Sugarcane leaf disease degree classification method, device and equipment based on SSA optimization and medium

Abstract

This invention discloses a method, apparatus, equipment, and medium for classifying the susceptibility of sugarcane leaves based on SSA optimization. The method includes: acquiring plant datasets of sugarcane leaf susceptibility areas and sugarcane susceptibility levels from several planting regions; constructing several sugarcane leaf susceptibility classification prediction models; optimizing the hyperparameters of each sugarcane leaf susceptibility classification prediction model based on the Sparrow Search algorithm; obtaining evaluation parameters corresponding to each optimized sugarcane leaf susceptibility classification prediction model; determining a target sugarcane leaf susceptibility classification prediction model from the several optimized models based on the evaluation parameters; and using the target sugarcane leaf susceptibility classification prediction model to predict the susceptibility of sugarcane leaf susceptibility areas. This invention belongs to the field of sugarcane leaf susceptibility prediction. This invention can improve the accuracy of sugarcane leaf susceptibility prediction.

Description

Technical Field

[0001] This invention relates to the field of predicting the degree of disease susceptibility in sugarcane leaves, and more particularly to a method, apparatus, equipment, and medium for classifying the degree of disease susceptibility in sugarcane leaves based on SSA optimization. Background Technology

[0002] Sugarcane is an important sugar crop and energy crop in my country, and its healthy growth is directly related to agricultural economic benefits and the security of agricultural product supply. During sugarcane cultivation, foliar diseases (such as brown spot, top rot, and rust) are widespread, severely affecting photosynthetic efficiency, leading to stunted plant growth, reduced yield, and deteriorated quality. Traditional disease diagnosis mainly relies on manual field observation and experience-based judgment, which suffers from high subjectivity, low efficiency, and poor timeliness, making it difficult to meet the needs of precision agricultural management under large-scale planting conditions.

[0003] In recent years, with the development of agricultural informatization and intelligent detection technologies, machine learning methods based on plant physiological parameters have been gradually applied to crop disease identification. Collecting physiological indicators closely related to plant health, such as chlorophyll content, canopy temperature, and nitrogen levels, and combining them with classification models to automatically determine the severity of diseases has become a research hotspot.

[0004] However, existing methods mostly employ single classification algorithms, resulting in limited model generalization ability. Furthermore, hyperparameter settings typically rely on manual experience or grid search, leading to low tuning efficiency, susceptibility to local optima, and unstable model performance. In addition, the lack of a mechanism for systematically comparing and optimizing multiple models limits further improvements in classification accuracy. Therefore, there is an urgent need for an intelligent identification method for sugarcane leaf disease susceptibility that can automatically optimize model parameters, comprehensively evaluate the performance of multiple models, and achieve high-precision classification, thereby improving the accuracy and practicality of disease monitoring. Summary of the Invention

[0005] This invention provides a method, apparatus, equipment, and medium for classifying the susceptibility of sugarcane leaves based on SSA optimization, which solves the technical problem of low accuracy in predicting the susceptibility of sugarcane leaves in the prior art and achieves the technical effect of improving the accuracy of predicting the susceptibility of sugarcane leaves.

[0006] In a first aspect, the present invention provides a method for classifying the susceptibility of sugarcane leaves to disease based on SSA optimization, including: Acquire plant datasets of disease-susceptible areas of sugarcane leaves and sugarcane disease susceptibility levels from several planting regions. The plant datasets include data on relative chlorophyll content, leaf surface temperature, and nitrogen content. Several sugarcane leaf disease susceptibility classification and prediction models were constructed. The algorithms corresponding to each sugarcane leaf disease susceptibility classification and prediction model are different. All sugarcane leaf disease susceptibility classification and prediction models are used to predict the disease susceptibility of sugarcane leaf disease-affected areas. The hyperparameters of the classification and prediction model for the disease susceptibility of various sugarcane leaves were optimized based on the sparrow search algorithm. The optimized sugarcane leaf disease susceptibility classification and prediction model was trained with several datasets to obtain the evaluation parameters corresponding to each optimized sugarcane leaf disease susceptibility classification and prediction model. The datasets included plant datasets and sugarcane disease susceptibility levels. Based on the evaluation parameters, a target sugarcane leaf disease severity classification and prediction model was determined from several optimized sugarcane leaf disease severity classification and prediction models, and the disease severity of the sugarcane leaf disease severity in the affected areas was predicted using the target sugarcane leaf disease severity classification and prediction model.

