculated as follows:
Or only take { Ψj}J=1,2 ..., i-1In withSeveral nearest point sets calculate overall situation Snake energy value jointly；
For RGB color image or HSI spatial color image, each component has corresponding Snake energy value, i.e.,
It is right respectivelyWithIt is normalized, Localized particle weight is calculated as follows；
Particle overall situation weight is calculated as follows:
Taking particle weights is the arithmetic average of part and global value, i.e.,
Finally, particle weights are normalized, to realize the Minimum Mean Squared Error estimation of parameter.
Another object of the present invention is to provide a kind of meters for realizing the intelligent vegetables nursery breeding automatic Loading method Calculation machine program.
Another object of the present invention is to provide a kind of meters for realizing the intelligent vegetables nursery breeding automatic Loading method Calculation machine.
Another object of the present invention is to provide a kind of computer readable storage mediums, including instruction, when it is in computer When upper operation, so that computer executes the intelligent vegetables nursery breeding automatic Loading method.
Another object of the present invention is to provide a kind of intelligent vegetables nursery breeding automatic load systems to be provided with seed storage Put box, conical hopper, circular valve, seed export pipe, firm banking, transverse bar, longitudinal rod, electric track, gear button；
The firm banking is connect by nut with electric track, and longitudinal rod, the longitudinal direction are embedded on the electric track Electric track is bonded on bar outer wall, the transverse bar is on the electric track on longitudinal rod, on the outer wall of the transverse bar It is bonded with electric track；On the embedding electric track on the transverse bar of seed storage box, the bottom end of the seed storage box is viscous It is connected to circular valve；It is embedding on the outer wall of the circular valve that there are three gear buttons；The seed export pipe is bonded in round valve Men Shang.
Further, there are three loophole of different sizes inside the circular valve, three gear buttons control three respectively The closure of loophole；
The regulation of the electric track external power supply and controlled processing procedure sequence.
Advantages of the present invention and good effect are as follows:
Apparatus of the present invention by the way that under the action of controlling program, transverse bar and longitudinal rod are slided in electric track, orderly and Accurately in seed filling hole tray, loading instead of artificial vegetables nursery breeding for the automatical and efficient rate of the process is worked Journey, while the survival rate of seed is also improved, it is suitble to large-scale promotion to use.
The present invention can effectively improve the time interval precision of prediction that computer adjustment seed is fallen into hole tray and compare General traditional prediction method.
The present invention judges the Similarity measures of different time segment data for the processing of shortage of data and using the degree of association, The similitude between same time segment data is compared, precision of prediction is improved.
What the present invention established during prediction is nonlinear model, is preferably reduced due to density randomness, fluctuation The problem of precision of prediction deficiency caused by property.
It goes to predict time interval by a nonlinear prediction technique, be carried out using Unscented kalman filtering pre- It surveys, it is ensured that its real-time predicted, and it is more in line with the practical rule of time interval, it is as a result more accurate.
It interferes and restrains vulnerable to initial profile point when extracting object boundary for conventional color image contour extraction method Speed is slow, big so as to cause the color image profile noise of extraction, the effect of image segmentation has been influenced, in consideration of it, of the invention Propose the color image contours extract algorithm based on particle filter.Firstly, providing the prediction of image outline and establishing two herein Dimension space, to make full use of image information；Then, the state transition model based on PSO optimization method is constructed, which promotes Particle is close to known optimum state, improves the distribution of particle, accelerates convergence rate；It finally establishes and is based on Snake The observation model of energy function, the observation model are capable of the effect of preferable quantitative description contour extraction of objects.
By simulation result it is found that its background is all more complicated in the color image of experiment, and come from experimental result It sees, predicts that the profile of color image is very important, if the profile predicted is stable and accurate, to subsequent profile It extracts and image segmentation is very helpful.Present invention employs the measurement models based on Snake energy function, although Predict to interfere when profile more, but image object profile still can accurately be predicted and be extracted to its algorithm, this is to subsequent figure As very big effect is played in segmentation.
