Crop disease risk assessment method and device

Abstract

The invention relates to the technical field of assessment model risk construction, in particular to a crop disease risk assessment method and device. The method comprises the following steps: carrying out crop growth form image data acquisition and abnormal state feature recognition on a crop planting area to obtain crop form abnormal state data; performing crop abnormal form identification and survival sensitivity analysis on the crop form abnormal state data to obtain nutrition deficiency-survival sensitivity data; trend clustering analysis is carried out according to the nutrition deficiency-survival sensitive data, and limited response trend clustering data is obtained; building a crop disease evaluation model for the limited response trend clustering data based on a multi-layer perceptron to obtain the crop disease evaluation model; and sending the crop disease assessment model to a data processing cloud platform to execute crop disease risk assessment. The crop disease risk assessment technology is optimized, so that the crop disease risk assessment technology is more perfect.

Description

Technical Field

[0001] The present invention relates to the technical field of risk construction of assessment models, and in particular to a method and device for assessing the risk of crop diseases. Background Art

[0002] Crop disease risk assessment methods play a vital role in modern agriculture, especially in the context of global climate change and increasing agricultural productivity requirements. These methods aim to predict and assess the potential threat of diseases to crops so that timely and effective measures can be taken to reduce losses. Traditional assessment methods rely on empirical data and field observations, which usually require a lot of manpower and time investment. With the development of science and technology, data-driven methods have gradually become mainstream. Using high-resolution remote sensing technology and drone monitoring, crop growth conditions and disease information can be obtained in real time. Combined with big data analysis and machine learning technology, these methods can more accurately predict the occurrence and spread of diseases. For example, by analyzing historical disease data and meteorological conditions, a disease prediction model can be established to provide accurate early warning information. The integration of these technologies not only improves the accuracy of disease risk assessment, but also speeds up the response speed, enabling farmers to take effective prevention and control measures before the disease breaks out, thereby protecting crop health and improving agricultural production efficiency. However, a traditional crop disease risk assessment method has the problems of low risk assessment accuracy and inaccurate trend analysis. Summary of the invention

[0003] Based on this, it is necessary to provide a crop disease risk assessment method and device to solve at least one of the above technical problems.

[0004] To achieve the above object, a method for assessing crop disease risk is provided, the method comprising the following steps:

[0005] Step S1: collecting crop growth morphology image data in the crop planting area through electronic monitoring equipment to obtain crop growth morphology images; performing abnormal state feature recognition on the crop growth morphology images to obtain crop morphology abnormal state data;

[0006] Step S2: performing crop deformity morphology recognition on the crop morphology abnormal state data to obtain crop deformity morphology data; performing survival sensitivity analysis based on the crop deformity morphology data to obtain nutrient deficiency-survival sensitivity data; performing crop physiological restriction response simulation based on the nutrient deficiency-survival sensitivity data to obtain crop physiological restriction response data;

[0007] Step S3: performing trend cluster analysis on the crop physiological restricted response data to obtain restricted response trend cluster data;

[0008] Step S4: constructing a crop disease assessment model based on the restricted response trend clustering data based on the multi-layer perceptron to obtain a crop disease assessment model; and sending the crop disease assessment model to the data processing cloud platform to perform crop disease risk assessment.