[0007] Furthermore, the disease susceptibility level of sugarcane is determined, including: Set up an area percentage gradient, where the area percentage gradient corresponds one-to-one with the sugarcane disease susceptibility level; Obtain the percentage of the sugarcane leaves infected with disease in the planting area; Determine the area percentage gradient of the infected sugarcane leaves and the corresponding sugarcane disease level.

[0008] Furthermore, several classification and prediction models for the degree of disease susceptibility in sugarcane leaves were constructed, including: KNN algorithm:

[0009] in, For the sample and The distance between them For the number of features, For the sample The first sample One characteristic, For the sample The One characteristic.

[0010] Furthermore, several classification and prediction models for the degree of disease susceptibility in sugarcane leaves were constructed, including: AdaBoost algorithm:

[0011] in, For the first The weights of each classifier Let be the error rate of the m-th weak classifier.

[0012] Furthermore, several classification and prediction models for the degree of disease susceptibility in sugarcane leaves were constructed, including: Logistic Regression Algorithm:

[0013] in, For response variables, For probability, For the first 1 eigenvector - All of these are model parameters.

[0014] Furthermore, the optimized sugarcane leaf disease susceptibility classification and prediction models were trained using several datasets to obtain evaluation parameters corresponding to each optimized sugarcane leaf disease susceptibility classification and prediction model, including: The optimized sugarcane leaf disease severity classification prediction model was trained using several datasets to obtain the accuracy, precision, recall, F1 score, and AUC for each sugarcane leaf disease severity classification prediction model. Based on the accuracy, precision, recall, F1 score, and AUC of each sugarcane leaf disease susceptibility classification prediction model, the evaluation parameters of the sugarcane leaf disease susceptibility classification prediction model are determined, including:

[0015] in, For the first Evaluation parameters of a sugarcane leaf disease susceptibility classification and prediction model. - All are weights. For the first The accuracy of a sugarcane leaf disease severity classification and prediction model. For the first The accuracy of a classification and prediction model for the degree of disease susceptibility in sugarcane leaves. For the first The recall rate of a sugarcane leaf disease severity classification and prediction model. For the first The F1 score of a classification and prediction model for the degree of disease susceptibility in sugarcane leaves. For the first The AUC of a sugarcane leaf disease susceptibility classification prediction model.

[0016] Furthermore, the hyperparameters of the classification and prediction models for the susceptibility of various sugarcane leaves were optimized based on the sparrow search algorithm, including: Determine the hyperparameters and adjustment range of the disease susceptibility classification prediction model for each sugarcane leaf; Initialize the SSA population, where each sparrow corresponds to a hyperparameter set; Determine the fitness for each hyperparameter set; Update the position of each sparrow to update the hyperparameter set; When the iteration reaches convergence, the optimal hyperparameter set is output based on the fitness of each hyperparameter set.

[0017] Secondly, the present invention provides a sugarcane leaf disease susceptibility classification device based on SSA optimization, comprising: The acquisition module is used to acquire plant datasets of diseased sugarcane leaf areas and sugarcane disease susceptibility levels from several planting regions. The plant datasets include data on relative chlorophyll content, leaf surface temperature, and nitrogen content. The module is used to build several sugarcane leaf disease susceptibility classification and prediction models. The algorithms corresponding to each sugarcane leaf disease susceptibility classification and prediction model are different. All sugarcane leaf disease susceptibility classification and prediction models are used to predict the susceptibility of sugarcane leaf disease-affected areas. The hyperparameter optimization module is used to optimize the hyperparameters of the classification and prediction models for the susceptibility of various sugarcane leaves based on the sparrow search algorithm. The evaluation parameter module is used to train the optimized sugarcane leaf disease susceptibility classification prediction model with several data sets to obtain the evaluation parameters corresponding to each optimized sugarcane leaf disease susceptibility classification prediction model. The data sets include plant datasets and sugarcane disease susceptibility levels. The target model determination module is used to determine the target sugarcane leaf disease severity classification prediction model from several optimized sugarcane leaf disease severity classification prediction models based on evaluation parameters, and to predict the disease severity of sugarcane leaf disease-affected areas using the target sugarcane leaf disease severity classification prediction model.

