CN110472557A - A kind of method and device of tomato growth monitoring - Google Patents
A kind of method and device of tomato growth monitoring Download PDFInfo
- Publication number
- CN110472557A CN110472557A CN201910742244.XA CN201910742244A CN110472557A CN 110472557 A CN110472557 A CN 110472557A CN 201910742244 A CN201910742244 A CN 201910742244A CN 110472557 A CN110472557 A CN 110472557A
- Authority
- CN
- China
- Prior art keywords
- tomato
- growth
- model
- analysis
- monitoring data
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Mining
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
Abstract
The present invention relates to the method and devices that agriculture Internet of Things applied technical field more particularly to a kind of tomato growth monitor, including pre-establish tomato growth analysis model；The video and image information for obtaining the environmental monitoring data of tomato and currently growing；Judge then to send early warning to user terminal when environmental monitoring data is more than preset threshold；Judged then to send early warning to user terminal when the variation of tomato plant within a preset time is more than default change threshold or the long damaged by vermin of tomato plant according to video and image information；Image information is imported in tomato growth analysis model, the current growth indexes of tomato are exported by analytical calculation.The utility model has the advantages that both monitoring the upgrowth situation of tomato plant by environmental monitoring and video image monitoring to play forewarning function, while the growth indexes of tomato are analyzed further through algorithm model, to realize that real Internet of Things is intelligently planted.
Description
Technical field
A kind of method and dress monitored the present invention relates to agriculture Internet of Things applied technical field more particularly to tomato growth
It sets.
Background technique
With the development of modernization, technology of Internet of things is graduallyd mature, automation, intelligence, the standardization of agricultural production
Also at development trend, the monitoring and control technology of agricultural environment is also being constantly progressive, but for the agriculture under environment of internet of things
Industry management is monitored and controlled, still without comparatively perfect and mature technical solution, currently, agricultural monitoring system on the market
Have and individually does disease pest monitoring, the monitoring of plantation situation, early warning system and by what algorithm model was analyzed one or two kinds of be
System, there are no the integrated functional systems of institute, cannot achieve real Internet of Things and intelligently plant.
Summary of the invention
It is an object of the invention to propose the method and device of a kind of kind of tomato growth monitoring, to solve existing tomato
Plantation cannot achieve the problem of real Internet of Things is intelligently planted.
To achieve this purpose, the present invention adopts the following technical scheme:
A kind of method of tomato growth monitoring, comprising:
Pre-establish tomato growth analysis model；
The video and image information for obtaining the environmental monitoring data of tomato and currently growing；
Judge then to send early warning to user terminal when the environmental monitoring data is more than preset threshold；
According to the video and image information judgement ought within a preset time tomato plant variation be more than default change threshold or kind
When the long damaged by vermin of eggplant plant, then early warning is sent to user terminal；
Described image information is imported in the tomato growth analysis model, the current growth of tomato is exported by analytical calculation and is referred to
Mark.
The present invention also provides a kind of devices of tomato growth monitoring, comprising:
Analysis model built in advance unit, for pre-establishing tomato growth analysis model；
Data image acquiring unit, for obtaining the environmental monitoring data and the currently video that grows and image information of tomato；
Data early warning unit, for judge when the environmental monitoring data is more than preset threshold, then send early warning to
Family terminal；
Image prewarning unit, for ought tomato plant variation be more than within a preset time according to the video and image information judgement
When default change threshold or the long damaged by vermin of tomato plant, then early warning is sent to user terminal；
Model analysis unit is led for importing described image information in the tomato growth analysis model by analytical calculation
The current growth indexes of tomato out.
The present invention is by pre-establishing tomato growth analysis model；The environmental monitoring data for obtaining tomato and currently growth
Video and image information；Judge then to send early warning to user terminal when environmental monitoring data is more than preset threshold；Root
According to video and image information judgement ought within a preset time tomato plant variation be more than default change threshold or tomato plant with
When insect pest, then early warning is sent to user terminal；Image information is imported in tomato growth analysis model, analytical calculation is passed through
Export the current growth indexes of tomato.
The utility model has the advantages that both having been monitored by environmental monitoring and video image pre- to play to monitor the upgrowth situation of tomato plant
Police acts on, while the growth indexes of tomato are analyzed further through algorithm model, to realize that real Internet of Things is intelligently planted.
Detailed description of the invention
Fig. 1 is the method flow diagram for the tomato growth monitoring that the specific embodiment of the invention provides.
Fig. 2 is the apparatus structure schematic diagram for the tomato growth monitoring that the specific embodiment of the invention provides.
Specific embodiment
In the following description, for purposes of illustration, it in order to provide the comprehensive understanding to one or more embodiments, explains
Many details are stated.It may be evident, however, that these embodiments can also be realized without these specific details.
To further illustrate the technical scheme of the present invention below with reference to the accompanying drawings and specific embodiments.