[0009] First, the present invention collects crop growth morphology image data through electronic monitoring equipment, which can realize real-time monitoring and recording of crop growth process. After these image data are identified by abnormal state characteristics, abnormal conditions that may exist in crop growth, such as diseases, malnutrition or external environmental factors, can be effectively detected. The accurate identification of these abnormal state data provides a basis for timely taking targeted management and intervention measures. The abnormal state data of crop morphology are used to identify deformed morphology, which can further analyze the specific morphological characteristics of crops and identify deformed or abnormal growth patterns. By performing survival sensitivity analysis based on deformed morphological data, the survival ability and adaptability of crops under different morphological abnormalities can be evaluated. In particular, the physiological restricted response simulation based on nutritional deficiency-survival sensitivity data can simulate the growth response of crops under malnutrition or other physiological restriction conditions. Trend cluster analysis processes the crop physiological restricted response data, and can identify and analyze the trend and pattern of crop growth response under different physiological restriction conditions. This analysis can help agricultural experts and managers to more comprehensively understand the growth performance of crops under different environments or management conditions, and identify possible common trends or abnormal phenomena. A crop disease assessment model is constructed based on a multi-layer perceptron, using trend clustering data as a basis. The multi-layer perceptron is a powerful deep learning model that can learn features from complex data and accurately classify and predict data. The construction of this model can not only effectively analyze the disease risks that crops may face under different physiological constraints, but also provide predictive assessment results. The crop disease assessment model is sent to a data processing cloud platform, so that the disease risk assessment process can be carried out in a wider data context, thereby improving the accuracy and reliability of the assessment. In this way, agricultural managers can obtain detailed information about the health status of crops in a timely manner, so as to take appropriate management measures to reduce the impact of diseases on yield and quality. Therefore, the present invention is an optimization process made to a traditional crop disease risk assessment method, which solves the problems of low risk assessment accuracy and inaccurate trend analysis in a traditional crop disease risk assessment method, and improves the accuracy of risk assessment and trend analysis.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: collecting crop growth morphology image data in the crop planting area through electronic monitoring equipment to obtain crop growth morphology images;

[0012] Step S12: performing imaging angle feature analysis on the crop growth morphology image to obtain crop image imaging angle data;

[0013] Step S13: performing abnormal state feature recognition on the crop growth morphology image according to the crop image imaging angle data to obtain crop morphology abnormal state data.

[0014] The present invention collects crop growth morphology image data in the crop planting area through electronic monitoring equipment, and this step is the basis of the entire monitoring system. By collecting crop growth morphology image data in real time, the system can accurately record the morphological characteristics and changes of crops at different growth stages, providing a rich data basis for subsequent analysis. The imaging angle feature analysis of the crop growth morphology image helps to further optimize the processing and recognition of image data. The imaging angle feature analysis can help the system better understand and utilize the collected image data, and improve the accuracy and reliability of the data. Abnormal state feature recognition is performed according to the imaging angle data of the crop image to obtain abnormal state data of the crop morphology. This step is a further extension of step S12. By combining the imaging angle data, the system can more accurately identify abnormal states in the growth process of crops, such as diseases, morphological distortions, etc. The accurate identification of these abnormal state data provides an important basis for early warning and intervention for agricultural production managers, and helps to take timely measures to protect the health of crops and optimize production efficiency.

[0015] Preferably, step S2 comprises the following steps:

[0016] Step S21: performing crop deformity morphology recognition on the crop morphology abnormal state data to obtain crop deformity morphology data;

[0017] Step S22: evaluating the nutritional deficiency symptoms according to the crop deformity morphology data to obtain the nutritional deficiency symptom evaluation data of the crop;

[0018] Step S23: performing survival sensitivity analysis on the crop nutrition deficiency symptom assessment data to obtain nutrition deficiency-survival sensitivity data;

[0019] Step S24: simulating crop physiological limitation response of crop nutrient deficiency symptom assessment data according to nutrient deficiency-survival sensitivity data to obtain crop physiological limitation response data.

