CN114708495B - Multi-source irrigation information fusion decision method and system - Google Patents
Multi-source irrigation information fusion decision method and system Download PDFInfo
- Publication number
- CN114708495B CN114708495B CN202210232293.0A CN202210232293A CN114708495B CN 114708495 B CN114708495 B CN 114708495B CN 202210232293 A CN202210232293 A CN 202210232293A CN 114708495 B CN114708495 B CN 114708495B
- Authority
- CN
- China
- Prior art keywords
- irrigation
- information
- decision
- probability
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000003973 irrigation Methods 0.000 title claims abstract description 741
- 230000002262 irrigation Effects 0.000 title claims abstract description 741
- 230000004927 fusion Effects 0.000 title claims abstract description 118
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000013528 artificial neural network Methods 0.000 claims abstract description 197
- 239000008186 active pharmaceutical agent Substances 0.000 claims abstract description 161
- 238000012549 training Methods 0.000 claims description 41
- 210000002569 neuron Anatomy 0.000 claims description 21
- 230000008635 plant growth Effects 0.000 claims description 15
- 239000013598 vector Substances 0.000 claims description 10
- 238000010586 diagram Methods 0.000 claims description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 230000012010 growth Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 2
- 239000002689 soil Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 239000007921 spray Substances 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a multi-source irrigation information fusion decision method and system. A plurality of irrigation information is obtained. And based on the irrigation information, obtaining a rough irrigation decision through an irrigation rough decision model. Based on irrigation information and irrigation modes, a plurality of pieces of irrigation probability information are obtained through DS evidence theory. And obtaining a detailed irrigation decision through an irrigation model based on the plurality of irrigation probability information and the rough irrigation decision. And obtaining the probability of the decision by using DS evidence theory, and obtaining the decision method with the highest probability in the roughly set decision methods according to the probability. And meanwhile, a neural network is used to obtain a decision method with highest probability in the roughly set decision methods. And obtaining classification information and irrigation probability of the trained fused subdivision decisions by using the DS irrigation neural network, the rough irrigation neural network and the fusion neural network. And multiplying the finally obtained irrigation probability with the irrigation probability obtained by the DS to obtain a detailed irrigation decision more accurately.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a multi-source irrigation information fusion decision method and system.
Background
The multi-source information fusion is based on multiple (homogeneous or heterogeneous) information sources, and is combined in space or time according to a specific standard to obtain consistency interpretation or description of the measured object, and the information system has better performance. At the fusion level, the fusion model generally performs information fusion processing at three levels of data, features and decisions. However, the data fusion disadvantage is heavy computational burden, poor real-time performance and good fault tolerance capability to handle instability and uncertainty of the sensor data itself, and is only applicable to the original data fusion of the same kind of sensor. Decision layer fusion requires compression of sensor measurement data, which not only has high processing costs, but also can lose a lot of detailed information. Due to the rational utilization of water resources, it is particularly important. The general irrigation method can not be adjusted according to weather conditions, season conditions, temperature conditions and the like.
Disclosure of Invention
The invention aims to provide a multi-source irrigation information fusion decision method and system, which are used for solving the problems in the prior art.
In a first aspect, an embodiment of the present invention provides a multi-source irrigation information fusion decision method, including:
Obtaining a plurality of irrigation information; the irrigation information represents information of the current plant; the irrigation information comprises an irrigation image and irrigation text information; the irrigation image represents information shot for plants in different ranges; the irrigation text information represents record information of plant conditions in different ranges; the irrigation information is information fused on a data layer;
based on irrigation information, obtaining a rough irrigation decision through an irrigation rough decision model;
based on irrigation information and irrigation modes, obtaining a plurality of pieces of irrigation probability information through DS evidence theory;
based on a plurality of irrigation probability information and rough irrigation decisions, obtaining detailed irrigation decisions through an irrigation model;
the irrigation model comprises a DS irrigation neural network, a rough irrigation neural network, a first fusion neural network, a second fusion neural network and a third fusion neural network:
the DS irrigates the neural network input to be a plurality of pieces of irrigation probability information; the rough irrigation neural network inputs a plurality of irrigation probability information; the input of the first fusion neural network is a plurality of pieces of irrigation probability information; the second fusion neural network is the output of the DS irrigation neural network and the output of the rough irrigation neural network; the input of the third fused neural network is the output of the first fused neural network and the output of the second fused neural network.
Optionally, the obtaining the rough irrigation decision based on the irrigation information through the rough irrigation decision model includes:
the irrigation rough decision model comprises a convolution neural network and an irrigation decision neural network;
inputting the plant image into a convolutional neural network to obtain a plant feature map;
multiplying the plant characteristic diagram with the convolution block with the same size to obtain single-column characteristic information;
and inputting the single-column characteristic information and the plant text information into an irrigation decision neural network to obtain a rough irrigation decision.
Optionally, the training method of the irrigation decision neural network comprises the following steps:
obtaining a training set, wherein the training set comprises training irrigation information and irrigation decision marking information; the training irrigation information comprises training irrigation image characteristic information and training irrigation text information; the irrigation decision marking information comprises water yield, irrigation time, irrigation duration and irrigation mode corresponding to different states of the plants;
inputting training irrigation information into an irrigation decision neural network to obtain rough irrigation information;
the quantity of neurons of an input layer of the irrigation decision neural network is the same as that of training irrigation information; the number of neurons of the output layer of the irrigation decision neural network is the same as the number of rough irrigation information; the irrigation decision neural network is in a full connection mode;
Obtaining a first loss value through a loss function by the rough irrigation information and the irrigation decision marking information;
obtaining the current training iteration times of an irrigation decision neural network and the preset maximum iteration times of the irrigation decision neural network training;
and stopping training when the first loss value is smaller than or equal to a threshold value or the training iteration number reaches the maximum iteration number, and obtaining the trained irrigation decision neural network.