Fig. 1 is the intelligent analysis method flow chart one of tomato growth state provided in an embodiment of the present invention, and details are as follows:
In step s101, tomato growth analysis model is pre-established；
In embodiments of the present invention, import tomato growth sample data, for growth period analysis to tomato growth sample data into
Row linear regression analysis, and linear regression model (LRM) is established, convolutional Neural is used to tomato growth sample data for fruit leaf analysis
Network carries out feature extraction, and establishes convolutional neural networks model.
Tomato growth sample data includes growth period sample data and fruit leaf sample data, is analyzed for growth period, is imported
Growth period sample data carries out linear regression analysis, and establishes linear regression model (LRM), wherein growth period sample data, including life
Long-term sample data table, respectively includes growth period sample data, growth period sample characteristics and growth period sample class；For fruit
Leaf sample data carries out feature extraction using convolutional neural networks, and establishes convolutional neural networks model, fruit leaf sample data,
Including fruit leaf sample data table, fruit leaf sample data, fruit leaf sample characteristics and fruit leaf sample class, fruit leaf sample are respectively included
Feature and fruit leaf sample discrimination results.
It is growth period sample set and growth period test set, in the present embodiment, life by growth period sample data random division
The ratio of long-term sample set and growth period test set is 8:2.Linear regression analysis is carried out to growth period sample set, establishes linear return
Return model；In embodiments of the present invention, linear regression model (LRM) formula are as follows: hƟ(x)=Ɵ0+Ɵ1X, wherein0With1For parameter.
The characteristics of linear regression model (LRM) is that modeling speed is fast, does not need very complicated calculating, still runs in the case where data volume is big
Speed is quickly.The understanding and explanation of each variable can be provided according to coefficient.It is very sensitive to exceptional value.
It in embodiments of the present invention, is fruit leaf sample set and fruit leaf test set by fruit leaf sample data random division, at this
In embodiment, the ratio of fruit leaf sample set and fruit leaf test set is 8:2.Spy is carried out using convolutional neural networks to fruit leaf sample set
Sign is extracted, and convolutional neural networks model is established；
In embodiments of the present invention, feature extraction is carried out using convolutional neural networks to fruit leaf sample set, by adopting from picture
Set algorithm carries out characteristics extraction according to position, profile from picture, characteristic value include leaf color coloured silk, leaf shape, fruit shapes,
Fruit color, fruit number, fruit size, fruit height are low.
As the preferred embodiment of the present invention, convolutional neural networks include at least two layers of convolutional layer and at least two layers of pond
Layer, wherein convolutional layer is using ReLu function as activation primitive, and convolutional neural networks model structure can be with are as follows: input layer, convolution
Layer, pond layer, convolutional layer, pond layer, output layer or input layer, convolutional layer, convolutional layer, pond layer, pond layer, output layer,
Depending on the number of convolutional layer and pond layer is needed according to model.A full articulamentum can also be increased after the layer of pond.Also
It can increase between each layer again several layers of, model is reduced to training set degree of fitting using Dropout connection type between layers, is increased
Strong model generalization ability.
Wherein, input layer is the tomato growth image acquired, and output layer is to calculate as a result, the formula of convolutional layer are as follows: s
(i,j)=(X*W)(i,j)+b=, wherein n_in is the number of input matrix, either
The last one-dimensional dimension of tensor.Xk represents k-th of input matrix.Wk represents k-th of sub- convolution nuclear matrix of convolution kernel.s(i,
J) value of s, that is, convolution kernel W corresponding output matrix corresponding position element.
After image data is by convolutional layer, it will usually use an active coating.The purpose is to one in convolutional layer
In just through linear calculating operation system introduce nonlinear characteristic.The present embodiment uses ReLU function as activation primitive, will
Element value corresponding less than 0 position in the tensor of output all becomes 0.ReLU function has huge acceleration to make model convergence
With, to gradient disappear the problem of it is also helpful.ReLU function is linear correction function, and effect is to guarantee that the network after training is complete
Have sparsity, operand and data dimension can also be reduced.ReLU function is defined as: ReLU (x)=max (0, x).
Most after-bay layer carries out aggregate statistics by the feature to different location, some calculated on one region of image is special
The average value or maximum value of sign.Pond layer is compressed to each submatrix of input tensor.
Linear regression model (LRM) is more convenient when analyzing multifactor model compared to other algorithms, in data volume
The speed of service still quickly, and can provide understanding reconciliation for each variable in linear regression when bigger
It releases, as long as data are proper, can achieve very high accuracy rate.Convolutional neural networks compare other neural network algorithms, because its
Shared convolution kernel is so be efficiently, without manual selected characteristic, to have space-invariance, mention automatically for high dimensional data processing
Feature is taken, there is preferable tagsort effect, and for image procossing aspect, the input picture of convolutional network and opening up for network
Flutterring structure can coincide well, can be carried out pattern classification again simultaneously in feature extraction.
