Tea tree nitrogen absorption type evaluation method and system based on nitrogen transporter activity

Through the evaluation method of tea tree nitrogen absorption type based on nitrogen transporter activity, a comprehensive prediction model is constructed using cluster analysis and CNN sub-model, which solves the problem of poor accuracy of tea tree nitrogen absorption evaluation, and achieves rapid nitrogen absorption capacity of tea tree Effective assessment and support the sustainable development of the tea industry.

CN118940074AActive Publication Date: 2024-11-12SHANDONG AGRICULTURAL UNIVERSITY
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202410935372.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-12
Publication Date
2024-11-12
Estimated Expiration
2044-07-12

AI Technical Summary

Technical Problem

The tea tree has large root system and complex site conditions. The existing soil diagnosis and field test methods have poor diagnostic accuracy, making it difficult to quickly and effectively evaluate the nitrogen absorption capacity of tea trees, which increases the investment cost and environmental risks of the tea industry, and does not meet the requirements of sustainable development.

Method used

A tea tree nitrogen absorption type evaluation method based on nitrogen transporter activity was used to construct a comprehensive prediction model through cluster analysis and CNN sub-model fusion prediction model, and the tea tree nitrogen absorption rate was evaluated, and type division was performed according to the fusion weight of the input parameters.

Benefits of technology

It improves the accuracy and reliability of the nitrogen absorption evaluation of tea trees, can quickly and effectively evaluate the nitrogen absorption capacity of unknown types of tea trees, shorten the breeding time, help adjust nutrient management strategies, and improve agricultural production efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118940074A_ABST
    Figure CN118940074A_ABST
Patent Text Reader

Abstract

The invention discloses a tea tree nitrogen absorption type evaluation method and system based on nitrogen transporter activity, and relates to the technical field of agriculture, and the method comprises the steps: collecting tea tree data, building a total data set, carrying out the type division of the data through a clustering analysis method, and respectively building corresponding training sets; training the CNN sub-models by adopting the corresponding training sets to obtain a low-nitrogen high-efficiency prediction model, a nitrogen high-efficiency prediction model and a high-nitrogen high-efficiency prediction model; calculating a fusion coefficient of input parameters in each prediction model; determining the fusion weight of each input parameter according to the fusion coefficient, and constructing a comprehensive prediction model; and performing nitrogen absorption rate calculation on the to-be-discriminated tea tree through the comprehensive prediction model to obtain a to-be-detected nitrogen absorption rate, and performing type division on the to-be-discriminated tea tree according to a clustering analysis method. According to the method, the comprehensive prediction model is constructed through the input parameter weight of the fusion sub-model, the influence of the nitrogen transporter activity on the nitrogen absorption of the tea tree is considered, and the comprehensiveness and accuracy of the comprehensive prediction model are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of agricultural technology, and more particularly to a method and system for evaluating nitrogen absorption type of tea trees based on nitrogen transporter activity. Background Art

[0002] At present, tea trees are harvested for their fresh leaves. Tea tree nutrient management is the basis for high quality, high yield and stable yield, and nutrient diagnosis is the theoretical and technical basis for nutrient management. Plant nutrient diagnosis methods include soil diagnosis, leaf diagnosis, field tests, tracer atoms, microbial determination and other methods. Since tea trees have a large root system, the tree requires and stores a lot of nutrients, and the site conditions of the garden are complex, the use of soil nutrient diagnosis and field tests often has problems such as poor diagnostic accuracy and low efficiency, and the results are only regarded as reference values ​​in production.

[0003] However, simply providing nutrition to tea trees by fertilizing will not only increase the input cost of the tea industry and the potential environmental risks brought by fertilizers, but also fail to meet the sustainable development requirements of the tea industry. Therefore, it is of great significance to select and breed nitrogen-efficient tea varieties. Tea trees are perennial economic crops, and breeding improved varieties takes a long time. The development of a fast and effective non-destructive identification technology for the nitrogen absorption capacity of tea trees can not only greatly shorten the breeding time of nitrogen-efficient tea varieties, but also keep the selected tea varieties intact for later field verification.