Compared with traditional color image contour extraction method, effectively alleviates and interfered and restrained speed by initial profile point The problems such as slow is spent, the image outline especially under Low SNR extracts, and the image object profile extracted also extremely makes us It is satisfied, it is better than general contours extract algorithm.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of intelligent vegetables nursery breeding automatic load system provided in an embodiment of the present invention；
Fig. 2 is the structural schematic diagram of circular valve provided in an embodiment of the present invention；
In figure: 1, seed storage box；2, conical hopper；3, circular valve；4, seed export pipe；5, firm banking；6, horizontal To bar；7, longitudinal rod；8, electric track；9, gear button；10, computer；11, camera.
Specific embodiment
In order to further understand the content, features and effects of the present invention, the following examples are hereby given, and cooperate attached drawing Detailed description are as follows.
Structure of the invention is explained in detail with reference to the accompanying drawing.
As depicted in figs. 1 and 2, intelligent vegetables nursery breeding automatic load system provided in an embodiment of the present invention, is provided with Seed storage box 1, conical hopper 2, circular valve 3, seed export pipe 4, firm banking 5, transverse bar 6, longitudinal rod 7, electronic rail Road 8, gear button 9.The firm banking 5 is connect by nut with electric track 8, and longitudinal rod is embedded on the electric track 8 7, electric track 8 is bonded on 7 outer wall of longitudinal rod, the transverse bar 6 is described on the electric track 8 on longitudinal rod 7 Electric track 8 is bonded on the outer wall of transverse bar 6, the seed storage box 1 is described on the electric track 8 on transverse bar 6 The bottom end of seed storage box 1 is bonded with circular valve 3, and embedding on the outer wall of the circular valve 3 there are three gear buttons 9, described Seed export pipe 4 is bonded on circular valve 3, and further, 3 inside of the circular valve is there are three loophole of different sizes, and three A gear button 9 controls the closure of three loopholes respectively, further, 8 external power supply of electric track and controlled processing procedure sequence Regulation.
In using the present invention, breeding hole tray is placed on firm banking 5 first, then it is required enter kind seed It is placed in seed storage box 1, according to the density of required sowing, selects suitable gear button 9, such seed falls into hole tray In time interval will be different, density is also just different, and controls process control longitudinal rod 7, transverse bar 6 and seed storage box 1 sliding on electric track 8 needs in time to supplement it, avoids cave when the seed deficiency in seed storage box 1 It is not loaded in disk, when hole tray is filled, the hole tray more renewed is placed on firm banking 5.
Below with reference to concrete analysis, the invention will be further described.
A kind of intelligent vegetables nursery breeding automatic Loading method provided in an embodiment of the present invention, comprising:
The built-in control program in computer 10, according to the density of required sowing, adjust seed fall into hole tray when Between be spaced；And control transverse bar and longitudinal rod and slided in electric track, it is orderly and registration in seed filling hole tray；
Computer adjustment seed is fallen into the time interval in hole tray, and control program includes: selection initial data, open Time interval data collection acquire data, shortage of data inspection is carried out to the data that are collected into and is supplemented complete；It is complete to supplementing Data calculate its rank correlation coefficient according to copula function, suitable input data is selected according to the size of rank correlation coefficient Time；
Data are finally entered according to what the size relation of grey Absolute data relating extent determined model to selected time interval； Modeling analysis is carried out to the data that finally enter of selection, the shape of Nonlinear state space model is established according to support vector regression State equation and measurement equation, and estimation prediction is carried out by state of the Unscented kalman filtering to model；Selection according to setting Standard optimizes and updates to the scale parameter in Unscented kalman filtering, obtains prediction result；
And the camera 11 by installing in hole tray acquires the missing image information of seed in hole tray, and feeds back to calculating Machine, closure of the control gear button to circular valve；Camera acquires in hole tray in the missing image information of seed, using being based on The color image contour extraction method of particle filter obtains the missing image information of seed in hole tray；With Sobel operator or it is based on The predicted value of Color Space Clustering method calculating objective contour:
Image object profile is considered as to the collection { dl being made of N number of unit line segmenti}1,2 ..., N, for i=1,2 ..., N；In C0 In find and dliCorresponding position, according to C0The tangent line of middle corresponding position is as dliSampled reference value, generate primary Collection；Constantly assemble to known optimum solution direction according to state transition model guidance particle, avoids standard particle filtering The method of middle degeneration realizes particle state transfer, and calculates the corresponding profile point set of each particle；According to the observation mould of foundation Type calculates particle weights；The parameter dl obtained with the weighted average calculation current iteration of particle collectionj (i)=(kj (i), bj (i))；