In step s 102, the video and image information for obtaining the environmental monitoring data of tomato and currently growing；
In embodiments of the present invention, environmental monitoring data includes weather station monitoring data and soil monitoring data, environmental monitoring
Data include weather station monitoring data and soil monitoring data, and weather station monitoring data include current air temperature, current sky
Air humidity degree, current light, gas concentration lwevel, wind speed, wind direction, rainfall, PM2.5, PM10, carbonomonoxide concentration, ozone are dense
One or more of degree, air pressure and content of nitrogen dioxide；Soil monitoring data include that current soil temperature, current soil are wet
One or more of degree, pH value and conductivity.Wherein, weather station monitoring data pass through temperature sensor respectively, humidity passes
Sensor, optical sensor, carbon dioxide sensor, air velocity transducer, wind transducer, rainfall amount sensor, PM2.5 sensing
Device, PM10 sensor, carbon monoxide transducer, ozone sensor, baroceptor and nitrogen dioxide sensor monitoring obtain.
Soil monitoring data pass through temperature sensor, humidity sensor, pH value sensor and conductivity sensor monitoring respectively and obtain.
In step s 103, judge when environmental monitoring data is more than preset threshold, then send early warning to user's end
End；
In embodiments of the present invention, alert threshold can be arranged in each sensor, once being more than preset threshold, then send early warning
It is prompted to user terminal, wherein user terminal can be the mobile terminal of user, for example, mobile phone, IPAD are also possible to user
Login account, for example, to user's registration Account Logon end send early warning.
As the preferred embodiment of the present invention, if judging to judge current control when environmental monitoring data is more than preset threshold
Whether control equipment is automatic mode, is to inquire the corresponding counter-measure of current environment monitoring data, and send counter-measure pair
The control command answered gives control equipment.Wherein, control equipment can be liquid manure all-in-one machine.If unprecedentedly control equipment is automatic mold
Formula then starts corresponding measure, for example, controlling water when water shortage when monitoring that current environment monitoring data are more than preset threshold
Fertile all-in-one machine watering etc..
It ought tomato plant variation be more than within a preset time default according to video and image information judgement in step S104
When change threshold or the long damaged by vermin of tomato plant, then early warning is sent to user terminal；
In embodiments of the present invention, the variation of current video and image information with tomato plant before preset time period, packet are compared
Plant color change, size variation, change in shape and quantity variation etc. are included, is more than default change threshold, then sends early warning and mention
Show to user terminal.If the same early warning that sends is to user terminal it was found that the long damaged by vermin of tomato plant.
In step s105, image information is imported in tomato growth analysis model, tomato is exported by analytical calculation and is worked as
Preceding growth indexes.
In embodiments of the present invention, image information is directed respectively into linear regression model (LRM) and convolutional neural networks model,
Linear regression model (LRM) calculates image information analysis the growth period growth indexes that tomato is current at export, convolutional neural networks model
The current fruit leaf growth indexes of export tomato are calculated to image information analysis, wherein growth period growth indexes include growth potential, open
Florescence, receipts phase beginning contain one or more of receipts phase and last receipts phase, and fruit leaf growth indexes include titbit type, leaf color, leaf
One or more of shape, fruit shape, fruit color and fruit shoulder.
Neural network is acquired suitable for mass data, is had more complex system, can be provided precise information, but calculation amount
It is bigger, it is suitable for fine data and analyzes.Linear regression is fairly simple, and it is smaller to be suitable for qualitatively analysis, data amount of analysis.Two
Person not only improves accuracy rate, but also improve arithmetic speed in conjunction with come the growth conditions of analyzing tomato.
Fig. 2 is the apparatus structure schematic diagram of tomato growth monitoring provided in an embodiment of the present invention, and details are as follows:
The device of tomato growth monitoring, comprising:
Analysis model built in advance unit 21 pre-establishes tomato growth analysis model；
Wherein, analysis model built in advance unit 21 includes that sample data import modul 211 imports tomato growth sample data；
Linear regression model module 212 is analyzed for growth period and carries out linear regression analysis to tomato growth sample data,
And linear regression model (LRM) is established,
Convolutional neural networks model building module 213 uses convolutional neural networks to tomato growth sample data for fruit leaf analysis
Feature extraction is carried out, and establishes convolutional neural networks model.
In embodiments of the present invention, tomato growth sample data is imported, is analyzed for growth period to tomato growth sample number
According to progress linear regression analysis, and linear regression model (LRM) is established, convolution is used to tomato growth sample data for fruit leaf analysis
Neural network carries out feature extraction, and establishes convolutional neural networks model.
Tomato growth sample data includes growth period sample data and fruit leaf sample data, is analyzed for growth period, is imported
Growth period sample data carries out linear regression analysis, and establishes linear regression model (LRM), wherein growth period sample data, including life
Long-term sample data table, respectively includes growth period sample data, growth period sample characteristics and growth period sample class；For fruit
Leaf sample data carries out feature extraction using convolutional neural networks, and establishes convolutional neural networks model, fruit leaf sample data,
Including fruit leaf sample data table, fruit leaf sample data, fruit leaf sample characteristics and fruit leaf sample class, fruit leaf sample are respectively included
Feature and fruit leaf sample discrimination results.