[0004] Therefore, how to evaluate the nitrogen absorption efficiency of tea plants is an urgent problem that technicians in this field need to solve. Summary of the invention

[0005] In view of this, the present invention provides a method and system for evaluating the nitrogen absorption type of tea trees based on the activity of nitrogen transport proteins. By integrating the input parameter weights of different sub-models, a comprehensive prediction model is constructed. The effects of various nitrogen transport protein activities on the nitrogen absorption of tea trees are comprehensively considered, which can well evaluate the nitrogen absorption capacity of unknown types of tea trees and improve the comprehensiveness and accuracy of the comprehensive prediction model.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] The method for evaluating the nitrogen absorption type of tea plants based on the activity of nitrogen transporters includes:

[0008] The activities of nitrogen transport proteins and nitrogen absorption rates of tea plants are collected to establish a total data set, and the data in the total data set are divided into types by cluster analysis and corresponding training sets are established to obtain a low-nitrogen high-efficiency set, a nitrogen high-efficiency set, and a high-nitrogen high-efficiency set;

[0009] The low-nitrogen high-efficiency set, nitrogen high-efficiency set, and high-nitrogen high-efficiency set were used to train CNN sub-models with the same structure, respectively. The activities of nitrogen transport proteins in tea plants were used as input data, and the corresponding nitrogen absorption rate of tea plants was used as output data to obtain low-nitrogen high-efficiency prediction models, nitrogen high-efficiency prediction models, and high-nitrogen high-efficiency prediction models.

[0010] Calculating the fusion coefficient of the input parameters of each model in the low-nitrogen high-efficiency prediction model, the nitrogen high-efficiency prediction model, and the high-nitrogen high-efficiency prediction model; and determining the fusion weight of each input parameter according to the fusion coefficient, and constructing a comprehensive prediction model based on the fusion weights of all input parameters;

[0011] The nitrogen absorption rate of the tea trees to be identified is calculated by the comprehensive prediction model to obtain the nitrogen absorption rate to be tested, and the tea trees to be identified are classified into types according to the cluster analysis method.

[0012] Preferably, the cluster analysis method is k-means, which specifically includes: initializing the classification, initializing three cluster centers, using Euclidean distance to characterize the measure of similarity between samples, calculating the distances from all samples to each initial cluster center, and dividing the sample categories according to the discrimination criterion of the minimum distance; updating the cluster centers, calculating the mean of the samples belonging to each category as the new cluster center of the category, and then re-obtaining the sum of the squares of the distances from all samples to the new cluster center of their category; judging whether the cluster center changes between the previous and subsequent cluster analyses, if it no longer changes, clustering ends, otherwise repeatedly updating the cluster center until clustering ends.

[0013] Preferably, the CNN sub-model specifically includes: selecting 6 input parameters related to the nitrogen absorption rate of tea trees, performing normalization processing, arranging the processed input parameter data in time series as input data, and constructing 6 input data time series input layers; first performing a convolution operation on the input data, and pooling the input data after the convolution operation, and performing a full connection operation on the output data after multiple convolution pooling operations to complete the training of the low-nitrogen and high-efficiency prediction model, the nitrogen and high-efficiency prediction model, and the high-nitrogen and high-efficiency prediction model.

[0014] Preferably, the input parameters specifically include ammonium nitrogen transporters: CsAMT1.1, CsAMT1.2; nitrate nitrogen transporters: CsNRT2.1, CsNRT2.2, CsNRT2.3, CsNRT2.4.

[0015] Preferably, completing the training of the low-nitrogen high-efficiency prediction model, the nitrogen high-efficiency prediction model, and the high-nitrogen high-efficiency prediction model specifically includes: performing convolution operations in different directions on the input data in the input layer, extracting features between the input data by longitudinal convolution, and extracting data features of the same input data at different times by transverse convolution, and performing pooling operations after each convolution calculation to reduce operation parameters.

[0016] Preferably, the performing convolution operations on the input data in different directions specifically includes:

[0017] S1: Use s convolution kernels to perform vertical convolution on the input data in the input layer. The vertical convolution is calculated using the following formula:

[0018] a i,j =f(w m x i,j +z m ),m=1,2,...,s;

[0019] Among them, x i,j is the element in row i and column j of the input data, w m represents the convolution kernel weight, z m is the bias term of the convolution kernel, a i,j is the i-th row and j-th column element of the convolution data, and f is the activation function;

[0020] S2: The input data is convolved by s convolution kernels and activated by the relu function. After activation, s neurons are output, each of which contains a data matrix.