Optionally, the obtaining, based on the irrigation information and the irrigation manner, a plurality of irrigation probability information according to DS evidence theory includes:
obtaining plant growth probabilities in a plurality of ranges based on irrigation information and rough irrigation decisions;
obtaining a plurality of pieces of irrigation probability information based on a plurality of ranges of plant growth probability and DS evidence theory; the irrigation probability information includes irrigation decisions and corresponding irrigation probabilities.
Optionally, the obtaining, based on the plurality of irrigation probability information and the rough irrigation decision, a detailed irrigation decision through an irrigation model includes:
obtaining DS irrigation probability decisions; the DS irrigation probability decision comprises a DS irrigation decision and a DS irrigation probability; the DS irrigation decision is an irrigation decision with the irrigation probability larger than other irrigation probabilities in the plurality of pieces of irrigation probability information; the DS irrigation probability is an irrigation probability corresponding to a DS irrigation decision;
Obtaining a rough irrigation probability decision based on the irrigation probability information and the rough irrigation decision; the rough irrigation probability decision comprises a rough irrigation decision and a rough irrigation probability; the rough irrigation probability is an irrigation probability corresponding to a rough irrigation decision in the irrigation probability information;
and inputting a plurality of irrigation probability information, DS irrigation probability decisions and rough irrigation probability decisions into an irrigation model to obtain detailed irrigation decisions.
Optionally, the inputting the multiple irrigation probability information, DS irrigation probability decisions and rough irrigation probability decisions into an irrigation model to obtain detailed irrigation decisions includes:
inputting a plurality of pieces of irrigation probability information into a DS irrigation neural network based on DS irrigation probability decisions to obtain DS irrigation characteristic information;
inputting a plurality of pieces of irrigation probability information into the rough irrigation neural network based on the rough irrigation probability decision to obtain rough irrigation characteristic information;
inputting a plurality of pieces of irrigation probability information into a first fusion neural network to obtain first fusion information;
and inputting the DS irrigation characteristic information, the rough irrigation characteristic information and the first fusion information into a second fusion neural network to obtain a detailed irrigation decision.
Optionally, the inputting the plurality of irrigation probability information into the DS irrigation neural network based on the DS irrigation probability decision to obtain DS irrigation characteristic information includes:
Dividing elements in the DS irrigation probability decision to obtain a plurality of DS detailed irrigation probability decisions;
the number of neurons of the DS irrigation neural network input layer is the number of a plurality of irrigation probability information; the number of neurons of the DS irrigation neural network output layer is the number of a plurality of DS detailed irrigation probability decisions;
inputting the multiple pieces of irrigation probability information into a DS irrigation neural network to obtain DS irrigation characteristic information; the DS irrigation characteristic information comprises probabilities of a plurality of DS detail irrigation probability decisions.
Optionally, the inputting the plurality of irrigation probability information into the first fusion neural network to obtain first fusion information includes:
dividing elements in the DS irrigation probability decision to obtain a plurality of DS detailed irrigation probability decisions;
the number of neurons of the first fusion neural network input layer is the number of a plurality of pieces of irrigation probability information; the number of neurons of the first fusion neural network output layer is the sum of the decision number of dividing the elements in the DS irrigation probability decision and the decision number of dividing the elements in the rough irrigation probability decision;
inputting a plurality of pieces of irrigation probability information into a first fusion neural network to obtain first fusion information; the first fusion information comprises the probability of the decision of dividing by the element in the DS irrigation probability decision and the decision of dividing by the element in the rough irrigation probability decision.
Optionally, inputting the DS irrigation characteristic information, the rough irrigation characteristic information, and the first fusion information into a second fusion neural network to obtain a detailed irrigation decision, including:
the DS irrigation characteristic information, the rough irrigation characteristic information and the first fusion information are connected with a second fusion neural network through full connection, and classification is carried out to obtain detailed irrigation characteristic vectors; the elements in the detailed irrigation feature vector comprise irrigation probability elements and irrigation decision type elements; the irrigation decision type element is expressed as a decision type which refines DS irrigation probability decision and rough irrigation probability decision; the irrigation probability elements are expressed as irrigation probabilities corresponding to DS irrigation probability decisions and rough irrigation probability decisions;
multiplying the irrigation probability in the detailed irrigation feature vector by the probability of the irrigation decision type element to obtain the probability of the new irrigation decision type element;
obtaining a detailed irrigation decision based on the probability of the new irrigation decision type element; the detailed irrigation decision is an irrigation decision with the probability of the new irrigation decision type element being greater than the probability of other new irrigation decision type elements.
In a second aspect, an embodiment of the present invention provides a multi-source irrigation information fusion decision system, including:
Irrigation information acquisition module: obtaining a plurality of irrigation information; the irrigation information represents information of the current plant; the irrigation information comprises an irrigation image and irrigation text information; the irrigation image represents information shot for plants in different ranges; the irrigation text information represents record information of plant conditions in different ranges; the irrigation information is information fused on a data layer;
rough irrigation decision module: based on irrigation information, obtaining a rough irrigation decision through an irrigation rough decision model;
DS module: based on irrigation information and irrigation modes, obtaining a plurality of pieces of irrigation probability information through DS evidence theory;
and a detailed irrigation decision module: based on a plurality of irrigation probability information and rough irrigation decisions, obtaining detailed irrigation decisions through an irrigation model;
the irrigation model comprises a DS irrigation neural network, a rough irrigation neural network, a first fusion neural network, a second fusion neural network and a third fusion neural network:
the DS irrigates the neural network input to be a plurality of pieces of irrigation probability information; the rough irrigation neural network inputs a plurality of irrigation probability information; the input of the first fusion neural network is a plurality of pieces of irrigation probability information; the second fusion neural network is the output of the DS irrigation neural network and the output of the rough irrigation neural network; the input of the third fused neural network is the output of the first fused neural network and the output of the second fused neural network.