It is growth period sample set and growth period test set, in the present embodiment, life by growth period sample data random division
The ratio of long-term sample set and growth period test set is 8:2.Linear regression analysis is carried out to growth period sample set, establishes linear return
Return model；In embodiments of the present invention, linear regression model (LRM) formula are as follows: hƟ(x)=Ɵ0+Ɵ1X, wherein0With1For parameter.
The characteristics of linear regression model (LRM) is that modeling speed is fast, does not need very complicated calculating, still runs in the case where data volume is big
Speed is quickly.The understanding and explanation of each variable can be provided according to coefficient.It is very sensitive to exceptional value.
It in embodiments of the present invention, is fruit leaf sample set and fruit leaf test set by fruit leaf sample data random division, at this
In embodiment, the ratio of fruit leaf sample set and fruit leaf test set is 8:2.Spy is carried out using convolutional neural networks to fruit leaf sample set
Sign is extracted, and convolutional neural networks model is established；
In embodiments of the present invention, feature extraction is carried out using convolutional neural networks to fruit leaf sample set, by adopting from picture
Set algorithm carries out characteristics extraction according to position, profile from picture, characteristic value include leaf color coloured silk, leaf shape, fruit shapes,
Fruit color, fruit number, fruit size, fruit height are low.
As the preferred embodiment of the present invention, convolutional neural networks include at least two layers of convolutional layer and at least two layers of pond
Layer, wherein convolutional layer is using ReLu function as activation primitive, and convolutional neural networks model structure can be with are as follows: input layer, convolution
Layer, pond layer, convolutional layer, pond layer, output layer or input layer, convolutional layer, convolutional layer, pond layer, pond layer, output layer,
Depending on the number of convolutional layer and pond layer is needed according to model.A full articulamentum can also be increased after the layer of pond.Also
It can increase between each layer again several layers of, model is reduced to training set degree of fitting using Dropout connection type between layers, is increased
Strong model generalization ability.
Wherein, input layer is the tomato growth image acquired, and output layer is to calculate as a result, the formula of convolutional layer are as follows: s
(i,j)=(X*W)(i,j)+b=, wherein n_in is the number of input matrix, or
It is the last one-dimensional dimension of tensor.Xk represents k-th of input matrix.Wk represents k-th of sub- convolution nuclear matrix of convolution kernel.s
The value of (i, j) s, that is, convolution kernel W corresponding output matrix corresponding position element.
After image data is by convolutional layer, it will usually use an active coating.The purpose is to one in convolutional layer
In just through linear calculating operation system introduce nonlinear characteristic.The present embodiment uses ReLU function as activation primitive, will
Element value corresponding less than 0 position in the tensor of output all becomes 0.ReLU function has huge acceleration to make model convergence
With, to gradient disappear the problem of it is also helpful.ReLU function is linear correction function, and effect is to guarantee that the network after training is complete
Have sparsity, operand and data dimension can also be reduced.ReLU function is defined as: ReLU (x)=max (0, x).
Most after-bay layer carries out aggregate statistics by the feature to different location, some calculated on one region of image is special
The average value or maximum value of sign.Pond layer is compressed to each submatrix of input tensor.
Linear regression model (LRM) is more convenient when analyzing multifactor model compared to other algorithms, in data volume
The speed of service still quickly, and can provide understanding reconciliation for each variable in linear regression when bigger
It releases, as long as data are proper, can achieve very high accuracy rate.Convolutional neural networks compare other neural network algorithms, because its
Shared convolution kernel is so be efficiently, without manual selected characteristic, to have space-invariance, mention automatically for high dimensional data processing
Feature is taken, there is preferable tagsort effect, and for image procossing aspect, the input picture of convolutional network and opening up for network
Flutterring structure can coincide well, can be carried out pattern classification again simultaneously in feature extraction.
The video and image information that data image acquiring unit 22 obtains the environmental monitoring data of tomato and currently grows；
In embodiments of the present invention, environmental monitoring data includes weather station monitoring data and soil monitoring data, environmental monitoring
Data include weather station monitoring data and soil monitoring data, and weather station monitoring data include current air temperature, current sky
Air humidity degree, current light, gas concentration lwevel, wind speed, wind direction, rainfall, PM2.5, PM10, carbonomonoxide concentration, ozone are dense
One or more of degree, air pressure and content of nitrogen dioxide；Soil monitoring data include that current soil temperature, current soil are wet
One or more of degree, pH value and conductivity.Wherein, weather station monitoring data pass through temperature sensor respectively, humidity passes
Sensor, optical sensor, carbon dioxide sensor, air velocity transducer, wind transducer, rainfall amount sensor, PM2.5 sensing
Device, PM10 sensor, carbon monoxide transducer, ozone sensor, baroceptor and nitrogen dioxide sensor monitoring obtain.