IfWherein, ε=0.5 is taken, then is obtained with the weighted average of particle collection As dliThe estimation of parameter.
Initial data is selected, data is acquired in disclosed time interval data collection, data is carried out to the data being collected into and are lacked Lapsing, it is complete to test and supplement；Specifically: the data for the same period for selecting each time interval annual；22 groups of selection altogether, Then every group of 192 data carry out missing to all collected data and check that the time for being more than 30 for missing data is straight Deletion is connect, it is remaining to be filled using averaging method, that is, assuming that aiFor missing data, then the data of the position are filled into
If aiWhen the data of front and back 12 also have missing, recursion of drawing back forward, until obtaining the data of front and back 12；If aiFor First data, then at this timeSimilarly, aiWhen for the last one data,If aiBefore in data set When 12 and rear 12 data, still selection and the identical fill method of head and the tail.
Scale parameter in Unscented kalman filtering is optimized and is updated, prediction result is obtained, comprising:
The feasible set λ ∈ [0,12] of a specified scale parameter λ, update method is as follows:
(3) initial λ is selected；
Take it to beWherein λmax=12, λmin=0；
(2) when updating, every time in original λjOn the basis of be added a random value ej, the value meet normal state variation, the phase Hope to be 0, variance very little；
κj+=κj+ej ej~N (0, Pe j), enable j=0,1,
(3) by above-mentioned λj+And λjIt substitutes into Unscented kalman filtering respectively and carries out prediction calculating；
(4) the prediction error for calculating the two takes the small person of error to enter and updates in next step；
If
Then take λj+1=λj+
Otherwise λj+1=λj
(5) (2)-(4) step is recycled, until predicting that error reaches the threshold value of setting or update times reach established standards When, it updates and stops, obtaining optimal scale parameter λ.
State transition model avoids the method degenerated in standard particle filtering from including:
It is located at and calculates i-th of unit line segment l of image outlineiParameterWhen, particle collection when iteration to s walks For
Wherein Θs (i)For straight line parameter set, Ws (i)For particle weights, Es (i)It is calculated for Snake energy function model Profile validity estimate, be the foundation of granular Weights Computing；Based on conditions above, local optimum particle are as follows:
Global optimum's particle are as follows:
Then PSO state transition model is as follows:
Wherein rk, rk1, rk2, rb, rb1, rb2Equal Normal Distribution, and
The method for establishing observation model includes:
If the estimates of parameters of estimated the 1st to the (i-1)-th obtained unit line segment are as follows:
Corresponding point sequence of an outline are as follows:
Calculating i-th of unit line segment l of image object profileiParameter when, iteration to s walk when particle collection are as follows:
Thus particle collection obtains image object profile point and is
Local Snake energy value is calculated as follows:
Global Snake energy value is calculated as follows:
Or only take { Ψj}J=1,2 ..., i-1In withSeveral nearest point sets calculate overall situation Snake energy value jointly；
For RGB color image or HSI spatial color image, each component has corresponding Snake energy value, i.e.,
It is right respectivelyWithIt is normalized, Localized particle weight is calculated as follows；
Particle overall situation weight is calculated as follows:
Taking particle weights is the arithmetic average of part and global value, i.e.,
Finally, particle weights are normalized, to realize the Minimum Mean Squared Error estimation of parameter.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real It is existing.When using entirely or partly realizing in the form of a computer program product, the computer program product include one or Multiple computer instructions.When loading on computers or executing the computer program instructions, entirely or partly generate according to Process described in the embodiment of the present invention or function.The computer can be general purpose computer, special purpose computer, computer network Network or other programmable devices.The computer instruction may be stored in a computer readable storage medium, or from one Computer readable storage medium is transmitted to another computer readable storage medium, for example, the computer instruction can be from one A web-site, computer, server or data center pass through wired (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL) Or wireless (such as infrared, wireless, microwave etc.) mode is carried out to another web-site, computer, server or data center Transmission).The computer-readable storage medium can be any usable medium or include one that computer can access The data storage devices such as a or multiple usable mediums integrated server, data center.The usable medium can be magnetic Jie Matter, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk Solid State Disk (SSD)) etc..
The above is only the preferred embodiments of the present invention, and is not intended to limit the present invention in any form, Any simple modification made to the above embodiment according to the technical essence of the invention, equivalent variations and modification, belong to In the range of technical solution of the present invention.