[0021] S3: Average pooling of s neurons. The specific formula is as follows:

[0022]

[0023] Among them, q represents the size of the pooling area, D and F are the length and width of a single neuron data matrix. Since the pooling kernel is vertically pooled, only the length of the neuron matrix will decrease, p i / q,j Represents the i / qth row and jth column element of the output neuron matrix of the pooling layer;

[0024] S4: The neurons output by the pooling layer are then horizontally convolved by g convolution kernels and average pooled again, outputting g neurons;

[0025] S5: The neurons output after two convolution poolings are used as the input of the connection layer. The connection layer integrates the feature information represented by the elements of all neuron matrices into the neurons of the connection layer. Assume that the number of neurons in the connection layer is T. After multiple convolution poolings, each neuron output contains k rows and l columns of elements. Each neuron in the connection layer is convolved by g [k, l] convolution kernels on the elements in each neuron matrix. The specific formula is as follows:

[0026]

[0027] Among them, x k,l is the kth row and lth column element of the input data, w n is the convolution kernel weight, zn is the bias term of the convolution kernel, y u is the value in each neuron of the fully connected layer.

[0028] Preferably, the output layer in S5 is calculated by the linear weighted sum of the output vectors of the connection layer, and the specific formula is as follows:

[0029]

[0030] Among them, the number of input neurons is T, and the output is The output is the nitrogen absorption rate result of tea tree.

[0031] Preferably, the step of calculating the fusion coefficient specifically includes:

[0032] Divide the weight value corresponding to each input parameter in each trained prediction model into different value ranges according to their size;

[0033] Calculate the weight value probability corresponding to each value range, and calculate the weight entropy corresponding to each input parameter in each prediction model according to the weight value probability;

[0034] A fusion coefficient corresponding to the input parameter is calculated according to the weight entropy.

[0035] Preferably, determining the fusion weight of each input parameter according to the fusion coefficient specifically includes: taking the same input parameter in all prediction models as the target, weightedly merging the weight entropy of the input parameters through the fusion coefficient, and obtaining the fusion weight of the target parameter in the comprehensive prediction model.

[0036] The tea tree nitrogen absorption rating system based on nitrogen transporter activity includes:

[0037] A training set establishment module collects the activities of various nitrogen transport proteins and nitrogen absorption rates of tea trees to establish a total data set, divides the data in the total data set into types through cluster analysis and establishes corresponding training sets to obtain a low-nitrogen high-efficiency set, a nitrogen high-efficiency set, and a high-nitrogen high-efficiency set;

[0038] The sub-model training module uses the low-nitrogen high-efficiency set, the nitrogen high-efficiency set, and the high-nitrogen high-efficiency set to train the CNN sub-models of the same structure respectively, takes the activities of the nitrogen transport proteins of the tea tree as the input data, and the corresponding nitrogen absorption rate of the tea tree as the output data, to obtain the low-nitrogen high-efficiency prediction model, the nitrogen high-efficiency prediction model, and the high-nitrogen high-efficiency prediction model;

[0039] A prediction model construction module calculates the fusion coefficient of the input parameters of each model in the low-nitrogen high-efficiency prediction model, the nitrogen high-efficiency prediction model, and the high-nitrogen high-efficiency prediction model; and determines the fusion weight of each input parameter according to the fusion coefficient, and constructs a comprehensive prediction model based on the fusion weights of all input parameters;

[0040] The type classification module calculates the nitrogen absorption rate of the tea trees to be identified through the comprehensive prediction model, obtains the nitrogen absorption rate to be tested, and classifies the tea trees to be identified according to the cluster analysis method.