Soil monitoring data pass through temperature sensor, humidity sensor, pH value sensor and conductivity sensor monitoring respectively and obtain.
Data early warning unit 23 judges when environmental monitoring data is more than preset threshold, then sends early warning to user's end
End；
In embodiments of the present invention, alert threshold can be arranged in each sensor, once being more than preset threshold, then send early warning
It is prompted to user terminal, wherein user terminal can be the mobile terminal of user, for example, mobile phone, IPAD are also possible to user
Login account, for example, to user's registration Account Logon end send early warning.
Image prewarning unit 24 ought tomato plant variation be more than within a preset time pre- according to video and image information judgement
If then sending early warning to user terminal when change threshold or the long damaged by vermin of tomato plant；
In embodiments of the present invention, the variation of current video and image information with tomato plant before preset time period, packet are compared
Plant color change, size variation, change in shape and quantity variation etc. are included, is more than default change threshold, then sends early warning and mention
Show to user terminal.If the same early warning that sends is to user terminal it was found that the long damaged by vermin of tomato plant.
Model analysis unit 25 imports image information in tomato growth analysis model, exports tomato by analytical calculation and works as
Preceding growth indexes.
Wherein, model analysis unit 25 includes that image information is directed respectively into linear regression model (LRM) by image import modul 251
In convolutional neural networks model；
252 linear regression model (LRM) of growth period analysis module calculates the current growth period growth of tomato at export to image information analysis
Index；
253 convolutional neural networks model of fruit leaf analysis module calculates the current fruit leaf growth of export tomato to image information analysis and refers to
Mark；
Wherein, the growth period growth indexes include growth potential, florescence, receipts phase beginning, contain one of receipts phase and last receipts phase
Or it is several, the fruit leaf growth indexes include one of titbit type, leaf color, leaf, fruit shape, fruit color and fruit shoulder or several
Kind.
In embodiments of the present invention, image information is directed respectively into linear regression model (LRM) and convolutional neural networks model,
Linear regression model (LRM) calculates image information analysis the growth period growth indexes that tomato is current at export, convolutional neural networks model
The current fruit leaf growth indexes of export tomato are calculated to image information analysis, wherein growth period growth indexes include growth potential, open
Florescence, receipts phase beginning contain one or more of receipts phase and last receipts phase, and fruit leaf growth indexes include titbit type, leaf color, leaf
One or more of shape, fruit shape, fruit color and fruit shoulder.
Neural network is acquired suitable for mass data, is had more complex system, can be provided precise information, but calculation amount
It is bigger, it is suitable for fine data and analyzes.Linear regression is fairly simple, and it is smaller to be suitable for qualitatively analysis, data amount of analysis.Two
Person not only improves accuracy rate, but also improve arithmetic speed in conjunction with come the growth conditions of analyzing tomato.
It further include that control unit 26 judges to work as environmental monitoring in data early warning unit 23 as the preferred embodiment of the present invention
Data are more than preset threshold, then whether send early warning to the current control equipment of judgement after user terminal is automatic mode,
It is to inquire the corresponding counter-measure of current environment monitoring data, and send the corresponding control command of counter-measure and set to control
It is standby.
In embodiments of the present invention, if judging when environmental monitoring data is more than preset threshold, judge current control equipment
Whether it is automatic mode, is to inquire the corresponding counter-measure of current environment monitoring data, and send the corresponding control of counter-measure
Control equipment is given in system order.Wherein, control equipment can be liquid manure all-in-one machine.If unprecedentedly control equipment is automatic mode, In
When monitoring that current environment monitoring data are more than preset threshold, then start corresponding measure, for example, controlling liquid manure one when water shortage
Machine watering etc..
The present invention is by pre-establishing tomato growth analysis model；The environmental monitoring data for obtaining tomato and currently growth
Video and image information；Judge then to send early warning to user terminal when environmental monitoring data is more than preset threshold；Root
According to video and image information judgement ought within a preset time tomato plant variation be more than default change threshold or tomato plant with
When insect pest, then early warning is sent to user terminal；Image information is imported in tomato growth analysis model, analytical calculation is passed through
Export the current growth indexes of tomato.
The utility model has the advantages that both having been monitored by environmental monitoring and video image pre- to play to monitor the upgrowth situation of tomato plant
Police acts on, while the growth indexes of tomato are analyzed further through algorithm model, to realize that real Internet of Things is intelligently planted.
It is worth noting that, those of ordinary skill in the art will appreciate that: the step of realizing above method embodiment or portion
This can be accomplished by hardware associated with program instructions step by step, and program above-mentioned can store in computer readable storage medium
In, which when being executed, executes step including the steps of the foregoing method embodiments, and storage medium above-mentioned include: ROM, RAM,
The various media that can store program code such as magnetic or disk.