Compared with the prior art, the embodiment of the invention achieves the following beneficial effects:
the embodiment of the invention also provides a multi-source irrigation information fusion decision method and a system, wherein the method comprises the following steps: a plurality of irrigation information is obtained. The irrigation information represents information of the current plant. The irrigation information comprises irrigation images and irrigation text information. The irrigation images represent information captured for different ranges of plants. The irrigation text information represents recorded information of plant conditions in different ranges. The irrigation information is information fused in a data layer. And based on the irrigation information, obtaining a rough irrigation decision through an irrigation rough decision model. Based on irrigation information and irrigation modes, a plurality of pieces of irrigation probability information are obtained through DS evidence theory. Based on the multiple pieces of irrigation probability information and the rough irrigation decision, a detailed irrigation block is obtained through an irrigation model. The irrigation model comprises a DS irrigation neural network, a rough irrigation neural network, a first fusion neural network, a second fusion neural network and a third fusion neural network: the DS irrigates the neural network input to a plurality of irrigation probability information. The rough irrigation neural network is used for inputting a plurality of irrigation probability information. The input of the first fusion neural network is a plurality of irrigation probability information. The second fused neural network is the output of the DS irrigation neural network and the output of the rough irrigation neural network. The input of the third fused neural network is the output of the first fused neural network and the output of the second fused neural network.
The convolution method is firstly used to obtain information which is not easy to be detected manually. And obtaining the probability of the decision by using DS evidence theory, and obtaining the decision method with the highest probability in the roughly set decision methods according to the probability. And meanwhile, a neural network is used to obtain a decision method with highest probability in the roughly set decision methods. And detecting decisions subdivided by the roughly set decision method by using the DS irrigation neural network and the roughly irrigation neural network respectively to obtain classification information and irrigation probability. And meanwhile, the classification information and irrigation probability of the subdivision decisions after the fusion through training are obtained through the first fusion neural network. And inputting the obtained classification information into a second fusion neural network for classification, multiplying the irrigation probability obtained finally by the irrigation probability obtained by the DS, and obtaining a detailed irrigation decision more accurately.
Drawings
Fig. 1 is a flowchart of a multi-source irrigation information fusion decision system provided by an embodiment of the present invention.
Fig. 2 is a schematic diagram of a process of a multi-source irrigation information fusion decision system according to an embodiment of the present invention.
Fig. 3 is a schematic structural diagram of a multi-source irrigation information fusion decision system according to an embodiment of the present invention.
Fig. 4 is a schematic block diagram of an electronic device according to an embodiment of the present invention.
The marks in the figure: a bus 500; a receiver 501; a processor 502; a transmitter 503; a memory 504; bus interface 505.
Detailed Description
The present invention will be described in detail with reference to the accompanying drawings.
Example 1
As shown in fig. 1, an embodiment of the present invention provides a multi-source irrigation information fusion decision method, which includes:
s101: obtaining a plurality of irrigation information; the irrigation information represents information of the current plant; the irrigation information comprises an irrigation image and irrigation text information; the irrigation image represents information shot for plants in different ranges; the irrigation text information represents record information of plant conditions in different ranges; the irrigation information is information fused in a data layer.
In this embodiment, the irrigation information includes soil humidity, soil temperature, growth period of planted crops, categories of planted crops, weather conditions, season conditions, temperature conditions, and the like.
S102: based on irrigation information, obtaining a rough irrigation decision through an irrigation rough decision model;
wherein the rough irrigation decision represents the irrigation decision obtained in a larger range.
S103: based on irrigation information, a plurality of pieces of irrigation probability information are obtained through DS evidence theory.
S104: and obtaining a detailed irrigation decision through an irrigation model based on the plurality of irrigation probability information and the rough irrigation decision.
The irrigation model enables the irrigation mode to be adopted in order to achieve the optimal growth state under different conditions of plants.
The process of the source irrigation information fusion decision method is shown in fig. 2.
The irrigation model comprises a DS irrigation neural network, a rough irrigation neural network, a first fusion neural network, a second fusion neural network and a third fusion neural network:
the DS irrigates the neural network input to be a plurality of pieces of irrigation probability information; the rough irrigation neural network inputs a plurality of irrigation probability information; the input of the first fusion neural network is a plurality of pieces of irrigation probability information; the second fusion neural network is the output of the DS irrigation neural network and the output of the rough irrigation neural network; the input of the third fused neural network is the output of the first fused neural network and the output of the second fused neural network.
Optionally, the obtaining the rough irrigation decision based on the irrigation information through the rough irrigation decision model includes:
The irrigation rough decision model comprises a convolutional neural network and an irrigation decision neural network.
And inputting the plant image into a convolutional neural network to obtain a plant characteristic diagram.
The convolutional neural network is trained, and can rapidly detect characteristic information of a large amount of time required for manual detection of plants.
Multiplying the plant characteristic diagram by the convolution block with the same size to obtain single-column characteristic information.
And inputting the single-column characteristic information and the plant text information into an irrigation decision neural network to obtain a rough irrigation decision.
By the method, the convolution myth network is used for extracting the information of the plants in the pictures, including the colors, the intensity, the shape and the like of the plants, and can quickly detect the characteristic information of the plants, which requires a large amount of time for manual detection. And simultaneously inputting the information and the plant text information into an irrigation decision neural network to carry out detection classification, so that a rough irrigation decision is obtained more accurately.