[0041] It can be seen from the above technical solution that compared with the prior art, the present invention discloses a method and system for evaluating the nitrogen absorption type of tea trees based on the activity of nitrogen transport proteins. By utilizing deep learning models and comprehensive analysis methods, the accuracy and reliability of the evaluation results are improved, and parameters such as the nitrogen transport protein activity of tea trees are comprehensively considered. Through the integrated prediction of multiple sub-models, the nitrogen absorption of tea trees can be comprehensively analyzed, and the characteristics of nutrient absorption of tea trees can be more comprehensively revealed. The nitrogen absorption capacity of unknown types of tea trees can be well evaluated. Through iterative optimization and model updating, the latest nitrogen absorption of tea trees can be obtained in a timely manner, which is helpful to timely adjust nutrient management strategies and improve agricultural production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0043] Figure 1 A flow chart of the method steps provided by the present invention;

[0044] Figure 2 A schematic diagram of the structure provided by the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] The present invention discloses a method for evaluating the nitrogen absorption type of tea plants based on the activity of nitrogen transport proteins, such as Figure 1 As shown, including:

[0047] The activities of nitrogen transport proteins and nitrogen absorption rates of tea trees were collected to establish a total data set. The data in the total data set were divided into types through cluster analysis and corresponding training sets were established to obtain low-nitrogen high-efficiency set, nitrogen high-efficiency set, and high-nitrogen high-efficiency set.

[0048] The low-nitrogen high-efficiency set, nitrogen high-efficiency set, and high-nitrogen high-efficiency set were used to train CNN sub-models with the same structure, respectively. The activities of nitrogen transport proteins in tea plants were used as input data, and the corresponding nitrogen absorption rate of tea plants was used as output data to obtain low-nitrogen high-efficiency prediction models, nitrogen high-efficiency prediction models, and high-nitrogen high-efficiency prediction models.

[0049] Calculate the fusion coefficient of the input parameters of each model in the low-nitrogen high-efficiency prediction model, the nitrogen high-efficiency prediction model, and the high-nitrogen high-efficiency prediction model; determine the fusion weight of each input parameter according to the fusion coefficient, and construct a comprehensive prediction model based on the fusion weights of all input parameters;

[0050] The nitrogen absorption rate of the tea trees to be identified is calculated through a comprehensive prediction model to obtain the nitrogen absorption rate to be tested, and the tea trees to be identified are classified into types based on the cluster analysis method.

[0051] In a specific embodiment, the clustering analysis method is k-means, which specifically includes: initializing the classification, initializing three cluster centers, using Euclidean distance to characterize the measure of similarity between samples, calculating the distances from all samples to each initial cluster center, and dividing the sample categories according to the discrimination criterion of the minimum distance; updating the cluster center, calculating the mean of the samples belonging to each category as the new cluster center of the category, and then re-obtaining the sum of the squares of the distances from all samples to the new cluster center of their category; judging whether the cluster center changes between the previous and subsequent cluster analyses, if it no longer changes, the clustering ends, otherwise the cluster center is repeatedly updated until the clustering ends.

[0052] In a specific embodiment, it includes:

[0053] Initialize the classification, initialize three cluster centers, use Euclidean distance to characterize the metric of similarity between samples, and calculate all samples To each initial cluster center C k The distance J(C k ), divide the sample X according to the minimum distance criterion d =[Jv d ,UCS d Category C k ; c is the total number of samples.

[0054] Update the cluster center and calculate each category {C 1 ,C 2 ,,C 3 The mean of the samples belonging to {u 1 ,u 2 ,u 3} as the new cluster center of this category C′={C 1 ′,C 2 ′,…,C 3′}, and then re-obtain the sum of squared distances J(C') from all samples to the new cluster center of their category;

[0055]

[0056] Where, X d represents the dth nitrogen transporter activity parameter sample, and X d =[Jv d ,UCS d ],u k Represents the k-th cluster center C k The mean of the samples belonging to , and k = 1, 2, 3;

[0057]

[0058] Determine whether the cluster centers C′ and J(C′) change between the two clustering analyses. If they do not change, clustering ends. Otherwise, the cluster centers are updated repeatedly until clustering ends.

[0059] In a specific embodiment, the CNN sub-model specifically includes: selecting 6 input parameters related to the nitrogen absorption rate of tea trees, normalizing them, arranging the processed input parameter data in time series as input data, and constructing 6 input data time series input layers; first performing a convolution operation on the input data, and pooling the input data after the convolution operation, and after multiple convolution pooling, performing a full connection operation on the output data to complete the training of the low-nitrogen high-efficiency prediction model, the nitrogen high-efficiency prediction model, and the high-nitrogen high-efficiency prediction model.