[0020] The present invention performs crop deformity morphology recognition on crop morphology abnormal state data, and this step is intended to further analyze and identify the possible deformities or abnormal growth forms of crops. By accurately identifying the deformity morphology data of crops, the system can more accurately evaluate the growth status and potential risks of crops under different morphological abnormalities. Nutritional deficiency symptoms are evaluated according to the crop deformity morphology data. This step focuses on analyzing the possible malnutrition symptoms of crops. By deeply evaluating the degree of influence of nutritional deficiency on crops, the system can quantify the specific influence of different nutrient element deficiencies on crop growth, and provide a data basis for subsequent physiological restricted response simulation. Survival sensitivity analysis is performed on the crop nutritional deficiency symptom assessment data to obtain nutritional deficiency-survival sensitivity data. The purpose of this step is to evaluate the survival ability and adaptability of crops under different nutritional deficiency conditions. The system can analyze the survival sensitivity of different nutritional deficiency symptoms to crop growth, and then provide targeted decision support and management suggestions for agricultural production managers. Crop physiological restricted response simulation is performed on the crop nutritional deficiency symptom assessment data according to the nutritional deficiency-survival sensitivity data to obtain crop physiological restricted response data. This step can predict and analyze the growth restrictions faced by crops in actual production environments by simulating the physiological responses of crops under different nutrient deficiency symptoms. These data not only help optimize agricultural management strategies and decisions, but also help agricultural producers effectively prevent and respond to various risks and challenges in the growth process.

[0021] Preferably, step S23 includes the following steps:

[0022] Step S231: performing nutrient element deficiency identification on the crop nutrient element deficiency symptom assessment data to obtain crop nutrient element deficiency data;

[0023] Step S232: performing an analysis of the environmental adaptability impairment of crops based on the crop nutrient element deficiency data to obtain element deficiency environmental adaptability impairment data;

[0024] Step S233: calculating the impairment difference of the element-missing environment adaptability impairment data to obtain the environment adaptability impairment difference data;

[0025] Step S234: Perform survival sensitivity analysis based on the environmental adaptability impairment difference data to obtain nutritional deficiency-environmental survival sensitivity data.

[0026] The present invention identifies the deficiency of nutrient elements in crop nutrition deficiency symptom assessment data, and this step is intended to accurately identify and analyze the possible deficiency of nutrient elements in crops. By deeply evaluating the degree of deficiency of different nutrient elements (such as nitrogen, phosphorus, potassium, etc.), the system can quantify the specific impact of different element deficiencies on crop growth, and provide basic data for subsequent environmental adaptability analysis. According to the crop nutrient element deficiency data, the environmental adaptability impairment analysis of crops is performed to obtain element deficiency environmental adaptability impairment data. This step focuses on analyzing the changes in the environmental adaptability of crops under the condition of nutrient element deficiency. By evaluating the adaptability loss of different nutrient element deficiencies to the crop growth environment, the system can quantify the impairment of environmental adaptability and provide data support for the subsequent impairment difference calculation. The element deficiency environmental adaptability impairment data is calculated to obtain environmental adaptability impairment difference data. In this step, by comparing the environmental adaptability loss under different element deficiency conditions, the system can analyze the differences in the impact of different nutrient element deficiencies on the adaptability of crop growth environments. These data help to understand the environmental adaptability challenges and risks that crops may face under different element deficiency conditions. Survival sensitivity analysis is performed based on the difference data of environmental adaptability impairment to obtain nutrient deficiency-environmental survival sensitivity data. The purpose of this step is to evaluate the survival ability and adaptability of crops under different nutrient deficiency conditions. The system can analyze the survival sensitivity of crops under different element deficiency conditions, and then provide targeted decision support and management suggestions for agricultural production managers.

[0027] Preferably, step S24 comprises the following steps:

[0028] Step S241: identifying key factors of the crop nutrition deficiency symptom assessment data according to the nutrition deficiency-survival sensitivity data to obtain nutrition deficiency key factor data;

[0029] Step S242: performing factor association analysis on the key factor data of nutritional deficiency to obtain nutritional deficiency factor association data;

[0030] Step S243: performing multi-level data decomposition according to the nutritional deficiency factor correlation data to obtain nutritional deficiency multi-level factor correlation data;

[0031] Step S244: simulating crop physiological restriction response according to the multi-level factor correlation data of nutrient deficiency to obtain crop physiological restriction response data.