Optionally, the training method of the irrigation decision neural network comprises the following steps:
obtaining a training set, wherein the training set comprises training irrigation information and irrigation decision marking information; the training irrigation information comprises training irrigation image characteristic information and training irrigation text information; the irrigation decision marking information comprises water yield, irrigation time, irrigation duration and irrigation mode corresponding to different states of the plants.
In this embodiment, the water yield corresponding to different states of the plant is 0-5ml for 1 minute, 5-10ml for 1 minute, and 10-15ml for 1 minute. The irrigation time includes 0.00-6.00,6.00-12.00, 12.00-18.00, 18.00-24.00. Irrigation times included 0.5 hours, 1 hour, 1.5 hours, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours. Irrigation modes include drip irrigation and spray irrigation
And inputting the training irrigation information into an irrigation decision neural network to obtain rough irrigation information.
The number of neurons of the input layer of the irrigation decision neural network is the same as the number of training irrigation information. The number of neurons at the output layer of the irrigation decision neural network is the same as the number of rough irrigation information. The irrigation decision neural network is in a fully-connected mode.
And obtaining a first loss value through a loss function by the rough irrigation information and the irrigation decision marking information.
Wherein the loss function is a cross entropy function.
Obtaining the current training iteration times of the irrigation decision neural network and the preset maximum iteration times of the irrigation decision neural network training.
In this embodiment, the maximum iteration number is 8000.
And stopping training when the first loss value is smaller than or equal to a threshold value or the training iteration number reaches the maximum iteration number, and obtaining the trained irrigation decision neural network.
Optionally, the obtaining irrigation probability information based on the irrigation information through DS evidence theory includes:
and obtaining a plurality of range plant growth probabilities based on the irrigation information and the rough irrigation decision.
Wherein, the plant growth probabilities of the multiple ranges are as follows:
TABLE 1
| Assume irrigation decisions | First range plant growth | Second range plant growth | Third range plant growth |
| Mode A | 0.64 | 0.24 | 0.07 |
| Mode B | 0.02 | 0.53 | 0.12 |
Wherein, the first range of plant growth, the second range of plant growth and the third range of plant growth represent different growth probabilities of plants in each region under different irrigation modes in different regions which divide the plants into the regions with the irrigation regions as the centers of circles. The calculation mode of the growth probability is to grow the same plant in different irrigation modes.
Based on the plant growth probability and DS evidence theory in a plurality of ranges, a plurality of pieces of irrigation probability information are obtained.
And obtaining normalized coefficients of plant growth probabilities in a plurality of ranges through DS evidence theory, and calculating irrigation probability information of each irrigation decision by using a Dempster synthesis rule.
Through the method, probability of plant growth of a plurality of irrigation decisions is obtained through DS evidence theory, so that which decision corresponds to the optimal irrigation mode is known.
Optionally, the obtaining, based on the plurality of irrigation probability information and the rough irrigation decision, a detailed irrigation decision through an irrigation model includes:
obtaining DS irrigation probability decisions; the DS irrigation probability decision comprises a DS irrigation decision and a DS irrigation probability; the DS irrigation decision is an irrigation decision with the irrigation probability larger than other irrigation probabilities in the plurality of pieces of irrigation probability information; the DS irrigation probability is an irrigation probability corresponding to a DS irrigation decision;
obtaining a rough irrigation probability decision based on the irrigation probability information and the rough irrigation decision; the rough irrigation probability decision comprises a rough irrigation decision and a rough irrigation probability; the rough irrigation probability is an irrigation probability corresponding to a rough irrigation decision in the irrigation probability information;
and inputting a plurality of irrigation probability information, DS irrigation probability decisions and rough irrigation probability decisions into an irrigation model to obtain detailed irrigation decisions.
By the method, irrigation decisions obtained by two different methods are calculated by using a probability-combined method, so that the accuracy of the probability decisions is improved.
Optionally, the inputting the multiple irrigation probability information, DS irrigation probability decisions and rough irrigation probability decisions into an irrigation model to obtain detailed irrigation decisions includes:
And inputting a plurality of pieces of irrigation probability information into the DS irrigation neural network based on the DS irrigation probability decision to obtain DS irrigation characteristic information.
And finding out a corresponding DS irrigation neural network through DS irrigation probability decision.
And inputting a plurality of pieces of irrigation probability information into the rough irrigation neural network based on the rough irrigation probability decision to obtain rough irrigation characteristic information.
And finding out a corresponding rough irrigation neural network through rough irrigation probability decision.
And inputting the irrigation probability information into a first fusion neural network to obtain first fusion information.
And finding out a corresponding first fusion neural network through the DS irrigation probability decision and the rough irrigation probability decision.
And inputting the DS irrigation characteristic information and the rough irrigation characteristic information into a second fusion neural network to obtain second fusion information.
And inputting the first fusion information and the second fusion information into a third fusion neural network to obtain a detailed irrigation decision.
By the method, because the irrigation rough decision model and the DS evidence theory are adopted, better irrigation information is probably obtained, and the water yield, the irrigation time, the irrigation duration and the irrigation mode in the rough irrigation decision are subdivided. The independent information and the fused information of irrigation are combined, so that the finding of an irrigation mode suitable for irrigation is enhanced.