[0060] In a specific embodiment, the CNN sub-model specifically includes an input layer, a convolutional layer, a pooling layer, a connection layer and an output layer, and the input layer, the convolutional layer, the pooling layer, the connection layer and the output layer are sequentially communicated and connected with each other.

[0061] The configuration of the input layer is used to input six input parameters related to the nitrogen absorption rate of tea trees and normalize the six input parameters.

[0062] The convolution layer uses sliding window technology to correspond the six parameter data of a past time interval with the nitrogen absorption rate of tea trees at a future moment to form a time series, and slides by unit time.

[0063] The time series of the CNN sub-model input data is:

[0064] Q I = {Q I (t),Q I (t+1),...,Q I (t+n)}, I = 1, 2, 3, ..., 6;

[0065] The pooling layer selects a sliding window with a width of n from time t to time t+n, and inputs the 6 parameter data row by row to form a matrix.

[0066] The output neurons after two convolution poolings serve as the input of the fully connected layer. The fully connected layer integrates the feature information represented by the elements of all neuron matrices into the neurons of the fully connected layer.

[0067] In a specific embodiment, it also includes using back-propagation technology to update weight parameters, and updating the weights and biases of the convolutional layer by calculating the first-order moment estimate and the second-order moment estimate of the gradient to achieve fine-tuning of the weight parameters.

[0068] In a specific embodiment, the input parameters specifically include ammonium nitrogen transporters: CsAMT1.1, CsAMT1.2; nitrate nitrogen transporters: CsNRT2.1, CsNRT2.2, CsNRT2.3, CsNRT2.4.

[0069] In a specific embodiment, completing the training of a low-nitrogen high-efficiency prediction model, a nitrogen high-efficiency prediction model, and a high-nitrogen high-efficiency prediction model specifically includes: performing convolution operations in different directions on the input data in the input layer, extracting features between input data through longitudinal convolution, and extracting data features of the same input data at different times through transverse convolution, and performing pooling operations after each convolution calculation to reduce operation parameters.

[0070] In a specific embodiment, performing convolution operations on input data in different directions specifically includes:

[0071] S1: Use s convolution kernels to perform vertical convolution on the input data in the input layer. The vertical convolution is calculated using the following formula:

[0072] a i,j =f(w m x i,j +z m ),m=1,2,...,s;

[0073] Among them, x i,j is the element in row i and column j of the input data, w m represents the convolution kernel weight, z m is the bias term of the convolution kernel, a i,j is the i-th row and j-th column element of the convolution data, and f is the activation function;

[0074] S2: The input data is convolved by s convolution kernels and activated by the relu function. After activation, s neurons are output, each of which contains a data matrix.

[0075] S3: Average pooling of s neurons. The specific formula is as follows:

[0076]

[0077] Among them, q represents the size of the pooling area, D and F are the length and width of a single neuron data matrix. Since the pooling kernel is vertically pooled, only the length of the neuron matrix will decrease, p i / q,j Represents the i / qth row and jth column element of the output neuron matrix of the pooling layer;

[0078] S4: The neurons output by the pooling layer are then horizontally convolved by g convolution kernels and average pooled again, outputting g neurons;

[0079] S5: The neurons output after two convolution poolings are used as the input of the connection layer. The connection layer integrates the feature information represented by the elements of all neuron matrices into the neurons of the connection layer. Assume that the number of neurons in the connection layer is T. After multiple convolution poolings, each neuron output contains h rows and l columns of elements. Each neuron in the connection layer is convolved by g [h, l] convolution kernels on the elements in each neuron matrix. The specific formula is as follows:

[0080]

[0081] Among them, x h,l is the element in the hth row and lth column of the input layer data; w n is the convolution kernel weight, z n is the bias term of the convolution kernel; the size of the convolution kernel is the same as the size of the neuron matrix, and a value is output after convolution, y u Represents the value in each neuron of the fully connected layer, that is, the value obtained by adding g values ​​after convolution.

[0082] In a specific embodiment, the output layer in S5 is calculated by the linear weighted sum of the output vectors of the connection layer, and the specific formula is as follows:

[0083]

[0084] Among them, the number of input neurons is T, The output of tea tree nitrogen absorption rate results, low nitrogen high efficiency prediction model, nitrogen high efficiency prediction model and high nitrogen high efficiency prediction model correspond to They are and

[0085] In a specific embodiment, the step of calculating the fusion coefficient specifically includes:

[0086] The weight value corresponding to each input parameter in each prediction model after training is divided into different value ranges according to the size;

[0087] Calculate the weight value probability o corresponding to each value range r , the weight entropy corresponding to each input parameter in each prediction model is calculated according to the weight value probability, as follows:

[0088]

[0089] Among them, R is the number of value ranges, o r represents the probability of the weight value distributed in the rth value range. It should be understood that the weight corresponding to H(w I ) is smaller, indicating that the weight changes less and the corresponding information is less.

[0090] The fusion coefficient of the corresponding input parameter is calculated according to the weight entropy. The calculation of the fusion coefficient is as follows:

[0091]

[0092] Among them, A and B are hyperparameters, is the parameter I in the CNN prediction sub-model The fusion coefficient in .

[0093] In a specific embodiment, determining the fusion weight of each input parameter according to the fusion coefficient specifically includes: taking the same input parameter in all prediction models as the target, weighting and merging the weight entropy of the input parameters through the fusion coefficient, and obtaining the fusion weight of the target parameter in the comprehensive prediction model. The fusion weight calculation is specifically as follows:

[0094]

[0095] in, is the weight corresponding to the input parameter I in the fusion model, is a comprehensive prediction model, Y is the number of CNN prediction sub-models, and Y=3.

[0096] The present invention fuses multiple CNN prediction sub-models in a weighted fusion manner, and performs weighted merger according to the weight entropy corresponding to each parameter in different models, so that weights with high weight entropy corresponding to high information content have a greater impact on the fusion weight, and weights with low weight entropy corresponding to small information content have a smaller impact on the fusion weight. This avoids the practice of cutting off weights with smaller influencing factors in related technologies, and improves the recognition accuracy and robustness of the comprehensive prediction model obtained by fusion.

[0097] The comprehensive prediction model is as follows:

[0098]

[0099] Among them, N τThe input parameters for the protein activity of the tea tree to be identified include the activity data of six nitrogen transporters: ammonium nitrogen transporters CsAMT1.1, CsAMT1.2; nitrate nitrogen transporters CsNRT2.1, CsNRT2.2, CsNRT2.3, and CsNRT2.4.

[0100] The tea tree nitrogen absorption grade assessment system based on nitrogen transporter activity, such as Figure 2 As shown, including:

[0101] The training set establishment module collects the activities of various nitrogen transport proteins and nitrogen absorption rates of tea trees to establish a total data set. The data in the total data set are divided into types through cluster analysis and corresponding training sets are established to obtain low-nitrogen high-efficiency sets, nitrogen high-efficiency sets, and high-nitrogen high-efficiency sets.

[0102] The sub-model training module uses the low-nitrogen high-efficiency set, the nitrogen high-efficiency set, and the high-nitrogen high-efficiency set to train the CNN sub-models of the same structure respectively, takes the activities of the nitrogen transport proteins of the tea tree as the input data, and the corresponding nitrogen absorption rate of the tea tree as the output data, to obtain the low-nitrogen high-efficiency prediction model, the nitrogen high-efficiency prediction model, and the high-nitrogen high-efficiency prediction model;

[0103] The prediction model construction module calculates the fusion coefficient of the input parameters of each model in the low-nitrogen high-efficiency prediction model, the nitrogen high-efficiency prediction model, and the high-nitrogen high-efficiency prediction model; determines the fusion weight of each input parameter according to the fusion coefficient, and constructs a comprehensive prediction model based on the fusion weights of all input parameters;

[0104] The type classification module calculates the nitrogen absorption rate of the tea trees to be identified through a comprehensive prediction model, obtains the nitrogen absorption rate to be tested, and classifies the tea trees to be identified based on the cluster analysis method.

[0105] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0106] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. Evaluation method of nitrogen absorption type of tea plants based on nitrogen transporter activity, It is characterized in that include: The activities of nitrogen transport proteins and nitrogen absorption rates of tea plants are collected to establish a total data set, and the data in the total data set are divided into types by cluster analysis and corresponding training sets are established to obtain a low-nitrogen high-efficiency set, a nitrogen high-efficiency set, and a high-nitrogen high-efficiency set; The low-nitrogen high-efficiency set, nitrogen high-efficiency set, and high-nitrogen high-efficiency set were used to train CNN sub-models with the same structure, respectively. The activities of nitrogen transport proteins in tea plants were used as input data, and the corresponding nitrogen absorption rate of tea plants was used as output data to obtain low-nitrogen high-efficiency prediction models, nitrogen high-efficiency prediction models, and high-nitrogen high-efficiency prediction models. Calculating the fusion coefficient of the input parameters of each model in the low-nitrogen high-efficiency prediction model, the nitrogen high-efficiency prediction model, and the high-nitrogen high-efficiency prediction model; and determining the fusion weight of each input parameter according to the fusion coefficient, and constructing a comprehensive prediction model based on the fusion weights of all input parameters; The nitrogen absorption rate of the tea trees to be identified is calculated by the comprehensive prediction model to obtain the nitrogen absorption rate to be tested, and the tea trees to be identified are classified into types according to the cluster analysis method.

2. The method for evaluating tea plant nitrogen absorption type based on nitrogen transporter activity according to claim 1, It is characterized in that The clustering analysis method is k-means, which specifically includes: initializing classification, initializing three cluster centers, using Euclidean distance to characterize the measure of similarity between samples, calculating the distances from all samples to each initial cluster center, and dividing the sample categories according to the discrimination criterion of the minimum distance; updating the cluster center, calculating the mean of the samples belonging to each category as the new cluster center of the category, and then re-obtaining the sum of the squares of the distances from all samples to the new cluster center of their category; judging whether the cluster center changes between the previous and subsequent cluster analyses, if it no longer changes, clustering ends, otherwise repeatedly updating the cluster center until clustering ends.

3. The method for evaluating tea plant nitrogen absorption type based on nitrogen transporter activity according to claim 1, It is characterized in that The CNN sub-model specifically includes: selecting 6 input parameters related to the nitrogen absorption rate of tea trees, performing normalization processing, arranging the processed input parameter data in time series as input data, and constructing 6 input data time series input layers; first performing a convolution operation on the input data, and pooling the input data after the convolution operation, and performing a full connection operation on the output data after multiple convolution pooling operations to complete the training of the low-nitrogen and high-efficiency prediction model, the nitrogen and high-efficiency prediction model, and the high-nitrogen and high-efficiency prediction model.

4. The method for evaluating tea plant nitrogen absorption type based on nitrogen transporter activity according to claim 3, It is characterized in that The input parameters specifically include ammonium nitrogen transporters: CsAMT1.1, CsAMT1.2; nitrate nitrogen transporters: CsNRT2.1, CsNRT2.2, CsNRT2.3, CsNRT2.

4.

5. The method for evaluating tea plant nitrogen absorption type based on nitrogen transporter activity according to claim 3, It is characterized in that The training of the low-nitrogen high-efficiency prediction model, the nitrogen high-efficiency prediction model, and the high-nitrogen high-efficiency prediction model specifically includes: performing convolution operations in different directions on the input data in the input layer, extracting features between the input data by longitudinal convolution, and extracting data features of the same input data at different times by transverse convolution, and performing pooling operations after each convolution calculation to reduce operation parameters.