[0032] The present invention identifies key factors of crop nutrient deficiency symptom assessment data based on nutrient deficiency-survival sensitive data. This step aims to identify and analyze key factors affecting crop nutrient deficiency. By deeply analyzing the key influencing factors of nutrient deficiency on crop growth, the system can determine the main factors affecting the response of nutrient deficiency and provide basic data for subsequent factor association analysis. Factor association analysis is performed on the key factor data of nutrient deficiency to obtain nutrient deficiency factor association data. This step focuses on analyzing the correlation and influence degree between different nutrient deficiency factors. Through association analysis, the system can reveal the possible interactions and influences between different nutrient elements, and help understand their comprehensive influence on crop growth and health. Multi-level data decomposition is performed according to the nutrient deficiency factor association data to obtain nutrient deficiency multi-level factor association data. This step further decomposes the association data into multiple levels to more carefully analyze the complex influence mechanism of different nutrient deficiency on crop growth. Through multi-level data decomposition, the system can deeply explore the role and interaction of nutrient deficiency factors at different levels, and improve the accuracy and depth of data analysis. Crop physiological restricted response simulation is performed according to the nutrient deficiency multi-level factor association data to obtain crop physiological restricted response data. This step can predict and analyze the physiological limitations that crops may face in actual production environments by simulating the growth response of crops under different nutrient deficiency conditions. These data can not only help agricultural production managers better understand and evaluate the impact of nutrient deficiency on crop yield and quality, but also provide a scientific basis for formulating precise agricultural management strategies.

[0033] Preferably, step S3 comprises the following steps:

[0034] Step S31: performing spatial cross-scale analysis on crop physiological restricted response data to obtain restricted response spatial cross-scale characteristic data;

[0035] Step S32: performing spatial response interpolation processing on the restricted response space cross-scale characteristic data to obtain restricted response space interpolation data;

[0036] Step S33: performing trend cluster analysis on the crop physiological restricted response data according to the restricted response space interpolation data to obtain restricted response trend cluster data.

[0037] The present invention performs spatial cross-scale analysis on crop physiological restricted response data to obtain restricted response spatial cross-scale characteristic data. This step aims to analyze the characteristics and changes of crop physiological restricted response at different spatial scales. Through spatial cross-scale analysis, the system can reveal the differences and commonalities of crop growth restriction in different geographical regions, and provide basic data for subsequent spatial response interpolation processing. Spatial response interpolation processing is performed on the restricted response spatial cross-scale characteristic data to obtain restricted response spatial interpolation data. This step uses interpolation technology to integrate scattered spatial data points into continuous spatial data expression. Through spatial response interpolation processing, the system can generate a smoother and continuous crop physiological restricted response distribution map, which helps to more accurately understand and describe the physiological restriction in different geographical regions. Trend cluster analysis is performed on crop physiological restricted response data according to the restricted response spatial interpolation data to obtain restricted response trend cluster data. This step clusters and classifies geographical regions with similar restricted response trends in spatial interpolation data through trend cluster analysis. By analyzing the trends and patterns of crop physiological constraint responses in different geographic regions, the system can discover potential spatial trends and correlations, and provide regionalized decision support and management recommendations for agricultural production managers.

[0038] Preferably, step S32 includes the following steps:

[0039] Step S321: performing spatial response hysteresis evaluation on the restricted response space cross-scale characteristic data to obtain spatial response hysteresis data;

[0040] Step S322: performing response progressive increment calculation on the spatial response hysteresis data to obtain response hysteresis progressive increment data;

[0041] Step S323: performing spatial response interpolation processing according to the response lag progressive increment data to obtain limited response spatial interpolation data.