Optionally, the inputting the plurality of irrigation probability information into the DS irrigation neural network based on the DS irrigation probability decision to obtain DS irrigation characteristic information includes:
dividing elements in the DS irrigation probability decision to obtain a plurality of DS detailed irrigation probability decisions;
the number of neurons of the DS irrigation neural network input layer is the number of a plurality of irrigation probability information; the number of neurons of the DS irrigation neural network output layer is the number of a plurality of DS detailed irrigation probability decisions;
inputting the multiple pieces of irrigation probability information into a DS irrigation neural network to obtain DS irrigation characteristic information; the DS irrigation characteristic information comprises probabilities of a plurality of DS detail irrigation probability decisions.
The irrigation probability information is fully connected with the first fusion neural network, but most neurons connected with the irrigation probability information except the random number are randomly hidden by using the mode of connecting the random number, so that the best combination mode of the irrigation probability information is obtained in the training process.
By the method, the probability of subdivision decisions under DS irrigation decisions is obtained independently, and the irrigation probability can be obtained accurately by setting a neural network.
Dividing elements in the rough irrigation probability decision to obtain a plurality of rough detailed irrigation probability decisions;
The number of neurons of the rough irrigation neural network input layer is the number of a plurality of irrigation probability information; the number of neurons of the rough irrigation neural network output layer is the number of a plurality of rough detailed irrigation probability decisions;
inputting the multiple pieces of irrigation probability information into a rough irrigation neural network to obtain rough irrigation characteristic information; the coarse irrigation characteristic information includes probabilities of a plurality of coarse detailed irrigation probability decisions.
Optionally, the inputting the plurality of irrigation probability information into the first fusion neural network to obtain first fusion information includes:
dividing elements in the DS irrigation probability decision to obtain a plurality of DS detailed irrigation probability decisions;
the number of neurons of the first fusion neural network input layer is the number of a plurality of pieces of irrigation probability information; the number of neurons of the first fusion neural network output layer is the sum of the decision number of dividing the elements in the DS irrigation probability decision and the decision number of dividing the elements in the rough irrigation probability decision;
inputting a plurality of pieces of irrigation probability information into a first fusion neural network to obtain first fusion information; the first fusion information comprises the probability of the decision of dividing by the element in the DS irrigation probability decision and the decision of dividing by the element in the rough irrigation probability decision.
By the method, the probability of the subdivided irrigation decision in the DS irrigation decision and the rough irrigation decision is judged simultaneously by setting the output structure of the neural network, so that information can be fused, and the detailed irrigation probability can be calculated accurately afterwards.
Optionally, inputting the DS irrigation characteristic information, the rough irrigation characteristic information, and the first fusion information into a second fusion neural network to obtain a detailed irrigation decision, including:
the DS irrigation characteristic information, the rough irrigation characteristic information and the first fusion information are connected with a second fusion neural network through full connection, and classification is carried out to obtain detailed irrigation characteristic vectors; the elements in the detailed irrigation feature vector comprise irrigation probability elements and irrigation decision type elements; the irrigation decision type element is expressed as a decision type which refines DS irrigation probability decision and rough irrigation probability decision; the irrigation probability element is expressed as irrigation probability corresponding to the DS irrigation probability decision and the rough irrigation probability decision.
The number of elements in the detailed irrigation characteristic vector is the number of irrigation decisions in the DS irrigation probability decision and the rough irrigation probability decision.
Multiplying the irrigation probability in the detailed irrigation feature vector by the probability of the irrigation decision type element to obtain the probability of the new irrigation decision type element.
Wherein the irrigation probability is a probability that the confidence represents that the current predicted detailed irrigation decision is a true detailed irrigation decision.
Obtaining a detailed irrigation decision based on the probability of the new irrigation decision type element; the detailed irrigation decision is an irrigation decision with the probability of the new irrigation decision type element being greater than the probability of other new irrigation decision type elements.
By the method, under the condition of considering DS probability, accurate detailed irrigation decisions can be obtained through the neural network.
According to the method, plant information which is manually repeated or difficult to detect in plant parts is obtained through a convolution network, then a plurality of rough decisions are set, and the optimal rough irrigation decision in the rough decisions is found through a trained neural network method. And finding the optimal DS irrigation decision in the rough decision through DS evidence theory. The rough decision and the corresponding irrigation probability are combined with the DS irrigation decision and the corresponding irrigation probability, and the detailed irrigation decision is obtained through an irrigation model, so that the more accurate and more effective detailed irrigation decision can be obtained through the joint consideration of the rough irrigation decision and the DS irrigation decision. Firstly, inputting a plurality of pieces of irrigation probability information into a rough irrigation neural network, and extracting optimal decisions subdivided in rough irrigation decisions. Inputting a plurality of pieces of irrigation probability information into the DS irrigation neural network, and extracting the optimal decision subdivided in the DS irrigation decision. And extracting the optimal irrigation decision in the DS irrigation decision and the rough irrigation decision by inputting a plurality of pieces of irrigation probability information into the first fused irrigation neural network. And finally multiplying the optimal decisions obtained by the three methods with the probabilities of the optimal decisions respectively, so that the optimal detailed irrigation decisions can be obtained more accurately.
Example 2
Based on the multi-source irrigation information fusion decision method, the embodiment of the invention also provides a multi-source irrigation information fusion decision system, which comprises an irrigation information acquisition module, a rough irrigation decision module, a DS module and a detailed irrigation decision module.