6. The method for evaluating tea plant nitrogen absorption type based on nitrogen transporter activity according to claim 5, It is characterized in that The convolution operation performed on the input data in different directions specifically includes: S1: Use s convolution kernels to perform vertical convolution on the input data in the input layer. The vertical convolution is calculated using the following formula: a i,j =f(w m x i,j +z m ),m=1,2,...,s; Among them, x i,j is the element in row i and column j of the input data, w m represents the convolution kernel weight, z m is the bias term of the convolution kernel, a i,j is the i-th row and j-th column element of the convolution data, and f is the activation function; S2: The input data is convolved by s convolution kernels and activated by the relu function. After activation, s neurons are output, each of which contains a data matrix. S3: Average pooling of s neurons. The specific formula is as follows: Among them, q represents the size of the pooling area, D and F are the length and width of a single neuron data matrix. Since the pooling kernel is vertically pooled, only the length of the neuron matrix will decrease, p i / q,j Represents the i / qth row and jth column element of the output neuron matrix of the pooling layer; S4: The neurons output by the pooling layer are then horizontally convolved by g convolution kernels and average pooled again, outputting g neurons; S5: The neurons output after two convolution poolings are used as the input of the connection layer. The connection layer integrates the feature information represented by the elements of all neuron matrices into the neurons of the connection layer. Assume that the number of neurons in the connection layer is T. After multiple convolution poolings, each neuron output contains k rows and l columns of elements. Each neuron in the connection layer is convolved by g [k, l] convolution kernels on the elements in each neuron matrix. The specific formula is as follows: Among them, x k,l is the kth row and lth column element of the input data, w n is the convolution kernel weight, z n is the bias term of the convolution kernel, y u is the value in each neuron of the fully connected layer.

7. The method for evaluating tea plant nitrogen absorption type based on nitrogen transporter activity according to claim 6, It is characterized in that The output layer in S5 is calculated by the linear weighted sum of the output vectors of the connection layer. The specific formula is as follows: Among them, the number of input neurons is T, and the output is The output is the nitrogen absorption rate result of tea tree.

8. The method for evaluating tea plant nitrogen absorption type based on nitrogen transporter activity according to claim 3, It is characterized in that The calculation steps of the fusion coefficient specifically include: Divide the weight value corresponding to each input parameter in each trained prediction model into different value ranges according to their size; Calculate the weight value probability corresponding to each value range, and calculate the weight entropy corresponding to each input parameter in each prediction model according to the weight value probability; A fusion coefficient corresponding to the input parameter is calculated according to the weight entropy.

9. The method for evaluating tea plant nitrogen absorption type based on nitrogen transporter activity according to claim 8, It is characterized in that Determining the fusion weight of each input parameter according to the fusion coefficient specifically includes: taking the same input parameter in all prediction models as the target, weightedly merging the weight entropy of the input parameters through the fusion coefficient, and obtaining the fusion weight of the target parameter in the comprehensive prediction model.

10. A tea tree nitrogen absorption grade assessment system based on nitrogen transporter activity, using a tea tree nitrogen absorption type assessment method based on nitrogen transporter activity according to any one of claims 1 to 9, It is characterized in that include: A training set establishment module collects the activities of various nitrogen transport proteins and nitrogen absorption rates of tea trees to establish a total data set, divides the data in the total data set into types through cluster analysis and establishes corresponding training sets to obtain a low-nitrogen high-efficiency set, a nitrogen high-efficiency set, and a high-nitrogen high-efficiency set; The sub-model training module uses the low-nitrogen high-efficiency set, the nitrogen high-efficiency set, and the high-nitrogen high-efficiency set to train the CNN sub-models of the same structure respectively, takes the activities of the nitrogen transport proteins of the tea tree as the input data, and the corresponding nitrogen absorption rate of the tea tree as the output data, to obtain the low-nitrogen high-efficiency prediction model, the nitrogen high-efficiency prediction model, and the high-nitrogen high-efficiency prediction model; A prediction model construction module calculates the fusion coefficient of the input parameters of each model in the low-nitrogen high-efficiency prediction model, the nitrogen high-efficiency prediction model, and the high-nitrogen high-efficiency prediction model; and determines the fusion weight of each input parameter according to the fusion coefficient, and constructs a comprehensive prediction model based on the fusion weights of all input parameters; The type classification module calculates the nitrogen absorption rate of the tea trees to be identified through the comprehensive prediction model, obtains the nitrogen absorption rate to be tested, and classifies the tea trees to be identified according to the cluster analysis method.

Citation Information

Patent Citations

  • Tea tree NRT1 gene, protein and gene expression method

    CN105385696A

  • Early non-destructive identification method for nitrogen absorption capacity of tea tree

    CN109541133A

  • Intelligent leaf chlorophyll content detection method based on multi-output ordinal regression model

    CN117726612A