[0042] The present invention evaluates the spatial response hysteresis of the restricted response spatial cross-scale characteristic data to obtain spatial response hysteresis data. This step is intended to evaluate the spatial hysteresis characteristics of the restricted response of crop physiology. By analyzing the hysteresis of the restricted response of crop growth in different geographical areas, the system can understand the time delay characteristics of the restricted response of crop growth between different regions, and provide basic data for the subsequent response progressive increment calculation. The spatial response hysteresis data is subjected to response progressive increment calculation to obtain response hysteresis progressive increment data. This step calculates the progressive increment of the spatial response, that is, the incremental change amount of the response between adjacent points in space. Through the response progressive increment calculation, the system can reveal the changing trend and incremental distribution of the restricted response of crop physiology in different geographical areas, and provide detailed data support for the subsequent spatial response interpolation processing. The spatial response interpolation processing is performed according to the response hysteresis progressive increment data to obtain the restricted response spatial interpolation data. This step uses interpolation technology to integrate the scattered response progressive increment data into a continuous spatial data expression. Through spatial response interpolation processing, the system can generate a smoother and more continuous distribution map of crop physiological limitation responses, thereby more accurately describing and analyzing the spatial distribution characteristics of crop growth limitations in different geographical areas.

[0043] Preferably, step S33 includes the following steps:

[0044] Step S331: performing spatial interpolation point similarity measurement on crop physiological restricted response data according to restricted response spatial interpolation data to obtain spatial interpolation point similarity data;

[0045] Step S332: performing non-uniformity analysis on the spatial interpolation point similarity data to obtain similarity non-uniformity data;

[0046] Step S333: performing anisotropic analysis on the spatial interpolation point similarity data according to the similarity non-uniformity data to obtain similarity anisotropic data;

[0047] Step S334: performing trend cluster analysis based on the similarity anisotropic data to obtain restricted response trend cluster data.

[0048] The present invention measures the spatial interpolation point similarity of crop physiological restricted response data according to restricted response spatial interpolation data to obtain spatial interpolation point similarity data. This step is intended to evaluate the similarity of crop physiological restricted responses between different spatial interpolation points. Through the spatial interpolation point similarity measurement, the system can quantify the similarity of crop growth restricted responses in different geographical locations, and provide basic data for subsequent non-uniformity analysis. Non-uniformity analysis is performed on the spatial interpolation point similarity data to obtain similarity non-uniformity data. This step focuses on analyzing the distribution characteristics of spatial interpolation point similarity in geographic space. Through non-uniformity analysis, the system can identify and understand the spatial distribution non-uniformity of crop physiological restricted responses in different geographical regions, that is, the change and distribution characteristics of similarity. Non-isotropic analysis is performed on the spatial interpolation point similarity data according to the similarity non-uniformity data to obtain similarity non-isotropic data. This step analyzes the non-isotropic nature of spatial interpolation point similarity, that is, the change in similarity in different directions. Through anisotropic analysis, the system can reveal the differences in spatial directions of crop physiological restricted responses in different geographical regions, providing data support for subsequent trend cluster analysis. Trend cluster analysis is performed based on similarity anisotropic data to obtain restricted response trend cluster data. In this step, spatial interpolation points with similar restricted response trends are classified and clustered through trend cluster analysis. By analyzing the trend clustering of crop physiological restricted responses in different geographical regions, the system can identify and summarize potential spatial trends and patterns, and provide regionalized decision support and management recommendations for agricultural production managers.

[0049] Preferably, step S4 comprises the following steps:

[0050] Step S41: performing clustering trend learning on the restricted response trend clustering data based on a multi-layer perceptron to obtain clustering trend learning data;

[0051] Step S42: normalizing the clustering trend learning data to obtain clustering trend learning normalized data;

[0052] Step S43: construct a crop disease assessment model based on cluster trend learning and normalized data to obtain a crop disease assessment model, and send the crop disease assessment model to a data processing cloud platform to perform crop disease risk assessment.