Irrigation information acquisition module: obtaining a plurality of irrigation information; the irrigation information represents information of the current plant; the irrigation information comprises an irrigation image and irrigation text information; the irrigation image represents information shot for plants in different ranges; the irrigation text information represents record information of plant conditions in different ranges; the irrigation information is information fused on a data layer;
rough irrigation decision module: based on irrigation information, obtaining a rough irrigation decision through an irrigation rough decision model;
DS module: based on irrigation information and irrigation modes, obtaining a plurality of pieces of irrigation probability information through DS evidence theory;
and a detailed irrigation decision module: based on a plurality of irrigation probability information and rough irrigation decisions, obtaining detailed irrigation decisions through an irrigation model;
the irrigation model comprises a DS irrigation neural network, a rough irrigation neural network, a first fusion neural network, a second fusion neural network and a third fusion neural network:
The DS irrigates the neural network input to be a plurality of pieces of irrigation probability information; the rough irrigation neural network inputs a plurality of irrigation probability information; the input of the first fusion neural network is a plurality of pieces of irrigation probability information; the second fusion neural network is the output of the DS irrigation neural network and the output of the rough irrigation neural network; the input of the third fused neural network is the output of the first fused neural network and the output of the second fused neural network.
The specific manner in which the various modules perform the operations in the systems of the above embodiments have been described in detail herein with respect to the embodiments of the method, and will not be described in detail herein.
An embodiment of the present invention further provides an electronic device, as shown in fig. 4, including a memory 504, a processor 502, and a computer program stored in the memory 504 and capable of running on the processor 502, where the steps of any one of the above-described methods for determining fusion of multi-source irrigation information are implemented by the processor 502 when the program is executed.
Where in FIG. 4 a bus architecture (represented by bus 500), bus 500 may include any number of interconnected buses and bridges, with bus 500 linking together various circuits, including one or more processors, represented by processor 502, and memory, represented by memory 504. Bus 500 may also link together various other circuits such as peripheral devices, voltage regulators, power management circuits, etc., as are well known in the art and, therefore, will not be described further herein. Bus interface 505 provides an interface between bus 500 and receiver 501 and transmitter 503. The receiver 501 and the transmitter 503 may be the same element, i.e. a transceiver, providing a means for communicating with various other apparatus over a transmission medium. The processor 502 is responsible for managing the bus 500 and general processing, while the memory 504 may be used to store data used by the processor 502 in performing operations.
The embodiment of the invention also provides a computer readable storage medium, on which a computer program is stored, which when being executed by a processor, implements the steps of any of the methods of the multi-source irrigation information fusion decision method described above and the data referred to above.
The algorithms and displays presented herein are not inherently related to any particular computer, virtual system, or other apparatus. Various general-purpose systems may also be used with the teachings herein. The required structure for a construction of such a system is apparent from the description above. In addition, the present invention is not directed to any particular programming language. It will be appreciated that the teachings of the present invention described herein may be implemented in a variety of programming languages, and the above description of specific languages is provided for disclosure of enablement and best mode of the present invention.
In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be construed as reflecting the intention that: i.e., the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.
Those skilled in the art will appreciate that the modules in the apparatus of the embodiments may be adaptively changed and disposed in one or more apparatuses different from the embodiments. The modules or units or components of the embodiments may be combined into one module or unit or component and, furthermore, they may be divided into a plurality of sub-modules or sub-units or sub-components. Any combination of all features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or units of any method or apparatus so disclosed, may be used in combination, except insofar as at least some of such features and/or processes or units are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings), may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Furthermore, those skilled in the art will appreciate that while some embodiments herein include some features but not others included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments can be used in any combination.
Various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that some or all of the functions of some or all of the components in an apparatus according to embodiments of the present invention may be implemented in practice using a microprocessor or Digital Signal Processor (DSP). The present invention can also be implemented as an apparatus or device program (e.g., a computer program and a computer program product) for performing a portion or all of the methods described herein. Such a program embodying the present invention may be stored on a computer readable medium, or may have the form of one or more signals. Such signals may be downloaded from an internet website, provided on a carrier signal, or provided in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, third, etc. do not denote any order. These words may be interpreted as names.
Claims (10)
1. The multi-source irrigation information fusion decision method is characterized by comprising the following steps of:
obtaining a plurality of irrigation information; the irrigation information represents information of the current plant; the irrigation information comprises an irrigation image and irrigation text information; the irrigation image represents information shot for plants in different ranges; the irrigation text information represents record information of plant conditions in different ranges; the irrigation information is information fused on a data layer;
based on irrigation information, obtaining a rough irrigation decision through an irrigation rough decision model;
based on irrigation information and irrigation modes, obtaining a plurality of pieces of irrigation probability information through DS evidence theory;
based on a plurality of irrigation probability information and rough irrigation decisions, obtaining detailed irrigation decisions through an irrigation model;
the irrigation model comprises a DS irrigation neural network, a rough irrigation neural network, a first fusion neural network, a second fusion neural network and a third fusion neural network:
the DS irrigates the neural network input to be a plurality of pieces of irrigation probability information; the input of the rough irrigation neural network is a plurality of irrigation probability information; the input of the first fusion neural network is a plurality of pieces of irrigation probability information; the inputs of the second fusion neural network are the output of the DS irrigation neural network and the output of the rough irrigation neural network; the input of the third fused neural network is the output of the first fused neural network and the output of the second fused neural network.
2. The multi-source irrigation information fusion decision method according to claim 1, wherein the obtaining the rough irrigation decision by irrigating the rough decision model based on the irrigation information comprises:
the irrigation rough decision model comprises a convolution neural network and an irrigation decision neural network;
inputting the plant image into a convolutional neural network to obtain a plant feature map;
multiplying the plant characteristic diagram with the convolution block with the same size to obtain single-column characteristic information;
and inputting the single-column characteristic information and the plant text information into an irrigation decision neural network to obtain a rough irrigation decision.