It should be noted that all the embodiments in this specification are described in a progressive manner, each embodiment weight
Point explanation is the difference from other embodiments, and the same or similar parts between the embodiments can be referred to each other.
For device class embodiment, since it is basically similar to the method embodiment, so no longer being repeated, related place referring to
The part of embodiment of the method illustrates.
The specific embodiment of invention is described in detail above, but it is only used as example, the present invention is not intended to limit
In specific embodiments described above.For a person skilled in the art, any equivalent modifications that the invention is carried out
Or substitute also all among scope of the invention, therefore, the made equalization in the case where not departing from the spirit and principles in the present invention range
Transformation and modification, improvement etc., all should be contained within the scope of the invention.
Claims (10)
1. a kind of method of tomato growth monitoring, which is characterized in that the described method includes:
Pre-establish tomato growth analysis model；
The video and image information for obtaining the environmental monitoring data of tomato and currently growing；
Judge then to send early warning to user terminal when the environmental monitoring data is more than preset threshold；
According to the video and image information judgement ought within a preset time tomato plant variation be more than default change threshold or kind
When the long damaged by vermin of eggplant plant, then early warning is sent to user terminal；
Described image information is imported in the tomato growth analysis model, the current growth of tomato is exported by analytical calculation and is referred to
Mark.
2. the method as described in claim 1, which is characterized in that in the judgement when the environmental monitoring data is more than default threshold
When value, then after sending early warning to user terminal further include:
Whether the current control equipment of judgement is automatic mode, be inquire the corresponding counter-measure of current environment monitoring data, and
The corresponding control command of the counter-measure is sent to the control equipment.
3. the method as described in claim 1, which is characterized in that the tomato growth analysis model that pre-establishes specifically includes:
Tomato growth sample data is imported, linear regression point is carried out to the tomato growth sample data for growth period analysis
Analysis, and linear regression model (LRM) is established, the tomato growth sample data is carried out using convolutional neural networks for fruit leaf analysis
Feature extraction, and establish convolutional neural networks model.
4. method as claimed in claim 3, which is characterized in that described that described image information is imported the tomato growth analysis
In model, the current growth indexes of tomato are exported by analytical calculation specifically:
Described image information is directed respectively into linear regression model (LRM) and convolutional neural networks model, the linear regression model (LRM) pair
Described image information analysis calculates the growth period growth indexes that tomato is current at export, and the convolutional neural networks model is to described
Image information analysis calculates the current fruit leaf growth indexes of export tomato, wherein the growth period growth indexes include growth potential,
Florescence, receipts phase beginning contain one or more of receipts phase and last receipts phase, and the fruit leaf growth indexes include titbit type, leaf
One or more of color, leaf, fruit shape, fruit color and fruit shoulder.
5. the method as described in claim 1, which is characterized in that the environmental monitoring data include weather station monitoring data and
Soil monitoring data, the weather station monitoring data include current air temperature, present air humidity, current light, titanium dioxide
Concentration of carbon, wind speed, wind direction, rainfall, PM2.5, PM10, carbonomonoxide concentration, ozone concentration, air pressure and content of nitrogen dioxide
One or more of, the soil monitoring data include in current soil temperature, current soil humidity, pH value and conductivity
One or more.
6. a kind of device of tomato growth monitoring, which is characterized in that described device includes:
Analysis model built in advance unit, for pre-establishing tomato growth analysis model；
Data image acquiring unit, for obtaining the environmental monitoring data and the currently video that grows and image information of tomato；
Data early warning unit, for judge when the environmental monitoring data is more than preset threshold, then send early warning to
Family terminal；
Image prewarning unit, for ought tomato plant variation be more than within a preset time according to the video and image information judgement
When default change threshold or the long damaged by vermin of tomato plant, then early warning is sent to user terminal；
Model analysis unit is led for importing described image information in the tomato growth analysis model by analytical calculation
The current growth indexes of tomato out.
7. device as claimed in claim 6, which is characterized in that described device further include:
Control unit is more than preset threshold for working as the environmental monitoring data in data early warning unit judges, then sends early warning
Whether the current control equipment of judgement is automatic mode after being prompted to user terminal, is to inquire current environment monitoring data to correspond to
Counter-measure, and send the corresponding control command of the counter-measure to the control equipment.
8. device as claimed in claim 6, which is characterized in that the analysis model built in advance unit includes:
Sample data import modul, for importing tomato growth sample data；
Linear regression model module, for carrying out linear regression to the tomato growth sample data for growth period analysis
Analysis, and linear regression model (LRM) is established,
Convolutional neural networks model building module, for using convolution mind to the tomato growth sample data for fruit leaf analysis
Feature extraction is carried out through network, and establishes convolutional neural networks model.