Claims (10)

1. a kind of intelligence vegetables nursery breeding automatic Loading method, which is characterized in that the intelligence vegetables nursery breeding fills automatically Embankment method includes:
It is built-in in a computer to control program, according to the density of required sowing, adjusts seed and fall into the time interval in hole tray； And control transverse bar and longitudinal rod and slided in electric track, it is orderly and registration in seed filling hole tray；
Computer adjustment seed is fallen into the time interval in hole tray, and control program includes: selection initial data, when disclosed Between interval data collection acquire data, shortage of data inspection is carried out to the data that are collected into and is supplemented complete；To the complete number of supplement Calculate its rank correlation coefficient according to according to copula function, according to the size of rank correlation coefficient select suitable input data when Between；
Data are finally entered according to what the size relation of grey Absolute data relating extent determined model to selected time interval；To choosing The data that finally enter taken carry out modeling analysis, and the state side of Nonlinear state space model is established according to support vector regression Journey and measurement equation, and estimation prediction is carried out by state of the Unscented kalman filtering to model；The selection criteria of foundation setting, Scale parameter in Unscented kalman filtering is optimized and updated, prediction result is obtained；
And the camera by installing in hole tray acquires the missing image information of seed in hole tray, and feeds back to computer, controls Closure of the gear button to circular valve；Camera acquires in hole tray in the missing image information of seed, filters using based on particle The color image contour extraction method of wave obtains the missing image information of seed in hole tray；With Sobel operator or based on color sky Between clustering method calculate objective contour predicted value:
Image object profile is considered as to the collection { dl being made of N number of unit line segmenti}1,2 ..., N, for i=1,2 ..., N；In C0In look for It arrives and dliCorresponding position, according to C0The tangent line of middle corresponding position is as dliSampled reference value, generate primary collection；It presses Constantly assemble to known optimum solution direction according to state transition model guidance particle, avoids degenerating in standard particle filtering Method realize particle state transfer, and calculate the corresponding profile point set of each particle；It is calculated according to the observation model of foundation Particle weights；The parameter dl obtained with the weighted average calculation current iteration of particle collectionj (i)=(kj (i), bj (i))；IfWherein, ε=0.5 is taken, then is obtained using the weighted average of particle collection as dliGinseng Several estimations.
2. intelligence vegetables nursery breeding automatic Loading method as described in claim 1, which is characterized in that
Initial data is selected, data is acquired in disclosed time interval data collection, shortage of data inspection is carried out to the data being collected into It tests and supplements complete；Specifically: the data for the same period for selecting each time interval annual；22 groups, every group of selection altogether Then 192 data carry out missing to all collected data and check that the time for missing data more than 30 directly deletes It removes, it is remaining to be filled using averaging method, that is, assuming that aiFor missing data, then the data of the position are filled into
If aiWhen the data of front and back 12 also have missing, recursion of drawing back forward, until obtaining the data of front and back 12；If aiIt is first A data, then at this timeSimilarly, aiWhen for the last one data,If aiFor 12 He preceding in data set Afterwards when 12 data, still selection and the identical fill method of head and the tail.
3. intelligence vegetables nursery breeding automatic Loading method as described in claim 1, which is characterized in that Unscented kalman filtering In scale parameter optimize and update, obtain prediction result, comprising:
The feasible set λ ∈ [0,12] of a specified scale parameter λ, update method is as follows:
(1) initial λ is selected；
Take it to beWherein λmax=12, λmin=0；
(2) when updating, every time in original λjOn the basis of be added a random value ej, the value meet normal state variation, be desired for 0, variance very little；
κj+=κj+ej ej~N (0, Pe j), enable j=0,1,
(3) by above-mentioned λj+And λjIt substitutes into Unscented kalman filtering respectively and carries out prediction calculating；
(4) the prediction error for calculating the two takes the small person of error to enter and updates in next step；
If
Then take λj+1=λj+
Otherwise λj+1=λj
(5) (2)-(4) step is recycled, when prediction error reaches the threshold value of setting or update times reach established standards, more It is new to stop, obtaining optimal scale parameter λ.
4. intelligence vegetables nursery breeding automatic Loading method as described in claim 1, which is characterized in that state transition model avoids The method degenerated in standard particle filtering includes:
It is located at and calculates i-th of unit line segment l of image outlineiParameterWhen, particle collection is when iteration to s walks
Wherein Θs (i)For straight line parameter set, Ws (i)For particle weights, Es (i)The wheel being calculated for Snake energy function model Wide validity is estimated, and is the foundation of granular Weights Computing；Based on conditions above, local optimum particle are as follows:
Global optimum's particle are as follows:
Then PSO state transition model is as follows:
Wherein rk, rk1, rk2, rb, rb1, rb2Equal Normal Distribution, and
5. intelligence vegetables nursery breeding automatic Loading method as described in claim 1, which is characterized in that establish the side of observation model Method includes:
If the estimates of parameters of estimated the 1st to the (i-1)-th obtained unit line segment are as follows:
Corresponding point sequence of an outline are as follows:
Calculating i-th of unit line segment l of image object profileiParameter when, iteration to s walk when particle collection are as follows:
Thus particle collection obtains image object profile point and is
Local Snake energy value is calculated as follows:
Global Snake energy value is calculated as follows:
Or only take { Ψj}J=1,2 ..., i-1In withSeveral nearest point sets calculate overall situation Snake energy value jointly；
For RGB color image or HSI spatial color image, each component has corresponding Snake energy value, i.e.,
It is right respectivelyWithIt is normalized, presses Formula calculates localized particle weight；
Particle overall situation weight is calculated as follows:
Taking particle weights is the arithmetic average of part and global value, i.e.,
Finally, particle weights are normalized, to realize the Minimum Mean Squared Error estimation of parameter.
6. a kind of computer journey for realizing intelligence vegetables nursery breeding automatic Loading method described in Claims 1 to 5 any one Sequence.
7. a kind of computer for realizing intelligence vegetables nursery breeding automatic Loading method described in Claims 1 to 5 any one.
8. a kind of computer readable storage medium, including instruction, when run on a computer, so that computer is executed as weighed Benefit requires intelligent vegetables nursery breeding automatic Loading method described in 1-5 any one.
9. a kind of intelligent vegetables nursery breeding for realizing intelligence vegetables nursery breeding automatic Loading method described in claim 1 is automatic Loading system, which is characterized in that the intelligence vegetables nursery breeding automatic load system is provided with seed storage box, taper leakage Bucket, circular valve, seed export pipe, firm banking, transverse bar, longitudinal rod, electric track, gear button；
The firm banking is connect by nut with electric track, embedded with longitudinal rod on the electric track, outside the longitudinal rod Electric track is bonded on wall；The transverse bar is bonded on the outer wall of the transverse bar on the electric track on longitudinal rod There is electric track；On the embedding electric track on the transverse bar of seed storage box, the bottom end of the seed storage box is bonded with Circular valve；It is embedding on the outer wall of the circular valve that there are three gear buttons；The seed export pipe is bonded on circular valve.
10. intelligence vegetables nursery breeding automatic load system as claimed in claim 9, which is characterized in that in the circular valve There are three loophole of different sizes, three gear buttons to control the closure of three loopholes respectively in portion；
The electric track external power supply.
CN201810927040.9A 2018-08-15 2018-08-15 A kind of intelligence vegetables nursery breeding automatic load system and packing method Pending CN109166110A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810927040.9A CN109166110A (en) 2018-08-15 2018-08-15 A kind of intelligence vegetables nursery breeding automatic load system and packing method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810927040.9A CN109166110A (en) 2018-08-15 2018-08-15 A kind of intelligence vegetables nursery breeding automatic load system and packing method