3. The multi-source irrigation information fusion decision method according to claim 1, wherein the training method of the irrigation decision neural network comprises the following steps:
obtaining a training set, wherein the training set comprises training irrigation information and irrigation decision marking information; the training irrigation information comprises training irrigation image characteristic information and training irrigation text information; the irrigation decision marking information comprises water yield, irrigation time, irrigation duration and irrigation mode corresponding to different states of the plants;
inputting training irrigation information into an irrigation decision neural network to obtain rough irrigation information;
The quantity of neurons of an input layer of the irrigation decision neural network is the same as that of training irrigation information; the number of neurons of the output layer of the irrigation decision neural network is the same as the number of rough irrigation information; the irrigation decision neural network is in a full connection mode;
obtaining a first loss value through a loss function by the rough irrigation information and the irrigation decision marking information;
obtaining the current training iteration times of an irrigation decision neural network and the preset maximum iteration times of the irrigation decision neural network training;
and stopping training when the first loss value is smaller than or equal to a threshold value or the training iteration number reaches the maximum iteration number, and obtaining the trained irrigation decision neural network.
4. The multi-source irrigation information fusion decision method according to claim 1, wherein the obtaining a plurality of irrigation probability information based on irrigation information and irrigation modes through DS evidence theory comprises:
obtaining plant growth probabilities in a plurality of ranges based on irrigation information and rough irrigation decisions;
obtaining a plurality of pieces of irrigation probability information based on a plurality of ranges of plant growth probability and DS evidence theory; the irrigation probability information includes irrigation decisions and corresponding irrigation probabilities.
5. The multi-source irrigation information fusion decision method according to claim 1, wherein the obtaining the detailed irrigation decision by the irrigation model based on the plurality of irrigation probability information and the rough irrigation decision comprises:
obtaining DS irrigation probability decisions; the DS irrigation probability decision comprises a DS irrigation decision and a DS irrigation probability; the DS irrigation decision is an irrigation decision with the irrigation probability larger than other irrigation probabilities in the plurality of pieces of irrigation probability information; the DS irrigation probability is an irrigation probability corresponding to a DS irrigation decision;
obtaining a rough irrigation probability decision based on the irrigation probability information and the rough irrigation decision; the rough irrigation probability decision comprises a rough irrigation decision and a rough irrigation probability; the rough irrigation probability is an irrigation probability corresponding to a rough irrigation decision in the irrigation probability information;
and inputting a plurality of irrigation probability information, DS irrigation probability decisions and rough irrigation probability decisions into an irrigation model to obtain detailed irrigation decisions.
6. The method of claim 5, wherein inputting the plurality of irrigation probability information, DS irrigation probability decisions, and rough irrigation probability decisions into the irrigation model to obtain detailed irrigation decisions, comprises:
Inputting a plurality of pieces of irrigation probability information into a DS irrigation neural network based on DS irrigation probability decisions to obtain DS irrigation characteristic information;
inputting a plurality of pieces of irrigation probability information into the rough irrigation neural network based on the rough irrigation probability decision to obtain rough irrigation characteristic information;
inputting a plurality of pieces of irrigation probability information into a first fusion neural network to obtain first fusion information;
and inputting the DS irrigation characteristic information, the rough irrigation characteristic information and the first fusion information into a second fusion neural network to obtain a detailed irrigation decision.
7. The method of claim 6, wherein inputting the plurality of irrigation probability information into the DS irrigation neural network based on the DS irrigation probability decision to obtain the DS irrigation characteristic information, comprises:
dividing elements in the DS irrigation probability decision to obtain a plurality of DS detailed irrigation probability decisions;
the number of neurons of the DS irrigation neural network input layer is the number of a plurality of irrigation probability information; the number of neurons of the DS irrigation neural network output layer is the number of a plurality of DS detailed irrigation probability decisions;
inputting the multiple pieces of irrigation probability information into a DS irrigation neural network to obtain DS irrigation characteristic information; the DS irrigation characteristic information comprises probabilities of a plurality of DS detail irrigation probability decisions.
8. The method for determining a fusion decision of multiple irrigation information according to claim 6, wherein the inputting the multiple irrigation probability information into the first fusion neural network to obtain the first fusion information comprises:
dividing elements in the DS irrigation probability decision to obtain a plurality of DS detailed irrigation probability decisions;
the number of neurons of the first fusion neural network input layer is the number of a plurality of pieces of irrigation probability information; the number of neurons of the first fusion neural network output layer is the sum of the decision number of dividing the elements in the DS irrigation probability decision and the decision number of dividing the elements in the rough irrigation probability decision;
inputting a plurality of pieces of irrigation probability information into a first fusion neural network to obtain first fusion information; the first fusion information comprises the probability of the decision of dividing by the element in the DS irrigation probability decision and the decision of dividing by the element in the rough irrigation probability decision.
9. The method of claim 6, wherein inputting the DS irrigation characteristic information, the rough irrigation characteristic information, and the first fusion information into the second fusion neural network to obtain a detailed irrigation decision, comprises:
The DS irrigation characteristic information, the rough irrigation characteristic information and the first fusion information are connected with a second fusion neural network through full connection, and classification is carried out to obtain detailed irrigation characteristic vectors; the elements in the detailed irrigation feature vector comprise irrigation probability elements and irrigation decision type elements; the irrigation decision type element is expressed as a decision type which refines DS irrigation probability decision and rough irrigation probability decision; the irrigation probability elements are expressed as irrigation probabilities corresponding to DS irrigation probability decisions and rough irrigation probability decisions;
multiplying the irrigation probability in the detailed irrigation feature vector by the probability of the irrigation decision type element to obtain the probability of the new irrigation decision type element;
obtaining a detailed irrigation decision based on the probability of the new irrigation decision type element; the detailed irrigation decision is an irrigation decision with the probability of the new irrigation decision type element being greater than the probability of other new irrigation decision type elements.