9. device as claimed in claim 8, which is characterized in that the model analysis unit includes:
Image import modul, for described image information to be directed respectively into linear regression model (LRM) and convolutional neural networks model；
Growth period analysis module, current to tomato at described image information analysis calculating export for the linear regression model (LRM)
Growth period growth indexes；
Fruit leaf analysis module, current to described image information analysis calculating export tomato for the convolutional neural networks model
Fruit leaf growth indexes；
Wherein, the growth period growth indexes include growth potential, florescence, receipts phase beginning, contain one of receipts phase and last receipts phase
Or it is several, the fruit leaf growth indexes include one of titbit type, leaf color, leaf, fruit shape, fruit color and fruit shoulder or several
Kind.
10. device as claimed in claim 6, which is characterized in that the environmental monitoring data include weather station monitoring data with
And soil monitoring data, the weather station monitoring data include current air temperature, present air humidity, current light, dioxy
It is dense to change concentration of carbon, wind speed, wind direction, rainfall, PM2.5, PM10, carbonomonoxide concentration, ozone concentration, air pressure and nitrogen dioxide
One or more of degree, the soil monitoring data include current soil temperature, current soil humidity, pH value and conductivity
One or more of.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910742244.XA CN110472557B (en) | 2019-08-13 | 2019-08-13 | Tomato growth monitoring method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910742244.XA CN110472557B (en) | 2019-08-13 | 2019-08-13 | Tomato growth monitoring method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110472557A true CN110472557A (en) | 2019-11-19 |
| CN110472557B CN110472557B (en) | 2023-06-02 |
Family
ID=68510481
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910742244.XA Active CN110472557B (en) | 2019-08-13 | 2019-08-13 | Tomato growth monitoring method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110472557B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111426347A (en) * | 2020-04-15 | 2020-07-17 | 河北冀云气象技术服务有限责任公司 | Crop growth condition characteristic acquisition system and method |
| CN112055324A (en) * | 2020-08-26 | 2020-12-08 | 陈一丰 | Tomato planting greenhouse temperature control method and system for Internet of things |
| CN112710661A (en) * | 2020-12-25 | 2021-04-27 | 河北北方学院 | Potato breeding monitoring and analyzing method and system |
| DE202022102591U1 (en) | 2022-05-12 | 2022-06-27 | Deepak Batham | System for monitoring plant health in precision agriculture using image processing and convolutional neural network |
| CN117540934A (en) * | 2024-01-08 | 2024-02-09 | 山东科翔智能科技有限公司 | Intelligent monitoring system for wheat growth period based on data analysis |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2013005725A (en) * | 2011-06-22 | 2013-01-10 | Nikon Corp | Growth degree detection device, plant cultivation system, plant cultivation plant, growth degree detection method, and program |
| CN103471514A (en) * | 2013-09-12 | 2013-12-25 | 齐鲁工业大学 | Garlic classifying method based on machine vision and unitary linear recursive analysis |
| CN106688705A (en) * | 2017-01-13 | 2017-05-24 | 湖南理工学院 | Intelligent planting greenhouse and monitoring method used for same |
| CN106778845A (en) * | 2016-12-01 | 2017-05-31 | 浙江省柯桥中学 | A kind of vegetation growth state monitoring method based on leaf color detection |
| WO2018040323A1 (en) * | 2016-08-30 | 2018-03-08 | 深圳前海弘稼科技有限公司 | Plants applicable growth model establishing method and device |
| CN108304812A (en) * | 2018-02-07 | 2018-07-20 | 郑州大学西亚斯国际学院 | A kind of crop disease recognition methods based on convolutional neural networks and more video images |
| CN109191074A (en) * | 2018-08-27 | 2019-01-11 | 宁夏大学 | Wisdom orchard planting management system |
| CN109284771A (en) * | 2018-08-03 | 2019-01-29 | 中国农业大学 | A kind of tomato growth model determination method and device |
| CN109344738A (en) * | 2018-09-12 | 2019-02-15 | 杭州睿琪软件有限公司 | The recognition methods of crop diseases and pest crop smothering and device |
| JP2019083745A (en) * | 2017-11-07 | 2019-06-06 | ヤンマー株式会社 | Grown state prospecting apparatus |
| CN110009043A (en) * | 2019-04-09 | 2019-07-12 | 广东省智能制造研究所 | A kind of pest and disease damage detection method based on depth convolutional neural networks |
-
2019
- 2019-08-13 CN CN201910742244.XA patent/CN110472557B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2013005725A (en) * | 2011-06-22 | 2013-01-10 | Nikon Corp | Growth degree detection device, plant cultivation system, plant cultivation plant, growth degree detection method, and program |
| CN103471514A (en) * | 2013-09-12 | 2013-12-25 | 齐鲁工业大学 | Garlic classifying method based on machine vision and unitary linear recursive analysis |
| WO2018040323A1 (en) * | 2016-08-30 | 2018-03-08 | 深圳前海弘稼科技有限公司 | Plants applicable growth model establishing method and device |