[0053] The present invention performs cluster trend learning on restricted response trend cluster data based on a multi-layer perceptron to obtain cluster trend learning data. This step uses a powerful neural network model, a multi-layer perceptron (MLP), to further analyze and learn the restricted response trend cluster data generated in the previous step. MLP can effectively learn and capture complex nonlinear relationships in the data, thereby extracting cluster trend characteristics of crop physiological restricted response. The cluster trend learning data is normalized to obtain cluster trend learning normalized data. In this step, the system will ensure that the cluster trend learning data has a uniform numerical range and distribution on the same scale through normalization. Normalization helps to reduce the differences between data, making the subsequent model construction and evaluation process more stable and reliable. A crop disease assessment model is constructed based on the cluster trend learning normalized data to obtain a crop disease assessment model, and the crop disease assessment model is sent to a data processing cloud platform to perform crop disease risk assessment. This step will use the cluster trend data learned in the previous step to construct a special crop disease assessment model. The model can predict and assess the disease risks that crops may face under different conditions, providing agricultural production managers with early warnings and precise disease management strategies.

[0054] Preferably, the present invention further provides a crop disease risk assessment device, comprising a data processing cloud platform, for executing the crop disease risk assessment method as described above, the crop disease risk assessment device comprising:

[0055] The abnormal characteristic state recognition module is used to collect crop growth morphology image data from the crop planting area through electronic monitoring equipment to obtain crop growth morphology images; perform abnormal state feature recognition on the crop growth morphology images to obtain crop morphology abnormal state data;

[0056] The crop physiological restriction response module is used to identify crop deformity morphology based on crop morphological abnormality data to obtain crop deformity morphology data; perform survival sensitivity analysis based on crop deformity morphology data to obtain nutrient deficiency-survival sensitivity data; perform crop physiological restriction response simulation based on nutrient deficiency-survival sensitivity data to obtain crop physiological restriction response data;

[0057] A trend cluster analysis module is used to perform trend cluster analysis on crop physiological restricted response data to obtain restricted response trend cluster data;

[0058] The assessment model building module is used to build a crop disease assessment model based on the restricted response trend clustering data using a multi-layer perceptron to obtain a crop disease assessment model; and send the crop disease assessment model to the data processing cloud platform to perform crop disease risk assessment.

[0059] The beneficial effects of the present invention are as follows: first, by collecting crop growth morphology image data through electronic monitoring equipment, real-time monitoring and recording of the crop growth process can be achieved. After these image data are identified by abnormal state features, abnormal conditions that may exist in crop growth, such as diseases, malnutrition or the influence of external environmental factors, can be effectively detected. The accurate identification of these abnormal state data provides a basis for timely and targeted management and intervention measures. The abnormal state data of crop morphology are used to identify deformed morphology, and the specific morphological characteristics of crops can be further analyzed in depth to identify deformed or abnormal growth patterns. By performing survival sensitivity analysis based on deformed morphological data, the survival ability and adaptability of crops under different morphological abnormalities can be evaluated. In particular, the physiological restricted response simulation based on nutrient deficiency-survival sensitivity data can simulate the growth response of crops under nutrient deficiency or other physiological restriction conditions. Trend cluster analysis processes the crop physiological restricted response data, and can identify and analyze the trends and patterns of crop growth responses under different physiological restriction conditions. This analysis can help agricultural experts and managers to more comprehensively understand the growth performance of crops under different environments or management conditions, and identify common trends or abnormal phenomena that may exist. A crop disease assessment model is constructed based on a multi-layer perceptron, using trend clustering data as a basis. The multi-layer perceptron is a powerful deep learning model that can learn features from complex data and accurately classify and predict data. The construction of this model can not only effectively analyze the disease risks that crops may face under different physiological constraints, but also provide predictive assessment results. The crop disease assessment model is sent to a data processing cloud platform, so that the disease risk assessment process can be carried out in a wider data context, thereby improving the accuracy and reliability of the assessment. In this way, agricultural managers can obtain detailed information about the health status of crops in a timely manner, so as to take appropriate management measures to reduce the impact of diseases on yield and quality. Therefore, the present invention is an optimization process made to a traditional crop disease risk assessment method, which solves the problems of low risk assessment accuracy and inaccurate trend analysis in a traditional crop disease risk assessment method, and improves the accuracy of risk assessment and trend analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A schematic diagram of the steps of a crop disease risk assessment method;

[0061] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0062] Figure 3 for Figure 1Detailed implementation steps of step S3 in FIG.