10. A multi-source irrigation information fusion decision system, comprising:
irrigation information acquisition module: obtaining a plurality of irrigation information; the irrigation information represents information of the current plant; the irrigation information comprises an irrigation image and irrigation text information; the irrigation image represents information shot for plants in different ranges; the irrigation text information represents record information of plant conditions in different ranges; the irrigation information is information fused on a data layer;
Rough irrigation decision module: based on irrigation information, obtaining a rough irrigation decision through an irrigation rough decision model;
DS module: based on irrigation information and irrigation modes, obtaining a plurality of pieces of irrigation probability information through DS evidence theory;
and a detailed irrigation decision module: based on a plurality of irrigation probability information and rough irrigation decisions, obtaining detailed irrigation decisions through an irrigation model;
the irrigation model comprises a DS irrigation neural network, a rough irrigation neural network, a first fusion neural network, a second fusion neural network and a third fusion neural network:
the DS irrigates the neural network input to be a plurality of pieces of irrigation probability information; the input of the rough irrigation neural network is a plurality of irrigation probability information; the input of the first fusion neural network is a plurality of pieces of irrigation probability information; the inputs of the second fusion neural network are the output of the DS irrigation neural network and the output of the rough irrigation neural network; the input of the third fused neural network is the output of the first fused neural network and the output of the second fused neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210232293.0A CN114708495B (en) | 2022-03-09 | 2022-03-09 | Multi-source irrigation information fusion decision method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210232293.0A CN114708495B (en) | 2022-03-09 | 2022-03-09 | Multi-source irrigation information fusion decision method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114708495A CN114708495A (en) | 2022-07-05 |
| CN114708495B true CN114708495B (en) | 2024-04-09 |
Family
ID=82167911
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210232293.0A Active CN114708495B (en) | 2022-03-09 | 2022-03-09 | Multi-source irrigation information fusion decision method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114708495B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102819673A (en) * | 2012-07-26 | 2012-12-12 | 中国农业科学院农田灌溉研究所 | Multisource irrigation information fusion method and device |
| CN105892287A (en) * | 2016-05-09 | 2016-08-24 | 河海大学常州校区 | Crop irrigation strategy based on fuzzy judgment and decision making system |
| CN110009251A (en) * | 2019-04-15 | 2019-07-12 | 西安邮电大学 | A kind of farmland multi-source information dynamic regulation fusion method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11516976B2 (en) * | 2017-12-11 | 2022-12-06 | The Texas A&M University System | Irrigation system control with predictive water balance capabilities |
-
2022
- 2022-03-09 CN CN202210232293.0A patent/CN114708495B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102819673A (en) * | 2012-07-26 | 2012-12-12 | 中国农业科学院农田灌溉研究所 | Multisource irrigation information fusion method and device |
| CN105892287A (en) * | 2016-05-09 | 2016-08-24 | 河海大学常州校区 | Crop irrigation strategy based on fuzzy judgment and decision making system |
| CN110009251A (en) * | 2019-04-15 | 2019-07-12 | 西安邮电大学 | A kind of farmland multi-source information dynamic regulation fusion method and system |
Non-Patent Citations (2)
| Title |
|---|
| 基于模糊粗糙集和D-S证据理论的多源灌溉信息融合方法;陈智芳;王景雷;孙景生;刘祖贵;宋妮;高阳;;计算机应用;20131001(第10期);全文 * |
| 基于线性规划和MODFLOW耦合技术的人民胜利渠灌区水资源优化配置研究;胡艳玲等;灌溉排水学报;20191215;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114708495A (en) | 2022-07-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111126662B (en) | Irrigation decision making method, device, server and medium based on big data | |
| CA3129643C (en) | Refined average for zoning method and system | |
| CN109508881B (en) | Sea island region classification and ecological resource value evaluation method | |
| US20220076068A1 (en) | Method for constructing a convolution neural network based on farmland images, electronic device using the same | |
| CN111199535A (en) | Animal state monitoring method and device, electronic equipment and storage medium | |
| CN110648020A (en) | Greenhouse crop water demand prediction method and device | |
| CN112308029A (en) | Rainfall station and satellite rainfall data fusion method and system | |
| CN111222539A (en) | Method for optimizing and expanding supervision classification samples based on multi-source multi-temporal remote sensing image | |
| CN114430536B (en) | Agricultural Internet of things method and system based on Beidou positioning module and cloud platform | |
| CN115797781A (en) | Crop identification method and device, computer equipment and storage medium | |
| Jia et al. | Research on water and fertilizer irrigation system of tea plantation | |
| CN108510490A (en) | Method and device for analyzing insect pest trend and computer storage medium | |
| He et al. | Recognition of soybean pods and yield prediction based on improved deep learning model | |
| CN114708495B (en) | Multi-source irrigation information fusion decision method and system | |
| CN104748865A (en) | Multipoint combined correction method used for infrared image | |
| CN116739739A (en) | Loan amount evaluation method and device, electronic equipment and storage medium | |
| CN114862611A (en) | Agricultural planting cloud diagnosis method and system based on edge computing | |
| CN114155440A (en) | Automatic detection method and system for farmland non-farming | |
| Guillaume et al. | Soft computing-based decision support tools for spatial data | |
| Abe et al. | Data science for food, energy and water: A workshop report | |
| CN113822360B (en) | Method and device for determining phenological period of crops | |
| Yu et al. | Maize tassel number and tasseling stage monitoring based on near-ground and UAV RGB images by improved YoloV8 | |
| Saito et al. | Representation of plant structure using XML and its application to cultivation management | |
| CN114648500B (en) | Crop weed detection method and device based on unmanned aerial vehicle and deep learning | |
| CN117132138B (en) | GIS-based intelligent agricultural yield prediction evaluation method and system |
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 |