[0018] Thirdly, the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to execute a sugarcane leaf susceptibility classification method based on SSA optimization as provided in the first aspect.

[0019] Fourthly, the present invention provides a non-transitory computer-readable storage medium, wherein when the instructions in the non-transitory computer-readable storage medium are executed by a processor of an electronic device, the electronic device is able to execute the sugarcane leaf disease severity classification method based on SSA optimization as provided in the first aspect.

[0020] One or more technical solutions provided in this invention have at least the following technical effects or advantages: This invention constructs a representative plant dataset by collecting key physiological indicators such as relative chlorophyll content, leaf surface temperature, and nitrogen content from diseased sugarcane leaves in multiple planting areas. This dataset is then combined with manually determined disease severity levels to form labeled training samples, ensuring the scientific validity and reliability of the data. Based on this, various classification and prediction models using different algorithms (such as KNN, AdaBoost, and logistic regression) are built, fully utilizing the modeling characteristics of each algorithm to enhance model diversity and adaptability. Furthermore, a sparrow search algorithm is introduced to automatically optimize the hyperparameters of each model, overcoming the shortcomings of traditional methods that rely on empirical parameter tuning, are inefficient, and are prone to getting trapped in local optima, significantly improving the model's convergence speed and prediction performance. By training the optimized models and calculating their accuracy, F1 score, AUC, and other multi-dimensional evaluation parameters, the model performance is comprehensively evaluated, and the optimal target classification model is finally selected for intelligent identification and prediction of sugarcane leaf disease severity. This invention achieves full automation and intelligence from data acquisition, model building, parameter optimization to model selection, which not only improves the accuracy and efficiency of disease identification, but also enhances the model's generalization ability in different environments, providing reliable technical support for early monitoring and precise control of sugarcane diseases. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart illustrating a method for classifying the susceptibility of sugarcane leaves based on SSA optimization provided by this invention; Figure 2 This is a flowchart illustrating another method for classifying the susceptibility of sugarcane leaves based on SSA optimization, provided by the present invention. Detailed Implementation

[0023] This invention provides a sugarcane leaf disease susceptibility classification method based on SSA optimization, which solves the technical problem of low accuracy in predicting sugarcane leaf disease susceptibility in the prior art.

[0024] The technical solution of this invention is to solve the above-mentioned technical problems, and the overall idea is as follows: The sugarcane leaf disease susceptibility classification method based on SSA optimization includes: acquiring plant datasets of sugarcane leaf disease-susceptible areas and sugarcane disease susceptibility levels from several planting regions, where the plant datasets include relative chlorophyll content, leaf surface temperature, and nitrogen content data; constructing several sugarcane leaf disease susceptibility classification prediction models, each with a different algorithm, all used to predict the susceptibility level of sugarcane leaf disease-susceptible areas; optimizing the hyperparameters of each sugarcane leaf disease susceptibility classification prediction model based on the Sparrow Search algorithm; training the optimized sugarcane leaf disease susceptibility classification prediction models with several datasets to obtain evaluation parameters for each optimized model, where the datasets include plant datasets and sugarcane disease susceptibility levels; determining a target sugarcane leaf disease susceptibility classification prediction model from the optimized models based on the evaluation parameters, and using the target model to predict the susceptibility level of sugarcane leaf disease-susceptible areas.

[0025] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0026] First, it should be clarified that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0027] This invention provides, for example Figure 1 The sugarcane leaf susceptibility classification method based on SSA optimization shown includes steps S11-S15: Step S11: Obtain plant datasets of susceptible sugarcane leaf areas and sugarcane disease severity levels from several planting regions. The plant datasets include data on relative chlorophyll content, leaf surface temperature, and nitrogen content.

[0028] The planting area refers to multiple different sugarcane fields or geographical areas (such as different farms or different climate zones) to obtain representative samples and avoid the model being applicable only to a single environment.

[0029] The diseased area of ​​sugarcane leaves refers to the part of the sugarcane leaf that has been infected with disease. The types of disease can be brown streak, ring spot, and mosaic virus.

[0030] The TYS-4N plant nutrient meter can be used to measure SPAD, leaf temperature, and nitrogen content in susceptible areas of sugarcane top rot, brown stripe, and mosaic virus. The average value of measurements taken at multiple different locations is used as the final data sample.

[0031] Determining the disease susceptibility level of sugarcane includes: Set up an area percentage gradient, where the area percentage gradient corresponds one-to-one with the sugarcane disease susceptibility level.

[0032] The area percentage gradient can be determined according to the actual situation. The area percentage gradient provided by this invention is: mild (less than 10%), moderate (10%-50%), moderately heavy (50%-70%), and severe (more than 80%).

[0033] Obtain the percentage of the sugarcane leaves infected with disease in the planting area.

[0034] The area percentage of diseased sugarcane leaves can be determined through image processing and other methods.

[0035] Determine the area percentage gradient of the infected sugarcane leaves and the corresponding sugarcane disease level.

[0036] After determining the area percentage of sugarcane leaves susceptible to disease in the planting area, the area percentage of sugarcane leaves susceptible to disease can be compared with the area percentage gradient to determine the corresponding sugarcane disease susceptibility level.

[0037] The plant dataset and sugarcane disease susceptibility levels can be used together as training data and input into the model described below to train the model.

[0038] Step S12: Construct several sugarcane leaf disease susceptibility classification prediction models. The algorithms corresponding to each sugarcane leaf disease susceptibility classification prediction model are different. All sugarcane leaf disease susceptibility classification prediction models are used to predict the disease susceptibility of sugarcane leaf disease-affected areas.

[0039] Since different algorithms are used to construct various sugarcane leaf disease susceptibility classification and prediction models, this invention provides six different machine learning classification algorithms for experimentation, including KNN, AdaBoost, RF, LR, DT, and XGBoost: KNN algorithm:

[0040] in, For the sample and The distance between them For the number of features, For the sample The first sample One characteristic, For the sample The One characteristic.

[0041] The sample refers to the planting area, and the features refer to the plant dataset (relative chlorophyll content, leaf temperature and nitrogen content data) and the sugarcane disease susceptibility level.

[0042] The KNN algorithm is an instance-based learning method that classifies samples by calculating the distance between them. The basic idea of ​​the KNN algorithm is: given a test sample, find the K nearest samples in the training set, and determine the class of the test sample based on the majority class of these K samples.

[0043] A decision tree is a tree-structured model that classifies samples through a series of conditional judgments. Each internal node represents a test on a feature, each branch represents a test result, and each leaf node represents a class.

[0044] AdaBoost algorithm:

[0045] in, For the first The weights of each classifier Let be the error rate of the m-th weak classifier.

[0046] AdaBoost is an ensemble learning method that combines multiple weak classifiers to build a strong classifier. The core idea of ​​AdaBoost is that each time a new weak classifier is trained, the weights of samples that were misclassified by the previous weak classifier are increased, thus making the new weak classifier pay more attention to these difficult-to-classify samples.

[0047] Random Forest is an ensemble learning method based on decision trees. It improves the stability and accuracy of the model by constructing multiple decision trees and averaging their results. Each decision tree is trained on a different subset obtained through sampling with replacement.

[0048] Logistic Regression Algorithm:

[0049] in, For response variables, For probability, For the first 1 eigenvector - These are all model parameters. Logistic regression is a generalized linear model used to solve binary or multi-class classification problems.

[0050] XGBoost (Extreme Gradient Boosting) is an ensemble learning method based on gradient boosting. XGBoost minimizes the loss function by successively adding new weak classifiers.

[0051] Step S13: Optimize the hyperparameters of the classification and prediction model for the disease susceptibility of each sugarcane leaf based on the sparrow search algorithm.

[0052] Specifically, this includes: determining the hyperparameters and adjustment range of the classification prediction model for the susceptibility of each sugarcane leaf; initializing the SSA population, where each sparrow corresponds to a hyperparameter group; determining the fitness of each hyperparameter group; updating the position of each sparrow to update the hyperparameter group; and when the iteration reaches convergence, outputting the optimal hyperparameter group based on the fitness of each hyperparameter group.

[0053] The Sparrow Search Algorithm (SSA) is a swarm intelligence-based optimization algorithm inspired by the foraging behavior of sparrows. This algorithm finds the global optimum by simulating the behavioral patterns of sparrows during foraging. SSA primarily simulates the behavior of three types of sparrows: discoverers, joiners, and scouts.

[0054] Discoverer sparrows explore the search space for new solutions. They move randomly to find potential food sources (i.e., candidate solutions in the optimization problem). Their movements are random but have a limited step size to ensure search diversity. Joiner sparrows follow the discoverer's guidance, moving towards better locations. They update their positions based on the discoverer's findings, and their behavior depends on the quality of the discoverer's findings and the uncertainty of the current environment. If the discoverer finds a better food source, the joiner will quickly follow; otherwise, they may remain in place or conduct small-scale exploration. Watcher sparrows monitor the environment for potential threats. When a threat is detected, they guide the entire flock away from the danger zone. Watcher sparrows also move randomly, but they adjust their behavior based on the flock's current state and changes in the environment to ensure the flock's safety.

[0055] First, SSA defines the search space to determine the range of values ​​for each hyperparameter. Then, it randomly generates initial hyperparameter combinations, each corresponding to an individual sparrow. It calculates the fitness value of each hyperparameter combination, that is, it uses the hyperparameter combination to train the model and evaluates its performance on the validation set.

[0056] The third step is iterative optimization. In each iteration, the position of each hyperparameter combination is updated, i.e., the hyperparameter values ​​are adjusted. The model is retrained using the new hyperparameter combinations, and its performance is evaluated. Based on the performance evaluation results, the optimal hyperparameter combination is updated. When the maximum number of iterations is reached or other termination conditions are met, the optimal hyperparameter combination (maximum optimal value) and its corresponding model performance are output.

[0057] The hyperparameters of several sugarcane leaf disease severity classification and prediction models in step S12 are optimized one by one using the sparrow search algorithm.

[0058] Step S14: Train the optimized sugarcane leaf disease susceptibility classification prediction model with several data sets to obtain the evaluation parameters corresponding to each optimized sugarcane leaf disease susceptibility classification prediction model. The data sets include plant datasets and sugarcane disease susceptibility levels.

[0059] Specifically, it includes: The optimized sugarcane leaf disease severity classification prediction model was trained using several datasets, and the accuracy, precision, recall, F1 score, and AUC of each sugarcane leaf disease severity classification prediction model were obtained.

[0060] Accuracy refers to the proportion of correctly classified samples out of the total number of samples; precision refers to the proportion of true positives among all samples predicted as positive; recall refers to the proportion of true positives among all samples that were actually positive; the F1 score is the harmonic mean of precision and recall, used to comprehensively evaluate the model's performance; AUC (Area Under the Curve) is the area under the ROC curve, used to evaluate the model's classification performance. The ROC curve is a curve plotted with the false positive rate (FPR) on the x-axis and the true positive rate (TPR) on the y-axis.

[0061]

[0062] in: TPR (True Positive Rate): Also known as sensitivity. ; FPR (False Positive Rate): Also known as 1-Specificity. ; The AUC value ranges from 0 to 1, with a value closer to 1 indicating better classification performance. An AUC of 0.5 indicates that the model's classification performance is the same as random guessing, while an AUC of 1 indicates that the model's classification performance is perfect.

[0063] Where: TP (True Positive): The number of samples that are actually positive and correctly predicted as positive by the model. TN (True Negative): The number of samples that are actually negative and correctly predicted as negative by the model. FP (False Positive): The number of samples that are actually negative but incorrectly predicted as positive by the model. FN (False Negative): The number of samples that are actually positive but incorrectly predicted as negative by the model.

[0064] The values ​​of accuracy, precision, recall, F1 score, and AUC range from 0 to 1. The closer the value is to 1, the better the model's performance. In addition to the method of determining evaluation parameters provided by this invention, one or more of the parameters such as accuracy, precision, recall, F1 score, and AUC can also be selected directly as evaluation parameters.

[0065] Based on the accuracy, precision, recall, F1 score, and AUC of each sugarcane leaf disease susceptibility classification prediction model, the evaluation parameters of the sugarcane leaf disease susceptibility classification prediction model are determined, including:

[0066] in, For the first Evaluation parameters of a sugarcane leaf disease susceptibility classification and prediction model. - All are weights. For the first The accuracy of a sugarcane leaf disease severity classification and prediction model. For the first The accuracy of a classification and prediction model for the degree of disease susceptibility in sugarcane leaves. For the first The recall rate of a sugarcane leaf disease severity classification and prediction model. For the first The F1 score of a classification and prediction model for the degree of disease susceptibility in sugarcane leaves. For the first The AUC of a sugarcane leaf disease susceptibility classification prediction model.

[0067] Step S15: Based on the evaluation parameters, determine the target sugarcane leaf disease severity classification prediction model from several optimized sugarcane leaf disease severity classification prediction models, and use the target sugarcane leaf disease severity classification prediction model to predict the disease severity of the sugarcane leaf disease-affected areas.

[0068] Specifically, the optimized sugarcane leaf disease susceptibility classification and prediction model with the largest evaluation parameters can be used as the target sugarcane leaf disease susceptibility classification and prediction model.

[0069] In addition, the entire process of this invention can also be referenced. Figure 2 .

[0070] In summary, this invention constructs a representative plant dataset by collecting key physiological indicators such as relative chlorophyll content, leaf surface temperature, and nitrogen content from diseased sugarcane leaves in multiple planting areas. This dataset is then combined with manually determined disease severity levels to form labeled training samples, ensuring the scientific validity and reliability of the data. Based on this, various classification and prediction models using different algorithms (such as KNN, AdaBoost, and logistic regression) are built, fully utilizing the modeling characteristics of each algorithm to enhance model diversity and adaptability. Furthermore, a sparrow search algorithm is introduced to automatically optimize the hyperparameters of each model, overcoming the shortcomings of traditional methods that rely on empirical parameter tuning, are inefficient, and are prone to getting trapped in local optima, significantly improving the model's convergence speed and prediction performance. By training the optimized models and calculating their accuracy, F1 score, AUC, and other multi-dimensional evaluation parameters, the model performance is comprehensively evaluated, and the optimal target classification model is finally selected for intelligent identification and prediction of sugarcane leaf disease severity. This invention achieves full automation and intelligence from data acquisition, model building, parameter optimization to model selection, which not only improves the accuracy and efficiency of disease identification, but also enhances the model's generalization ability in different environments, providing reliable technical support for early monitoring and precise control of sugarcane diseases.

[0071] Based on the same inventive concept, this invention provides a sugarcane leaf disease susceptibility classification device optimized by SSA, comprising: The acquisition module is used to acquire plant datasets of diseased sugarcane leaf areas and sugarcane disease susceptibility levels from several planting regions. The plant datasets include data on relative chlorophyll content, leaf surface temperature, and nitrogen content. The module is used to build several sugarcane leaf disease susceptibility classification and prediction models. The algorithms corresponding to each sugarcane leaf disease susceptibility classification and prediction model are different. All sugarcane leaf disease susceptibility classification and prediction models are used to predict the susceptibility of sugarcane leaf disease-affected areas. The hyperparameter optimization module is used to optimize the hyperparameters of the classification and prediction models for the susceptibility of various sugarcane leaves based on the sparrow search algorithm. The evaluation parameter module is used to train the optimized sugarcane leaf disease susceptibility classification prediction model with several data sets to obtain the evaluation parameters corresponding to each optimized sugarcane leaf disease susceptibility classification prediction model. The data sets include plant datasets and sugarcane disease susceptibility levels. The target model determination module is used to determine the target sugarcane leaf disease severity classification prediction model from several optimized sugarcane leaf disease severity classification prediction models based on evaluation parameters, and to predict the disease severity of sugarcane leaf disease-affected areas using the target sugarcane leaf disease severity classification prediction model.

[0072] Based on the same inventive concept, the present invention also provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to execute a sugarcane leaf disease susceptibility classification method based on SSA optimization as described above.

[0073] Based on the same inventive concept, the present invention also provides a non-transitory computer-readable storage medium, which, when the instructions in the storage medium are executed by the processor of an electronic device, enables the electronic device to execute the sugarcane leaf disease severity classification method based on SSA optimization as described above.

[0074] Since the electronic device described in this embodiment is an electronic device used to implement the information processing method in the embodiments of the present invention, those skilled in the art can understand the specific implementation methods and various variations of the electronic device in this embodiment based on the information processing method described in the embodiments of the present invention. Therefore, how the electronic device implements the method in the embodiments of the present invention will not be described in detail here. Any electronic device used by those skilled in the art to implement the information processing method in the embodiments of the present invention falls within the scope of protection of the present invention.

[0075] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0076] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0077] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0078] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0079] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0080] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for classifying the susceptibility of sugarcane leaves to disease based on SSA optimization, characterized in that, include: Acquire plant datasets of disease-susceptible areas of sugarcane leaves and sugarcane disease susceptibility levels from several planting regions. The plant datasets include data on relative chlorophyll content, leaf surface temperature, and nitrogen content. Several classification and prediction models for the susceptibility of sugarcane leaves to disease were constructed. Each model uses a different algorithm, and all models are used to predict the susceptibility of sugarcane leaves to disease-affected areas. These include: the KNN algorithm. in, For the sample and The distance between them For the number of features, For the sample The first sample One characteristic, For the sample The One feature; AdaBoost algorithm: in, For the first The weights of each classifier Let m be the error rate of the m-th weak classifier; Logistic Regression Algorithm: in, For response variables, For probability, For the first 1 eigenvector - All are model parameters; The hyperparameters of the classification and prediction model for the disease susceptibility of various sugarcane leaves were optimized based on the sparrow search algorithm. The optimized sugarcane leaf disease susceptibility classification and prediction model was trained using several datasets to obtain the evaluation parameters corresponding to each optimized sugarcane leaf disease susceptibility classification and prediction model. The datasets included plant datasets and sugarcane disease susceptibility levels. The training process included: training the optimized sugarcane leaf disease susceptibility classification and prediction model using several datasets to obtain the accuracy, precision, recall, F1 score, and AUC corresponding to each sugarcane leaf disease susceptibility classification and prediction model. Based on the accuracy, precision, recall, F1 score, and AUC of each sugarcane leaf disease susceptibility classification prediction model, the evaluation parameters of the sugarcane leaf disease susceptibility classification prediction model are determined, including: in, For the first Evaluation parameters of a sugarcane leaf disease susceptibility classification and prediction model. - All are weights. For the first The accuracy of a classification and prediction model for the degree of disease susceptibility in sugarcane leaves. For the first The accuracy of a classification and prediction model for the degree of disease susceptibility in sugarcane leaves. For the first The recall rate of a sugarcane leaf disease severity classification prediction model. For the first The F1 score of a classification and prediction model for the degree of disease susceptibility in sugarcane leaves. For the first AUC of a sugarcane leaf disease susceptibility classification prediction model; Based on the evaluation parameters, a target sugarcane leaf disease severity classification and prediction model is determined from several optimized sugarcane leaf disease severity classification and prediction models, and the disease severity of the sugarcane leaf disease severity in the affected areas is predicted using the target sugarcane leaf disease severity classification and prediction model.

2. The sugarcane leaf disease susceptibility classification method based on SSA optimization as described in claim 1, characterized in that, Determining the disease susceptibility level of sugarcane includes: Set up an area percentage gradient, where the area percentage gradient corresponds one-to-one with the sugarcane disease susceptibility level; Obtain the percentage of the sugarcane leaves infected with disease in the planting area; Determine the area percentage gradient of the diseased areas on the sugarcane leaves and determine the corresponding sugarcane disease susceptibility level.

3. The sugarcane leaf disease susceptibility classification method based on SSA optimization as described in claim 1, characterized in that, The hyperparameters of the classification and prediction models for the susceptibility of various sugarcane leaves were optimized based on the sparrow search algorithm, including: Determine the hyperparameters and adjustment range of the disease susceptibility classification prediction model for each sugarcane leaf; Initialize the SSA population, where each sparrow corresponds to a hyperparameter set; Determine the fitness for each hyperparameter set; Update the position of each sparrow to update the hyperparameter set; When the iteration reaches convergence, the optimal hyperparameter set is output based on the fitness of each hyperparameter set.

4. A sugarcane leaf disease susceptibility classification device based on SSA optimization, characterized in that, include: The acquisition module is used to acquire plant datasets of diseased sugarcane leaf areas and sugarcane disease susceptibility levels from several planting regions. The plant datasets include data on relative chlorophyll content, leaf surface temperature, and nitrogen content. The module is used to construct several classification and prediction models for the susceptibility of sugarcane leaves to disease. Each model uses a different algorithm, but all models are used to predict the susceptibility of sugarcane leaves to disease-affected areas. The model includes the KNN algorithm. in, For the sample and The distance between them For the number of features, For the sample The first sample One characteristic, For the sample The One feature; AdaBoost algorithm: in, For the first The weights of each classifier Let m be the error rate of the m-th weak classifier; Logistic Regression Algorithm: in, For response variables, For probability, For the first 1 eigenvector - All are model parameters; the hyperparameter optimization module is used to optimize the hyperparameters of the classification and prediction model of the susceptibility of various sugarcane leaves based on the sparrow search algorithm. The evaluation parameter module is used to train the optimized sugarcane leaf disease susceptibility classification and prediction model with several sets of data to obtain the evaluation parameters corresponding to each optimized sugarcane leaf disease susceptibility classification and prediction model. The data sets include plant datasets and sugarcane disease susceptibility levels. The module includes: training the optimized sugarcane leaf disease susceptibility classification and prediction model with several sets of data to obtain the accuracy, precision, recall, F1 score, and AUC corresponding to each sugarcane leaf disease susceptibility classification and prediction model. Based on the accuracy, precision, recall, F1 score, and AUC of each sugarcane leaf disease susceptibility classification prediction model, the evaluation parameters of the sugarcane leaf disease susceptibility classification prediction model are determined, including: in, For the first Evaluation parameters of a sugarcane leaf disease susceptibility classification and prediction model. - All are weights. For the first The accuracy of a classification and prediction model for the degree of disease susceptibility in sugarcane leaves. For the first The accuracy of a classification and prediction model for the degree of disease susceptibility in sugarcane leaves. For the first The recall rate of a sugarcane leaf disease severity classification prediction model. For the first The F1 score of a classification and prediction model for the degree of disease susceptibility in sugarcane leaves. For the first AUC of a sugarcane leaf disease susceptibility classification prediction model; The target model determination module is used to determine the target sugarcane leaf disease severity classification prediction model from several optimized sugarcane leaf disease severity classification prediction models based on evaluation parameters, and to predict the disease severity of sugarcane leaf disease-affected areas using the target sugarcane leaf disease severity classification prediction model.

5. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the SSA-optimized sugarcane leaf disease severity classification method as described in any one of claims 1 to 3.

6. A non-transitory computer-readable storage medium, characterized in that, When the instructions in the non-transitory computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to perform the SSA-optimized sugarcane leaf disease severity classification method as described in any one of claims 1 to 3.

Images