[0063] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0064] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0065] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0066] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0067] To achieve this, please refer to Figures 1 to 3 , a crop disease risk assessment method, the method comprising the following steps:

[0068] Step S1: collecting crop growth morphology image data in the crop planting area through electronic monitoring equipment to obtain crop growth morphology images; performing abnormal state feature recognition on the crop growth morphology images to obtain crop morphology abnormal state data;

[0069] Step S2: performing crop deformity morphology recognition on the crop morphology abnormal state data to obtain crop deformity morphology data; performing survival sensitivity analysis based on the crop deformity morphology data to obtain nutrient deficiency-survival sensitivity data; performing crop physiological restriction response simulation based on the nutrient deficiency-survival sensitivity data to obtain crop physiological restriction response data;

[0070] Step S3: performing trend cluster analysis on the crop physiological restricted response data to obtain restricted response trend cluster data;

[0071] Step S4: constructing a crop disease assessment model based on the restricted response trend clustering data based on the multi-layer perceptron to obtain a crop disease assessment model; and sending the crop disease assessment model to the data processing cloud platform to perform crop disease risk assessment.

[0072] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of a crop disease risk assessment method of the present invention. In this example, the crop disease risk assessment method includes the following steps:

[0073] Step S1: collecting crop growth morphology image data in the crop planting area through electronic monitoring equipment to obtain crop growth morphology images; performing abnormal state feature recognition on the crop growth morphology images to obtain crop morphology abnormal state data;

[0074] In an embodiment of the present invention, in a crop planting area, electronic monitoring devices are arranged at multiple locations to cover the crop growth area in all directions. These electronic monitoring devices are equipped with high-resolution cameras that can regularly capture images of the growth morphology of crops. These image data are transmitted to a central data processing system via a network for storage and processing. The processing system first pre-processes the image data, including denoising, color correction, and image enhancement, to improve image quality. Then, an image recognition algorithm based on deep learning is used to automatically detect and identify abnormal states of crop growth morphology. This algorithm learns the characteristics of normal and abnormal growth morphology through a large number of labeled samples in the training data set, thereby accurately identifying crops with abnormal morphology. The data of these abnormal states are recorded and summarized into crop morphology abnormal state data, providing a basis for further analysis.

[0075] Step S2: performing crop deformity morphology recognition on the crop morphology abnormal state data to obtain crop deformity morphology data; performing survival sensitivity analysis based on the crop deformity morphology data to obtain nutrient deficiency-survival sensitivity data; performing crop physiological restriction response simulation based on the nutrient deficiency-survival sensitivity data to obtain crop physiological restriction response data;

[0076] In an embodiment of the present invention, after obtaining the data of abnormal state of crop morphology, the next task is to perform detailed deformed morphological recognition on these abnormal states. This process uses advanced computer vision technology, especially morphological analysis and feature extraction algorithms, to further classify and identify deformed morphology. By comparing the abnormal morphology with the database of deformed morphology, it is determined which morphology belongs to the deformed category. The identified deformed morphological data is then used as input for survival sensitivity analysis. In this analysis process, multivariate statistical analysis technology is used to combine environmental factors (such as soil quality, climatic conditions) and nutritional conditions to evaluate the impact of deformed morphology on crop survival. By establishing a mathematical model, the impact of nutrient deficiency on survival sensitivity is calculated, and finally nutrient deficiency-survival sensitivity data is obtained. These data show the sensitivity of different nutrient deficiency conditions to crop growth, providing a key basis for subsequent physiological restricted response simulation. Using the obtained nutrient deficiency-survival sensitivity data, crop physiological restricted response simulation is performed. This process uses simulation methods in systems biology and computational biology. First, a mathematical model of crop physiological response is constructed, which is based on the known relationship between physiological mechanism and response path. Then, the nutrient deficiency-survival sensitivity data is input into the model using simulation software, and the physiological responses of crops under different nutrient deficiency conditions are calculated through dynamic simulation. This simulation process usually requires the use of high-performance computing resources to handle complex models and a large number of computing tasks. Finally, the crop physiological restriction response data are extracted from the simulation results. These data include the changes in various physiological indicators under different nutrient deficiency scenarios, providing a scientific basis for further research and decision-making.