Publications (1)

Publication Number Publication Date
CN109166110A true CN109166110A (en) 2019-01-08

Family

ID=64895751

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810927040.9A Pending CN109166110A (en) 2018-08-15 2018-08-15 A kind of intelligence vegetables nursery breeding automatic load system and packing method

Country Status (1)

Country Link
CN (1) CN109166110A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111013120A (en) * 2019-12-23 2020-04-17 重庆工商大学 Sliding seesaw control system and method for snowfield entertainment

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101152294B1 (en) * 2009-11-26 2012-06-08 오창준 Apparatus for transplating plug seedling
CN103918385A (en) * 2014-04-24 2014-07-16 淮海工学院 Intelligent anti-blockage non-miss-seeding pneumatic type plug seedling precision seed-metering device
CN105320809A (en) * 2015-09-30 2016-02-10 西安电子科技大学 Wind speed prediction method for wind farm spatial correlation
CN106664902A (en) * 2016-11-23 2017-05-17 宁波市镇海区精量农业科技开发有限公司 Precision vegetable sower and rail thereof
CN107211632A (en) * 2016-03-21 2017-09-29 新昌县澄潭镇博纳机械厂 The adjustable component of sowing of tray seeder

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101152294B1 (en) * 2009-11-26 2012-06-08 오창준 Apparatus for transplating plug seedling
CN103918385A (en) * 2014-04-24 2014-07-16 淮海工学院 Intelligent anti-blockage non-miss-seeding pneumatic type plug seedling precision seed-metering device
CN105320809A (en) * 2015-09-30 2016-02-10 西安电子科技大学 Wind speed prediction method for wind farm spatial correlation
CN107211632A (en) * 2016-03-21 2017-09-29 新昌县澄潭镇博纳机械厂 The adjustable component of sowing of tray seeder
CN106664902A (en) * 2016-11-23 2017-05-17 宁波市镇海区精量农业科技开发有限公司 Precision vegetable sower and rail thereof

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
伊力哈木•亚尔买买提: "基于粒子滤波的彩色图像轮廓提取算法研究", 《计算机仿真》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111013120A (en) * 2019-12-23 2020-04-17 重庆工商大学 Sliding seesaw control system and method for snowfield entertainment

Similar Documents

Publication Publication Date Title
CN110751094B (en) Crop yield estimation method based on GEE comprehensive remote sensing image and deep learning method
D'Ambrosio et al. A cellular automata model for soil erosion by water
CN104914720B (en) Electronic nose intelligence baking control system and control method with autolearn feature
CN111145245B (en) Short-time adjacent rainfall forecasting method, system and computer-readable storage medium
Biondi et al. Performance assessment of a Bayesian Forecasting System (BFS) for real-time flood forecasting
CN109871995A (en) The quantum optimization parameter adjustment method of distributed deep learning under Spark frame
CN104091216A (en) Traffic information predication method based on fruit fly optimization least-squares support vector machine
CN104679834A (en) Time series data cleaning method and system
CN110046570B (en) Method and device for dynamically supervising grain stock of granary
WO2014054042A1 (en) Device and method for detecting plantation rows
CN108716904B (en) Dam body deflection obtaining method based on measuring point measuring value of finite inclinometer
CN109255476B (en) Variable parameter nonlinear river flow prediction method
CN109166110A (en) A kind of intelligence vegetables nursery breeding automatic load system and packing method
CN114970302A (en) Regional underground water condition prediction method based on underground water monitoring system
Wang et al. Industry 3.5 to empower smart production for poultry farming and an empirical study for broiler live weight prediction
CN105405152B (en) Adaptive scale method for tracking target based on structuring support vector machines
CN110826526A (en) Method for cloud detection radar to identify clouds
CN117953034B (en) Machine vision-based laying hen weight estimation system and method
CN110414084A (en) A kind of semivariable function model solution method based on tri-level programming
CN117236668B (en) Regional water resource allocation analysis decision method and system based on supply and consumption
He et al. A calculation method of phenotypic traits of soybean pods based on image processing technology
CN116415508B (en) Method and system for generating two-dimensional sediment model of estuary
CN117243145A (en) Accurate feeding method and device for meat poultry based on feed intake
CN108920737A (en) A kind of particle filter assimilation method of hydrodynamic model, device and calculate equipment
CN111027612A (en) Energy metering data feature reduction method and device based on weighted entropy FCM

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20190108

RJ01 Rejection of invention patent application after publication