| CN106778845A (en) * | 2016-12-01 | 2017-05-31 | 浙江省柯桥中学 | A kind of vegetation growth state monitoring method based on leaf color detection |
| CN106688705A (en) * | 2017-01-13 | 2017-05-24 | 湖南理工学院 | Intelligent planting greenhouse and monitoring method used for same |
| JP2019083745A (en) * | 2017-11-07 | 2019-06-06 | ヤンマー株式会社 | Grown state prospecting apparatus |
| CN108304812A (en) * | 2018-02-07 | 2018-07-20 | 郑州大学西亚斯国际学院 | A kind of crop disease recognition methods based on convolutional neural networks and more video images |
| CN109284771A (en) * | 2018-08-03 | 2019-01-29 | 中国农业大学 | A kind of tomato growth model determination method and device |
| CN109191074A (en) * | 2018-08-27 | 2019-01-11 | 宁夏大学 | Wisdom orchard planting management system |
| CN109344738A (en) * | 2018-09-12 | 2019-02-15 | 杭州睿琪软件有限公司 | The recognition methods of crop diseases and pest crop smothering and device |
| CN110009043A (en) * | 2019-04-09 | 2019-07-12 | 广东省智能制造研究所 | A kind of pest and disease damage detection method based on depth convolutional neural networks |
Non-Patent Citations (1)
| Title |
|---|
| 丁永军 等: "基于多光谱图像技术的番茄营养素诊断模型", 《农业工程学报》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111426347A (en) * | 2020-04-15 | 2020-07-17 | 河北冀云气象技术服务有限责任公司 | Crop growth condition characteristic acquisition system and method |
| CN112055324A (en) * | 2020-08-26 | 2020-12-08 | 陈一丰 | Tomato planting greenhouse temperature control method and system for Internet of things |
| CN112055324B (en) * | 2020-08-26 | 2023-08-18 | 陈一丰 | Tomato planting greenhouse temperature control method and system for Internet of things |
| CN112710661A (en) * | 2020-12-25 | 2021-04-27 | 河北北方学院 | Potato breeding monitoring and analyzing method and system |
| DE202022102591U1 (en) | 2022-05-12 | 2022-06-27 | Deepak Batham | System for monitoring plant health in precision agriculture using image processing and convolutional neural network |
| CN117540934A (en) * | 2024-01-08 | 2024-02-09 | 山东科翔智能科技有限公司 | Intelligent monitoring system for wheat growth period based on data analysis |
| CN117540934B (en) * | 2024-01-08 | 2024-04-05 | 山东科翔智能科技有限公司 | Intelligent monitoring system for wheat growth period based on data analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110472557B (en) | 2023-06-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110472557A (en) | A kind of method and device of tomato growth monitoring | |
| CN110427063A (en) | A kind of tomato growth monitoring management platform based on Internet of Things | |
| CN209517198U (en) | A kind of wisdom agricultural standardization management system | |
| CN104077725B (en) | The monitoring of potato planting Internet of Things, control and information service cloud platform integrated system | |
| CN111008733B (en) | Crop growth control method and system | |
| CN112450056A (en) | Water, fertilizer and pesticide integrated intelligent irrigation system based on machine learning algorithm | |
| CN103605353B (en) | A kind of agriculture intelligent optimization supervisory system based on cloud service and optimization method for supervising | |
| CN213092145U (en) | Intelligent management measurement and control system for plant growth | |
| CN209802305U (en) | system for be used for crops output aassessment | |
| CN102789579A (en) | Identification method for stressed state of water fertilizer of greenhouse crop on basis of computer vision technology | |
| CN113034301A (en) | Crop growth management system and method | |
| CN115936916A (en) | Intelligent agricultural service platform based on big data | |
| CN106971409B (en) | Maize canopy leaf color modeling and method | |
| CN114365614A (en) | Water and fertilizer accurate regulation and control method, intelligent equipment and system based on Internet of things | |
| CN115294518B (en) | Intelligent monitoring method and system for precise greenhouse cultivation of horticultural plants | |
| CN112465109A (en) | Green house controlling means based on cloud limit is in coordination | |
| Musa et al. | An intelligent plant dissease detection system for smart hydroponic using convolutional neural network | |
| CN109213240A (en) | A kind of strawberry greenhouse wireless monitor and control system based on self adaptive control | |
| CN111415368B (en) | Method for quickly measuring dropping speed of liquid drops of mobile equipment based on deep learning | |
| CN115604301A (en) | Planting environment monitoring system based on artificial intelligence | |
| CN111802218A (en) | Automatic irrigation method and system for fruit trees | |
| Ma et al. | Integration agricultural knowledge and Internet of Things for multi-agent deficit irrigation control | |
| CN207039650U (en) | Long-range multiterminal implant system based on IOT technologies | |
| KR20220036433A (en) | Fruit growth monitoring system and fruit growth monitoring method thereof | |
| CN111726407A (en) | Fog calculation monitoring technology for famous flowers and medicinal plants cultivation in intelligent plant factory environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |