A method for monitoring the asexual reproduction and growth of *Poa annua* rhizomes based on clustering.

Abstract

The present invention discloses a method for monitoring the asexual reproduction and growth of rhizomes of bluegrass based on clustering processing, and the method includes multi-source monitoring data acquisition, multi-source data preprocessing, construction of a growth status monitoring clustering model, clustering structure optimization and hierarchical clustering growth status monitoring. The present invention relates to the technical field of plant reproduction data processing, and specifically refers to a method for monitoring the asexual reproduction and growth of rhizomes of bluegrass based on clustering processing. This solution innovatively proposes a dual clustering mechanism based on a master-sub hierarchical structure, which significantly improves the accuracy, stability and agricultural usability of growth status monitoring; designs a misclassification point recognition mechanism, a misclassification trigger mechanism and introduces a cluster adjustment factor to enhance the adaptability and expression ability of complex plant growth monitoring behavior patterns; adopts a mirror inversion mechanism and a sine-cosine function method to improve the optimization algorithm, the optimal parameter combination of the growth status monitoring clustering model, optimizes the clustering structure, and improves the accuracy of the growth status monitoring results.

Description

Technical Field

[0001] This invention relates to the field of plant reproduction data processing technology, specifically to a method for monitoring the asexual reproduction and growth of Kentucky bluegrass rhizomes based on clustering. Background Technology

[0002] The clustering-based method for monitoring the asexual reproduction and growth of Kentucky bluegrass rhizomes involves collecting multi-source data on the growth and reproduction of Kentucky bluegrass rhizomes and using a clustering algorithm to classify the growth status of the rhizomes, thereby enabling automatic monitoring and dynamic identification of their growth trends, distribution patterns, and population structure during asexual reproduction.

[0003] However, traditional methods for monitoring the growth of plants through asexual reproduction of rhizomes suffer from several technical problems. These include the use of a single-layer clustering strategy, which fails to simultaneously characterize the complex relationship between plant physiological states and external inducing factors, resulting in insufficient expression of multidimensional plant growth patterns and inaccurate classification of growth monitoring states. Existing clustering algorithms for plant growth status monitoring also suffer from issues such as ambiguous cluster boundaries, inaccurate misclassification, delayed cluster structure response, and unstable model updates, leading to inaccurate growth status identification results. Furthermore, existing clustering algorithms for growth status monitoring often have unreasonable cluster structure parameter settings, resulting in unstable cluster identification results and insufficient clustering accuracy. Summary of the Invention

[0004] To address the above issues and overcome the shortcomings of existing technologies, this invention provides a clustering-based method for monitoring the asexual propagation growth of *Poa annua*. Addressing the limitations of traditional methods that rely solely on single-layer clustering, which fails to simultaneously characterize the complex relationship between plant physiological states and external stimuli, resulting in insufficient expression of multidimensional growth patterns and inaccurate classification of growth monitoring states, this invention innovatively proposes a dual-clustering mechanism based on a hierarchical principal-sub-state structure. This mechanism performs two-stage clustering on both plant physiological data and environmental stimuli data, constructing a dual-layer labeling system of principal and sub-states. This effectively distinguishes the subtle growth response patterns of individual plants under similar physiological states caused by external changes, offering higher interpretability and labeling accuracy. It achieves structured hierarchical analysis of multi-source data during plant asexual propagation, significantly improving the accuracy, stability, and agricultural usability of growth status monitoring. Furthermore, addressing the technical problems of existing clustering algorithms for plant growth status monitoring, such as ambiguous cluster boundaries, inaccurate misclassification, delayed cluster structure response, and unstable model updates, leading to inaccurate growth status identification results, this invention innovatively introduces a cluster radius-based... This invention employs a misclassification point identification mechanism, constructs a reasonable misclassification judgment boundary, and improves the accuracy and rationality of misclassification identification. It designs a cross-period cumulative misclassification trigger mechanism, introduces historical misclassification point statistics and a Xie-Beni index-driven structure adaptive adjustment strategy, and enhances the model's responsiveness to complex time-varying data structures. It also introduces a cluster center incremental update method driven by a cluster adjustment factor, effectively improving the numerical stability and dynamic adaptability of the model update process. Through these improvements, the growth state monitoring clustering model constructed in this invention possesses strong dynamic adaptability, structural adjustment flexibility, and state identification accuracy, significantly improving the accuracy and temporal stability of growth state classification results, and enhancing its adaptability and expressive ability to complex plant growth monitoring behavior patterns. Addressing the technical problem in existing clustering algorithms suitable for growth state monitoring where unreasonable clustering structure parameter settings lead to unstable clustering identification results and insufficient clustering accuracy, this solution uses a mirror inversion mechanism and sine / cosine function methods to improve and optimize the algorithm. It globally optimizes the key parameters in the clustering model to obtain the optimal parameter combination, thereby effectively improving the structural expressive ability and classification accuracy of the clustering model, and enhancing the accuracy and applicability of the growth state monitoring results.

[0005] The technical solution adopted in this invention is as follows: The method for monitoring the asexual propagation and growth of Kentucky bluegrass rhizomes based on clustering processing provided by this invention includes the following steps:

[0006] Step S1: Acquisition of multi-source monitoring data;

[0007] Step S2: Multi-source data preprocessing;

[0008] Step S3: Construct a clustering model for monitoring growth status;

[0009] Step S4: Cluster structure optimization;

[0010] Step S5: Monitoring the growth status of hierarchical clustering.

[0011] Further, in step S1, the acquisition of multi-source monitoring data specifically involves collecting data information from the rhizome growth area of ​​Kentucky bluegrass to obtain raw data for asexual reproduction growth monitoring; the raw data for asexual reproduction growth monitoring includes historical growth monitoring data and real-time growth monitoring data, both of which include plant physiological data, plant environmental data, and operational behavior data for each monitoring point; the historical growth monitoring data also includes the growth monitoring status.

[0012] Further, in step S2, the multi-source data preprocessing specifically involves data optimization processing to obtain optimized asexual reproduction growth monitoring data; including the following steps:

[0013] Step S21: Data cleaning process, specifically, handling missing values, outliers, and duplicate data in the original data;

[0014] Step S22: Data normalization processing, specifically, the data is normalized using the Min-Max normalization method;

[0015] Step S23: Time series sample construction, specifically, the continuous monitoring data of each monitoring point is sliced ​​according to the set 24-hour time window to construct sample data with consistent time, and a sliding window mechanism is used to set the time overlap interval to 4 hours to generate multiple time series samples with overlapping relationships.

[0016] Step S24: Reference label association, specifically, the growth monitoring status in the historical growth monitoring data is used as the reference label for the main clustering operation, which is only used for cluster label selection and does not participate in the clustering data processing process.

[0017] Further, in step S3, the construction of the growth status monitoring clustering model specifically involves establishing a clustering model through a misclassification point identification mechanism, a misclassification triggering mechanism, and an improved fuzzy C-means clustering algorithm by introducing a cluster adjustment factor, thereby obtaining the growth status monitoring clustering model, including the following steps:

[0018] Step S31: Initialize clustering, specifically by using the initial number of clusters c and fuzzy index m to perform the standard fuzzy C-means clustering algorithm on the sample data in the historical growth monitoring data to obtain the set of cluster centers and the membership matrix of each data point to each cluster;

[0019] Step S32: Calculate the distance between cluster centers, specifically for any two cluster centers in the cluster center set. and The distance between the two is calculated using Euclidean distance;

[0020] Step S33: Calculate the distance from each data point to the cluster center. Specifically, for each data point... The distance from each cluster center is calculated using Euclidean distance.

[0021] Step S34: Misclassification point detection, specifically by calculating the minimum distance from each data point to all cluster centers and comparing it with the cluster radius. The data points are compared to determine whether they are misclassified. If the minimum distance of a data point from all cluster centers is greater than the cluster radius, then the data point is considered misclassified. The formula used is as follows:

[0022] ;

[0023] ;

[0024] In the formula, Represents the misclassification function. This represents the number of clusters in the t-th period. This represents the j-th cluster, where j represents the cluster index. Represents any data point in the j-th cluster. express arrive European distance, express arrive European distance, Indicates the current period data point Membership degree of cluster j This represents the radius of the j-th cluster. This represents the threshold parameter for membership values. This represents the centroid of the current periodic cluster j. This represents the i-th sample data point. This represents the data in the t-th period;

[0025] Step S35: Adjust the clustering structure, specifically, if the proportion of misclassified points accumulated up to the current period t to the total amount of data in the current period exceeds the misclassification proportion threshold parameter. Then, based on the Xie-Beni index as the structural evaluation index, the structural strength of each cluster number scheme within the candidate cluster number range is evaluated, and the optimal cluster number is selected. And re-execute the standard fuzzy C-means clustering algorithm based on the updated cluster number; otherwise, update the current cluster center.

[0026] Step S36: Update the current cluster center, specifically by calculating the temporary cluster center for the current period. and calculation cluster adjustment factor Update the current cluster center using the following formula:

[0027] ;

[0028] ;

[0029] ;

[0030] In the formula, Indicates the data points in the current period The membership value of cluster j, Indicates the first Data from each period, The table represents the center of cluster j in the next cycle. Indicates the data points in the previous period The membership degree value of cluster j;

[0031] Step S37: Cluster deletion mechanism, specifically, for each cluster existing within the current period t, by analyzing its activity changes over T consecutive periods, it is determined whether the cluster is in an inactive state, thereby determining whether it is an outdated cluster. If cluster j is considered obsolete, a cluster deletion operation needs to be performed, and the cluster deletion constraint conditions must be met during the deletion operation; where, This represents the fuzzy cardinality of the current periodic cluster j, which is the sum of the membership degrees of all data points in that cluster. Indicates the period The fuzzy cardinality of cluster j at time t, where T represents the number of cycles for cluster activity assessment;

[0032] Step S38: Obtain the final clustering result, specifically if the change in all cluster centers is lower than the preset convergence threshold or the current iteration number has reached the preset maximum number of iterations. If the clustering iteration is terminated, the final clustering result is output. Specifically, the clustering result is that all sample data points are assigned to the corresponding clusters according to their final maximum membership degree, forming multiple cluster sets. Each cluster represents a different growth state under the asexual reproduction of Kentucky bluegrass rhizomes.

[0033] Furthermore, in step S4, the clustering structure optimization specifically includes the following steps:

[0034] Step S41: Construct the search individual location vector, specifically by encoding the growth state monitoring clustering model parameters into the search individual location vector; the growth state monitoring clustering model parameters include the fuzzy index m and the membership threshold. Compared to the misclassification threshold ;

[0035] Step S42: Design the fitness function, specifically by establishing the fitness function through a weighted combination of the average silhouette coefficient of the current cluster structure and the label hopping rate;

[0036] Step S43: Obtain the optimal parameter combination of the model. Specifically, the optimal parameter combination of the growth state monitoring clustering model is obtained by globally optimizing the parameters of the growth state monitoring clustering model through an improved optimization algorithm.

[0037] Step S431: Initialize the search population individuals. Specifically, generate the current positions of N search individuals through random initialization, then optimize the individual positions using a mirror inversion mechanism to obtain the reflected positions of the individuals, and calculate the fitness value F of the individuals in the population using the fitness function. i Individuals are sorted from best to worst based on their fitness values. The top N individuals with the highest fitness values ​​are selected based on their current position and reflected position to form a new search population. The formula used is as follows:

[0038] ;

[0039] In the formula, This represents the reflection position of the i-th individual in the d-th dimension. This represents the current position of the i-th individual in the d-th dimension. and Let represent the minimum and maximum values ​​of the search space in dimension d, respectively. This represents the scaling factor, which is a constant.

[0040] Step S432: Update the individual search position, specifically by updating the individual search position using sine and cosine functions and a non-linear decreasing search. The formula used is as follows:

[0041] ;

[0042] In the formula, Indicates that the i-th individual is in the first place. The position of the d-th dimension in the population. Indicates that the i-th individual is in the first place. The position of the d-th dimension in the population. This indicates the position of the optimal individual in the population. Indicates the number of iterations. Indicates the maximum number of iterations. Indicates the search adjustment factor, and , Indicates the warning value. Indicates the safety threshold. Represents [0,2] Random numbers within the range of ] This represents a random number in the range [0,1], and e represents the base of the natural logarithm.

[0043] Step S433: Search complete, specifically when the fitness value F of the searched individual is... i When the fitness threshold is exceeded and the maximum number of iterations is reached, the search is terminated and the global optimal position of the search individual is obtained. The global optimal position of the search individual specifically refers to the optimal parameter combination of the growth state monitoring clustering model.

[0044] Step S44: Clustering parameter optimization, specifically, adjusting the parameters of the growth state monitoring clustering model according to the optimal parameter combination of the growth state monitoring clustering model to obtain the optimal growth state monitoring clustering model.

[0045] Further, in step S5, the hierarchical clustering growth status monitoring specifically involves identifying multiple growth states of Kentucky bluegrass through a principal clustering structure, forming a two-layer labeling system, and obtaining growth status monitoring results; including the following steps:

[0046] Step S51: The main state identification stage, specifically, takes the plant physiological data in the real-time growth monitoring data as input data and inputs it into the optimal growth state monitoring clustering model, performs the first stage main clustering operation, generates multiple main clusters, and in each cluster, according to the associated historical growth monitoring state labels, counts the frequency of the labels, selects the label with the highest frequency as the main state label of the cluster, and obtains the main state identification result.

[0047] Step S52: Sub-state identification stage, used to analyze the subdivided states constituted by external factors behind the main state, and identify subdivided growth response patterns; specifically, it includes the following steps:

[0048] Step S521: Obtaining sub-clusters. Specifically, based on the plant physiological data of each main cluster, select the corresponding plant environmental data and operational behavior data as the input data for the second stage of clustering, input them into the optimal growth state monitoring clustering model, perform the second stage sub-clustering operation, and generate multiple sub-clusters.

[0049] Step S522: Output the sub-state recognition result. Specifically, perform statistical processing on the data of each sub-cluster, calculate the mean, standard deviation and maximum value of the data to form the feature representation vector of the cluster. Then, construct a set of predefined sub-state labels. Each sub-state label consists of multiple feature parameter ranges. Compare the cosine similarity of the feature representation vector of each sub-cluster with all sub-state labels, and select the label with the highest similarity as the sub-state label of the cluster to obtain the sub-state recognition result.

[0050] Step S53: Output the growth status monitoring results. Specifically, based on the main state identification results and the sub-state identification results, the samples within the cluster are combined according to their correspondence to output the double-layer growth status corresponding to each monitoring point, thus obtaining the growth status monitoring results.

[0051] The beneficial effects achieved by the present invention using the above solution are as follows:

[0052] (1) In view of the technical problem that traditional rhizome asexual reproduction growth monitoring methods only use a single-layer clustering strategy, which makes it difficult to simultaneously characterize the complex relationship between plant physiological state and external inducement, resulting in insufficient expression of plant multidimensional growth patterns and inaccurate classification of growth monitoring status, this solution innovatively proposes a dual clustering mechanism based on a master-sub-layer hierarchical structure. Two-stage clustering is performed on plant physiological data and environmental manipulation inducement data respectively to construct a dual-layer label system of master state and sub-state. This can effectively distinguish the subdivided growth response patterns of individual plants under similar physiological states caused by external changes, and has higher interpretability of growth monitoring status and label expression accuracy. It realizes the structured hierarchical analysis of multi-source data in the process of plant asexual reproduction, and significantly improves the accuracy, stability and agricultural usability of growth status monitoring.

[0053] (2) To address the technical problems of existing clustering algorithms for monitoring plant growth status, such as ambiguous cluster boundaries, inaccurate misclassification, delayed cluster structure response, and unstable model updates, which lead to inaccurate growth status identification results, this solution innovatively introduces a misclassification point identification mechanism based on cluster radius to construct a reasonable misclassification judgment boundary and improve the accuracy and rationality of misclassification identification; it designs a cross-period cumulative misclassification triggering mechanism and introduces a structure adaptive adjustment strategy driven by historical misclassification point statistics and Xie-Beni index to enhance the model's responsiveness to complex time-varying data structures; it introduces a cluster center incremental update method driven by cluster adjustment factor to effectively improve the numerical stability and dynamic adaptability of the model update process; through the above improvements, the growth status monitoring clustering model constructed by this invention has strong dynamic adaptability, structural adjustment flexibility, and status identification accuracy, which can significantly improve the accuracy and temporal stability of growth status classification results and enhance the adaptability and expressive ability to complex plant growth monitoring behavior patterns.

[0054] (3) In view of the technical problem that the clustering structure parameters in the existing clustering algorithms applicable to growth status monitoring are not set reasonably, resulting in unstable clustering identification results and insufficient clustering accuracy, this scheme adopts the mirror inversion mechanism and sine and cosine function method to improve the optimization algorithm, and performs global optimization on the key parameters in the clustering model to obtain the optimal parameter combination, thereby effectively improving the structural expression ability and classification accuracy of the clustering model, and enhancing the accuracy and applicability of growth status monitoring results. Attached Figure Description

[0055] Figure 1 This is a flowchart illustrating the method for monitoring the asexual propagation and growth of Kentucky bluegrass rhizomes based on clustering processing provided by the present invention.

[0056] Figure 2 This is a flowchart illustrating step S2;

[0057] Figure 3 This is a flowchart illustrating step S3;

[0058] Figure 4 This is a flowchart illustrating step S4;

[0059] Figure 5 This is a flowchart illustrating step S43;

[0060] Figure 6 This is a flowchart illustrating step S5;

[0061] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation

[0062] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0063] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the system or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0064] Example 1, see Figure 1 The technical solution adopted by this invention is as follows: The method for monitoring the asexual propagation and growth of Kentucky bluegrass rhizomes based on clustering processing provided by this invention includes the following steps:

[0065] Step S1: Acquisition of multi-source monitoring data, specifically by collecting data information to obtain raw data for monitoring asexual reproduction and growth;

[0066] Step S2: Multi-source data preprocessing, specifically, data cleaning, data normalization, time-series sample construction, and reference label association are performed to obtain optimized data for asexual reproduction growth monitoring;

[0067] Step S3: Construct a growth state monitoring clustering model to build a dynamic and adaptive growth state clustering model, thereby achieving effective classification of multiple growth states; specifically, a clustering model is established by using a misclassification point identification mechanism, a misclassification triggering mechanism, and introducing a cluster adjustment factor to improve the fuzzy C-means clustering algorithm, thus obtaining the growth state monitoring clustering model;

[0068] Step S4: Cluster structure optimization, specifically, based on the improved optimization algorithm, a search vector is constructed for the key model parameters. The multi-objective fitness function, which combines the average silhouette coefficient and the label jump rate, is used to complete the initialization of the search population, the position update and the global search. Finally, the optimal combination of model parameters is obtained and used to update the clustering model structure to obtain the optimal growth state monitoring clustering model.

[0069] Step S5: Hierarchical clustering for growth status monitoring is used to perform structured clustering analysis on the optimized data of asexual reproduction growth monitoring to identify different types of growth status. Specifically, a two-layer clustering structure combining main state identification and sub-state identification is constructed. First, the main clustering is performed with plant physiological data as input to obtain the main state identification result. Then, sub-clustering is performed based on the environmental and operational data of the main cluster, and the label is assigned through template matching to generate the sub-state identification result. Finally, the main state and sub-state identification results are jointly output to obtain the growth status monitoring result.

[0070] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S1, the acquisition of multi-source monitoring data specifically involves collecting data information from the rhizome growth area of ​​*Poa annua* by deploying a sensor system and a management operation recording system to obtain raw data for asexual reproduction growth monitoring. The raw data for asexual reproduction growth monitoring includes historical growth monitoring data and real-time growth monitoring data. Both the historical and real-time growth monitoring data include plant physiological data, plant environmental data, and operational behavior data for each monitoring point. The historical growth monitoring data also includes the growth monitoring status.

[0071] The monitoring points are determined by dividing the target planting area into multiple 1m×1m spatial sub-areas, and combining plant distribution density, soil environmental heterogeneity, and management operation zoning settings.

[0072] The plant physiological data include root and stem volume, root resistance value, and root zone. Concentration, root zone Concentration, root zone conductivity, and root temperature;

[0073] The plant environmental data includes soil temperature, soil moisture, soil oxygen content, soil pH, and soil electrical conductivity.

[0074] The operational data includes fertilizer type, fertilizer dosage, irrigation frequency, and irrigation water volume;

[0075] The growth monitoring states include active growth state, stagnant state, degenerate state, abnormal expansion state, and steady state.

[0076] Example 3, see Figure 1 and Figure 2 This embodiment is based on the above embodiment. The multi-source data preprocessing is used to preprocess the raw data of asexual reproduction growth monitoring. Specifically, it involves data cleaning, data normalization, time-series sample construction, and reference label association to obtain optimized asexual reproduction growth monitoring data. The steps include:

[0077] Step S21: Data cleaning process, used to improve data quality and validity, specifically to handle missing values, outliers and duplicate data in the original data;

[0078] The missing value processing specifically involves using time series linear interpolation to fill in missing values ​​in continuous sampling by interpolating between previous and subsequent time points;

[0079] The outlier handling specifically involves automatically identifying upper and lower bounds based on interquartile range using a box plot method to remove outliers.

[0080] The deduplication process specifically involves using a hash check deduplication method to quickly filter out duplicate datasets.

[0081] Step S22: Data normalization processing, which is used to standardize the original data with different dimensions and numerical ranges, specifically by using the Min-Max normalization method to standardize the data;

[0082] Step S23: Time series sample construction, which is used to transform continuous raw monitoring data into samples with a uniform time scale and structured format. Specifically, the continuous monitoring data of each monitoring point is sliced ​​according to a set 24-hour time window to construct time-consistent sample data. A sliding window mechanism is used to set the time overlap interval to 4 hours to generate multiple time series samples with overlapping relationships.

[0083] Step S24: Reference label association, used to provide semantic mapping support for clustering results. Specifically, the growth monitoring status in the historical growth monitoring data is used as the reference label for the main clustering operation. It is only used for cluster label selection and does not participate in the clustering data processing process.

[0084] Example 4, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S3, the construction of the growth status monitoring clustering model is used to construct a dynamic clustering model based on multi-source monitoring data during the asexual reproduction process of Kentucky bluegrass rhizomes, so as to achieve effective classification of different growth states. Specifically, the clustering model is established by using a misclassification point identification mechanism, a misclassification triggering mechanism, and introducing a cluster adjustment factor to improve the fuzzy C-means clustering algorithm, thereby obtaining the growth status monitoring clustering model, including the following steps:

[0085] Step S31: Initialize clustering, specifically by using the initial number of clusters c and fuzzy index m to perform the standard fuzzy C-means clustering algorithm on the sample data in the historical growth monitoring data to obtain the set of cluster centers and the membership matrix of each data point to each cluster;

[0086] Step S32: Calculate the distance between cluster centers, specifically for any two cluster centers in the cluster center set. and The distance between the two is calculated using Euclidean distance; the formula used is as follows:

[0087] ;

[0088] In the formula, This represents the function for calculating Euclidean distance. This represents the centroid of the current periodic cluster j, and also the center of cluster j; This represents the centroid of the current periodic cluster k, and also the center of cluster k; Represents the Euclidean norm;

[0089] Step S33: Calculate the distance from each data point to the cluster center. Specifically, for each data point... The distance from each cluster center is calculated using Euclidean distance; the formula used is as follows:

[0090] ;

[0091] In the formula, This represents the i-th sample data point. This represents the data in the t-th period;

[0092] Step S34: Misclassification point detection, specifically by calculating the minimum distance from each data point to all cluster centers and comparing it with the cluster radius. The data points are compared to determine whether they are misclassified. If the minimum distance of a data point from all cluster centers is greater than the cluster radius, then the data point is considered misclassified. The formula used is as follows:

[0093] ;

[0094] ;

[0095] In the formula, Represents the misclassification function, if Represents sample data points It is a misclassified point; otherwise, it represents a sample data point. Correctly classified; This represents the number of clusters in the t-th period. This represents the j-th cluster, where j represents the cluster index. Represents any data point in the j-th cluster. express arrive European distance, express arrive European distance, Indicates the current period data point Membership degree of cluster j This represents the radius of the j-th cluster. This represents the threshold parameter for membership values, and its range. ;

[0096] Step S35: Adjust the clustering structure, specifically, if the proportion of misclassified points accumulated up to the current period t to the total amount of data in the current period exceeds the misclassification proportion threshold parameter. Then, based on the Xie-Beni index as the structural evaluation index, the structural strength of each cluster number scheme within the candidate cluster number range is evaluated, and the optimal cluster number is selected. The standard fuzzy C-means clustering algorithm is re-executed based on the updated cluster number; otherwise, the current cluster centers are updated. The formula used is as follows:

[0097] ;

[0098] In the formula, This represents the proportion of misclassified points accumulated up to the current period t to the total amount of data in the current period. Indicates the optimal number of clusters in the structure. This indicates the maximum number of clusters set, with a value of 5. The Xie-Beni index is a function used to evaluate the effectiveness of clustering results, measuring the compactness within clusters and the separation between clusters. This represents the number of clusters selected from the set of candidate clusters that minimizes the evaluation index.

[0099] Step S36: Update the current cluster center, specifically by calculating the temporary cluster center for the current period. and calculation cluster adjustment factor Update the current cluster center using the following formula:

[0100] ;

[0101] ;

[0102] ;

[0103] In the formula, Indicates the data points in the current period The membership value of cluster j, Indicates the first Data from each period, The table represents the center of cluster j in the next cycle. Indicates the data points in the previous period The membership degree value of cluster j;

[0104] Step S37: Cluster deletion mechanism, specifically, for each cluster existing within the current period t, by analyzing its activity changes over T consecutive periods, it is determined whether the cluster is in an inactive state, thereby determining whether it is an outdated cluster. If cluster j is deemed obsolete, a cluster deletion operation is required. This deletion operation must satisfy cluster deletion constraints, including a minimum cluster size constraint and a priority deletion constraint. The minimum cluster size constraint requires that after deletion, the remaining cluster size must be at least two. The priority deletion constraint requires that if multiple clusters simultaneously meet the deletion conditions, the cluster with the smallest change in fuzzy cardinality is deleted first. This represents the fuzzy cardinality of the current periodic cluster j, which is the sum of the membership degrees of all data points in that cluster. Indicates the period The fuzzy cardinality of cluster j at time t, where T represents the number of cycles for cluster activity assessment;

[0105] Step S38: Obtain the final clustering result, specifically if the change in all cluster centers is lower than the preset convergence threshold or the current iteration number has reached the preset maximum number of iterations. If the clustering iteration is terminated, the final clustering result is output. Specifically, the clustering result is that all sample data points are assigned to the corresponding clusters according to their final maximum membership degree, forming multiple cluster sets. Each cluster represents a different growth state under the asexual reproduction of Kentucky bluegrass rhizomes.

[0106] By performing the above operations, this solution addresses the technical problems in existing clustering algorithms for plant growth status monitoring, such as ambiguous cluster boundaries, inaccurate misclassification, lagging cluster structure response, and unstable model updates, which lead to inaccurate growth status identification results. This solution innovatively introduces a misclassification point identification mechanism based on cluster radius to construct a reasonable misclassification judgment boundary, improving the accuracy and rationality of misclassification identification. It also designs a cross-period cumulative misclassification triggering mechanism, introducing historical misclassification point statistics and a Xie-Beni index-driven structure adaptive adjustment strategy to enhance the model's responsiveness to complex time-varying data structures. Furthermore, it introduces a cluster center incremental update method driven by a cluster adjustment factor, effectively improving the numerical stability and dynamic adaptability of the model update process. Through these improvements, the growth status monitoring clustering model constructed in this invention possesses strong dynamic adaptability, structural adjustment flexibility, and state identification accuracy, significantly improving the accuracy and temporal stability of growth status classification results, and enhancing its adaptability and expressive ability to complex plant growth monitoring behavior patterns.

[0107] Example 5, see Figure 1 and Figure 4 This embodiment is based on the above embodiment. In step S4, the clustering structure optimization uses an improved intelligent optimization algorithm to perform global search and adaptive adjustment of key parameters in the growth state monitoring clustering model, thereby improving the structural performance and temporal stability of the clustering model in the identification of complex plant growth behaviors. Specifically, it includes the following steps:

[0108] Step S41: Construct the search individual location vector, specifically by encoding the growth state monitoring clustering model parameters into the search individual location vector; the growth state monitoring clustering model parameters include the fuzzy index m and the membership threshold. Compared to the misclassification threshold ;

[0109] Step S42: Design a fitness function to construct a multi-objective fitness evaluation function required for cluster structure optimization. This function comprehensively measures the performance of the current cluster structure in terms of both spatial compactness and temporal stability. Specifically, the fitness function is established by a weighted combination of the average silhouette coefficient and the label hopping rate of the current cluster structure. The formula used is as follows:

[0110] ;

[0111] In the formula, This represents the position of the i-th searched individual. Represents the fitness function. This represents the average silhouette coefficient calculated from the current clustering results, used to evaluate the geometric compactness of the clusters. This represents the label jump rate of the corresponding clustering result in the time series, used to evaluate the stability of clustering in the time series dimension. The weighting factor representing the cluster compactness index, with a range of values. , The weighting factor represents the time series stability index. ;

[0112] Step S43: Obtain the optimal parameter combination of the model. Specifically, the optimal parameter combination of the growth state monitoring clustering model is obtained by globally optimizing the parameters of the growth state monitoring clustering model through an improved optimization algorithm.

[0113] Step S431: Initialize the search population individuals. Specifically, generate the current positions of N search individuals through random initialization, then optimize the individual positions using a mirror inversion mechanism to obtain the reflected positions of the individuals, and calculate the fitness value F of the individuals in the population using the fitness function. i Individuals are sorted from best to worst based on their fitness values. The top N individuals with the highest fitness values ​​are selected based on their current position and reflected position to form a new search population. The formula used is as follows:

[0114] ;

[0115] In the formula, This represents the reflection position of the i-th individual in the d-th dimension. This represents the current position of the i-th individual in the d-th dimension. and Let represent the minimum and maximum values ​​of the search space in dimension d, respectively. This represents the scaling factor, which is a constant.

[0116] Step S432: Update the individual search position, specifically by updating the individual search position using sine and cosine functions and a non-linear decreasing search. The formula used is as follows:

[0117] ;

[0118] In the formula, Indicates that the i-th individual is in the first place. The position of the d-th dimension in the population. Indicates that the i-th individual is in the first place. The position of the d-th dimension in the population. This indicates the position of the optimal individual in the population. Indicates the number of iterations. Indicates the maximum number of iterations. Indicates the search adjustment factor, and , Indicates the warning value. Indicates the safety threshold. Represents [0,2] Random numbers within the range of ] This represents a random number in the range [0,1], and e represents the base of the natural logarithm.

[0119] Step S433: Search complete, specifically when the fitness value F of the searched individual is... i When the fitness threshold is exceeded and the maximum number of iterations is reached, the search is terminated and the global optimal position of the search individual is obtained. The global optimal position of the search individual specifically refers to the optimal parameter combination of the growth state monitoring clustering model.

[0120] Step S44: Clustering parameter optimization, specifically, adjusting the parameters of the growth state monitoring clustering model according to the optimal parameter combination of the growth state monitoring clustering model to obtain the optimal growth state monitoring clustering model.

[0121] By performing the above operations, this solution addresses the technical problem in existing clustering algorithms suitable for growth state monitoring that suffer from unreasonable clustering structure parameter settings, leading to unstable clustering identification results and insufficient clustering accuracy. It employs a mirror inversion mechanism and sine / cosine function methods to improve and optimize the algorithm. This approach globally optimizes key parameters in the clustering model, obtaining the optimal parameter combination. This effectively enhances the structural expressiveness and classification accuracy of the clustering model, thereby improving the accuracy and applicability of growth state monitoring results.

[0122] Example 6, see Figure 1 and Figure 6 This embodiment is based on the above embodiment. In step S5, the hierarchical clustering growth status monitoring is used to perform structured clustering analysis on multi-source monitoring data collected during the asexual reproduction of Kentucky bluegrass rhizomes, and to identify the various growth states of individual plants under different physiological and environmental conditions. Specifically, it identifies the various growth states of Kentucky bluegrass through the master clustering structure, forms a two-layer labeling system, and obtains the growth status monitoring results; including the following steps:

[0123] Step S51: Main state identification stage, used to identify the core growth state presented by the physiological behavior of Kentucky bluegrass. Specifically, plant physiological data in real-time growth monitoring data is used as input data and input into the optimal growth state monitoring clustering model. The first stage of main clustering operation is performed to generate multiple main clusters. In each cluster, according to the associated historical growth monitoring state labels, the frequency of label occurrence is counted, and the label with the highest frequency is selected as the main state label of the cluster to obtain the main state identification result.

[0124] Step S52: Sub-state identification stage, used to analyze the subdivided states constituted by external factors behind the main state, and identify subdivided growth response patterns; specifically, it includes the following steps:

[0125] Step S521: Obtaining sub-clusters. Specifically, based on the plant physiological data of each main cluster, select the corresponding plant environmental data and operational behavior data as the input data for the second stage of clustering, input them into the optimal growth state monitoring clustering model, perform the second stage sub-clustering operation, and generate multiple sub-clusters.

[0126] Step S522: Outputting sub-state identification results. Specifically, statistical processing is performed on the data of each sub-cluster, calculating the mean, standard deviation, and maximum value to form the feature representation vector of the cluster. Subsequently, a set of predefined sub-state labels is constructed, each sub-state label consisting of multiple feature parameter ranges. The feature representation vector of each sub-cluster is compared with all sub-state labels using cosine similarity, and the label with the highest similarity is selected as the sub-state label of the cluster, thus obtaining the sub-state identification result. The sub-state labels include moisture-sensitive sub-labels, stress-sensitive sub-labels, nutrient-deficient sub-labels, management-interference-sensitive sub-labels, and external factor-stable sub-labels. The feature parameters include soil moisture, soil temperature, root zone conductivity, CO2 concentration, fertilization frequency, and irrigation dosage. The feature parameter ranges are set based on historical statistical experience and agricultural management knowledge.

[0127] Step S53: Output the growth status monitoring results to form the final multi-level status monitoring results. Specifically, based on the main state identification results and sub-state identification results, the results are combined according to the correspondence of samples within the cluster to output the double-layer growth status corresponding to each monitoring point, thus obtaining the growth status monitoring results.

[0128] By performing the above operations, this solution addresses the technical problem in traditional rhizome asexual reproduction growth monitoring methods that rely solely on single-layer clustering strategies, making it difficult to simultaneously characterize the complex relationship between plant physiological states and external inducing factors. This results in insufficient expression of multidimensional plant growth patterns and inaccurate classification of growth monitoring states. This innovative approach proposes a dual-clustering mechanism based on a master-sub-state hierarchical structure. Two-stage clustering is performed on both plant physiological data and environmental induced factor data, constructing a dual-layer labeling system of master and sub-states. This effectively distinguishes the subdivided growth response patterns of individual plants under similar physiological states caused by external changes, providing higher interpretability and label expression accuracy for growth monitoring states. It achieves structured hierarchical analysis of multi-source data during plant asexual reproduction, significantly improving the accuracy, stability, and agricultural usability of growth status monitoring.

[0129] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0130] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

[0131] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A method for monitoring the asexual propagation and growth of Kentucky bluegrass rhizomes based on clustering, characterized in that: The method includes the following steps: Step S1: Acquisition of multi-source monitoring data, specifically by collecting data information to obtain raw data for monitoring asexual reproduction and growth; Step S2: Multi-source data preprocessing, specifically, data cleaning, data normalization, time-series sample construction, and reference label association are performed to obtain optimized data for asexual reproduction growth monitoring; Step S3: Construct a growth status monitoring clustering model. Specifically, this involves improving the fuzzy C-means clustering algorithm by introducing a misclassification point identification mechanism, a misclassification triggering mechanism, and a cluster adjustment factor to construct a clustering model for growth status monitoring. This includes the following steps: initializing clusters, calculating the distance between cluster centers, calculating the distance from data points to cluster centers, detecting misclassification points, adjusting the cluster structure, updating the current cluster centers, implementing a cluster deletion mechanism, and obtaining the final clustering results. Step S4: Cluster structure optimization, specifically, constructing search vectors for key parameters in the growth state monitoring cluster model, designing a fitness function by integrating the average profile coefficient and label jump rate, improving the optimization algorithm by using a mirror inversion mechanism and sine and cosine function methods, obtaining the optimal parameter combination of the model, and updating the cluster model structure based on the optimal parameter combination to obtain the optimal growth state monitoring cluster model; Step S5: Hierarchical clustering for growth status monitoring. Specifically, a two-layer clustering structure combining main state identification and sub-state identification is constructed. First, the main clustering is performed with plant physiological data as input to obtain the main state identification result. Then, the sub-clustering is performed based on the environmental and operational data of the main cluster to generate the sub-state identification result. Finally, the main state and sub-state identification results are jointly output to obtain the growth status monitoring result.

2. The method for monitoring the asexual propagation and growth of Kentucky bluegrass rhizomes based on clustering processing according to claim 1, characterized in that: In step S3, the construction of the growth state monitoring clustering model specifically includes the following steps: Step S31: Initialize clustering, specifically by using the initial number of clusters c and fuzzy index m to perform the standard fuzzy C-means clustering algorithm on the sample data in the historical growth monitoring data to obtain the set of cluster centers and the membership matrix of each data point to each cluster; Step S32: Calculate the distance between cluster centers, specifically for any two cluster centers in the cluster center set. and The distance between the two is calculated using Euclidean distance; Step S33: Calculate the distance from each data point to the cluster center. Specifically, for each data point... The distance from each cluster center is calculated using Euclidean distance. Step S34: Misclassification point detection, specifically by calculating the minimum distance from each data point to all cluster centers and dividing it by the radius of that cluster. The comparison is performed to determine whether a data point is a misclassified point. If the minimum distance of a data point from all cluster centers is greater than the radius of the cluster, then the data point is considered a misclassified point. The formula used is as follows: ； ； In the formula, Represents the misclassification function. This represents the number of clusters in the t-th period. This represents the j-th cluster, where j represents the cluster index. Represents any data point in the j-th cluster. express arrive European distance, express arrive European distance, Indicates the current period data point Membership degree of cluster j This represents the radius of the j-th cluster. This represents the threshold parameter for membership values. This represents the centroid of the current periodic cluster j. This represents the i-th sample data point. This represents the data in the t-th period; Step S35: Adjust the clustering structure, specifically, if the proportion of misclassified points accumulated up to the current period t to the total amount of data in the current period exceeds the misclassification proportion threshold parameter. Then, based on the Xie-Beni index as the structural evaluation index, the structural strength of each cluster number scheme within the candidate cluster number range is evaluated, and the optimal cluster number is selected. And re-execute the standard fuzzy C-means clustering algorithm based on the updated cluster number; otherwise, update the current cluster center. Step S36: Update the current cluster center, specifically by calculating the temporary cluster center for the current period. and calculation cluster adjustment factor Update the current cluster center using the following formula: ； ； ； In the formula, Indicates the data points in the current period The membership value of cluster j, Indicates the first Data from each period, Indicates the center of cluster j in the next cycle. Indicates the data points in the previous period The membership degree value of cluster j; Step S37: Cluster deletion mechanism, specifically, for each cluster existing within the current period t, by analyzing its activity changes over T consecutive periods, it is determined whether the cluster is in an inactive state, thereby determining whether it is an outdated cluster. If cluster j is considered obsolete, a cluster deletion operation needs to be performed, and the cluster deletion constraint conditions must be met during the deletion operation; where, This represents the fuzzy cardinality of the current periodic cluster j, which is the sum of the membership degrees of all data points in that cluster. Indicates the period The fuzzy cardinality of cluster j at time t, where T represents the number of cycles for cluster activity assessment; Step S38: Obtain the final clustering result, specifically if the change in all cluster centers is lower than the preset convergence threshold or the current iteration number has reached the preset maximum number of iterations. If the clustering iteration is terminated, the final clustering result is output. Specifically, the clustering result is that all sample data points are assigned to the corresponding clusters according to their final maximum membership degree, forming multiple cluster sets. Each cluster represents a different growth state under the asexual reproduction of Kentucky bluegrass rhizomes.

3. The method for monitoring the asexual propagation and growth of Kentucky bluegrass rhizomes based on clustering processing according to claim 1, characterized in that: In step S4, the clustering structure optimization specifically includes the following steps: Step S41: Construct the search individual location vector, specifically by encoding the growth state monitoring clustering model parameters into the search individual location vector; the growth state monitoring clustering model parameters include the fuzzy index m and the membership threshold. Compared to the misclassification threshold ; Step S42: Design the fitness function, specifically by establishing the fitness function through a weighted combination of the average silhouette coefficient of the current cluster structure and the label hopping rate; Step S43: Obtain the optimal parameter combination of the model. Specifically, the optimal parameter combination of the growth state monitoring clustering model is obtained by globally optimizing the parameters of the growth state monitoring clustering model through an improved optimization algorithm. Step S44: Clustering parameter optimization, specifically, adjusting the parameters of the growth state monitoring clustering model according to the optimal parameter combination of the growth state monitoring clustering model to obtain the optimal growth state monitoring clustering model.

4. The method for monitoring the asexual propagation and growth of Kentucky bluegrass rhizomes based on clustering processing according to claim 1, characterized in that: Step S43: Obtaining the optimal parameter combination for the model specifically includes the following steps: Step S431: Initialize the search population individuals. Specifically, generate the current positions of N search individuals by random initialization, then optimize the individual positions using a mirror inversion mechanism to obtain the reflected positions of the individuals, and calculate the fitness value F of the search individuals in the population using the fitness function. i Individuals are sorted from best to worst based on their fitness values. The top N individuals with the highest fitness values ​​are selected based on their current position and reflected position to form a new search population. The formula used is as follows: ； In the formula, This represents the reflection position of the i-th individual in the d-th dimension. This represents the current position of the i-th individual in the d-th dimension. and Let represent the minimum and maximum values ​​of the search space in dimension d, respectively. This represents the scaling factor, which is a constant. Step S432: Update the individual search position, specifically by updating the individual search position using sine and cosine functions and a non-linear decreasing search. The formula used is as follows: ； In the formula, Indicates that the i-th individual is in the first place. The position of the d-th dimension in the population. Indicates that the i-th individual is in the first place. The position of the d-th dimension in the population. This indicates the position of the optimal individual in the population. Indicates the number of iterations. Indicates the maximum number of iterations. Indicates the search adjustment factor, and , Indicates the warning value. Indicates the safety threshold. Represents [0,2] Random numbers within the range of ] This represents a random number in the range [0,1], and e represents the base of the natural logarithm. Step S433: Search complete, specifically when the fitness value F of the searched individual is... i When the fitness threshold is exceeded and the maximum number of iterations is reached, the search is terminated and the global optimal position of the search individual is obtained. Specifically, the global optimal position of the search individual refers to the optimal parameter combination of the growth state monitoring clustering model.

5. The method for monitoring the asexual propagation and growth of Kentucky bluegrass rhizomes based on clustering processing according to claim 1, characterized in that: In step S5, the hierarchical clustering growth status monitoring specifically involves identifying multiple growth states of Kentucky bluegrass through a principal clustering structure, forming a two-layer labeling system, and obtaining growth status monitoring results; including the following steps: Step S51: The main state identification stage, specifically, takes the plant physiological data in the real-time growth monitoring data as input data and inputs it into the optimal growth state monitoring clustering model, performs the first stage main clustering operation, generates multiple main clusters, and in each cluster, according to the associated historical growth monitoring state labels, counts the frequency of the labels, selects the label with the highest frequency as the main state label of the cluster, and obtains the main state identification result. Step S52: Sub-state identification stage, used to analyze the subdivided states constituted by external factors behind the main state, and identify subdivided growth response patterns; specifically, it includes the following steps: Step S521: Obtaining sub-clusters. Specifically, based on the plant physiological data of each main cluster, select the corresponding plant environmental data and operational behavior data as the input data for the second stage of clustering, input them into the optimal growth state monitoring clustering model, perform the second stage sub-clustering operation, and generate multiple sub-clusters. Step S522: Output the sub-state recognition result. Specifically, perform statistical processing on the data of each sub-cluster, calculate the mean, standard deviation and maximum value of the data to form the feature representation vector of the cluster. Then, construct a set of predefined sub-state labels. Each sub-state label consists of multiple feature parameter ranges. Compare the cosine similarity of the feature representation vector of each sub-cluster with all sub-state labels, and select the label with the highest similarity as the sub-state label of the cluster to obtain the sub-state recognition result. Step S53: Output the growth status monitoring results. Specifically, based on the main state identification results and the sub-state identification results, the samples within the cluster are combined according to their correspondence to output the double-layer growth status corresponding to each monitoring point, thus obtaining the growth status monitoring results.

6. The method for monitoring the asexual propagation and growth of Kentucky bluegrass rhizomes based on clustering processing according to claim 1, characterized in that: In step S1, the acquisition of multi-source monitoring data specifically involves collecting data information from the rhizome growth area of ​​Kentucky bluegrass to obtain raw data for asexual reproduction growth monitoring. The raw data for asexual reproduction growth monitoring includes historical growth monitoring data and real-time growth monitoring data. Both the historical and real-time growth monitoring data include plant physiological data, plant environmental data, and operational behavior data for each monitoring point. The historical growth monitoring data also includes the growth monitoring status.

7. The method for monitoring the asexual propagation and growth of Kentucky bluegrass rhizomes based on clustering processing according to claim 6, characterized in that: In step S2, the multi-source data preprocessing specifically includes the following steps: Step S21: Data cleaning process, specifically, handling missing values, outliers, and duplicate data in the original data; Step S22: Data normalization processing, specifically, the data is normalized using the Min-Max normalization method; Step S23: Time series sample construction, specifically, the continuous monitoring data of each monitoring point is sliced ​​according to the set 24-hour time window to construct sample data with consistent time, and a sliding window mechanism is used to set the time overlap interval to 4 hours to generate multiple time series samples with overlapping relationships. Step S24: Reference label association, specifically, the growth monitoring status in the historical growth monitoring data is used as the reference label for the main clustering operation, which is only used for cluster label selection and does not participate in the clustering data processing process.

Images