[0077] Step S3: performing trend cluster analysis on the crop physiological restricted response data to obtain restricted response trend cluster data;

[0078] In an embodiment of the present invention, after processing the crop physiological restriction response data, trend cluster analysis is performed to identify potential patterns and trends in the data. First, the crop physiological restriction response data is input into a data analysis software, which should have powerful statistical analysis and clustering functions. By selecting a suitable clustering algorithm, such as K-means clustering or hierarchical clustering, appropriate clustering parameters, such as the number of clusters or similarity thresholds, are set. Data preprocessing includes standardization and normalization to eliminate the influence of different feature scales on the analysis results. During the processing, the algorithm groups the physiological restriction response data and classifies the response data with similar trends into the same category. Data visualization tools used in this process, such as scatter plots or heat maps, help to intuitively display the results of different clusters. Finally, the obtained restricted response trend clustering data shows the response patterns under different physiological restriction states, which helps to reveal potential disease patterns and risks.

[0079] Step S4: constructing a crop disease assessment model based on the restricted response trend clustering data based on the multi-layer perceptron to obtain a crop disease assessment model; and sending the crop disease assessment model to the data processing cloud platform to perform crop disease risk assessment.

[0080] In an embodiment of the present invention, after obtaining the restricted response trend clustering data, a multi-layer perceptron (MLP) model is used to construct a crop disease assessment model. First, a suitable machine learning framework, such as TensorFlow or PyTorch, is selected to build a multi-layer perceptron model. The input layer of the model accepts the restricted response trend clustering data, the number of nodes and the activation function (such as ReLU) of the hidden layer are adjusted according to the specific data and model requirements, and the output layer predicts the risk level of crop diseases. During the model training process, a training set containing label data is required to optimize the parameters of the model, and the generalization ability of the model is ensured by methods such as cross-validation. After completing the model training, the constructed crop disease assessment model is saved as a standard format file (such as .h5 or .pth). The file is then uploaded to the cloud through the interface of the data processing cloud platform, and the cloud platform performs the disease risk assessment task and returns the assessment results in the form of a report. The integration and upload steps of the data processing cloud platform generally include interface calls, data transmission, and result verification to ensure the smooth progress of the assessment process and the accuracy of the results.

[0081] Preferably, step S1 comprises the following steps:

[0082] Step S11: collecting crop growth morphology image data in the crop planting area through electronic monitoring equipment to obtain crop growth morphology images;

[0083] Step S12: performing imaging angle feature analysis on the crop growth morphology image to obtain crop image imaging angle data;

[0084] Step S13: performing abnormal state feature recognition on the crop growth morphology image according to the crop image imaging angle data to obtain crop morphology abnormal state data.

[0085] In an embodiment of the present invention, electronic monitoring equipment is arranged in the crop planting area to achieve comprehensive coverage. The selected equipment includes a high-resolution digital camera or a video camera that can stably shoot under different lighting and environmental conditions. The electronic monitoring equipment is equipped with an automatic timed shooting function to regularly capture image data of the growth status of crops. During the image acquisition process, a multi-angle and multi-position equipment layout is adopted to obtain a comprehensive growth morphological image. These image data are transmitted to a central data storage system in real time via a wired or wireless network to ensure the integrity and timeliness of the image. In the data storage system, the image data is organized and saved in a standard format (such as JPEG or PNG) for subsequent processing and analysis. After obtaining the crop growth morphological image, the imaging angle feature analysis is performed. This process uses image processing software and algorithms, such as computer vision tools in OpenCV or MATLAB. First, each image is geometrically corrected to correct the image distortion caused by the shooting angle and the device position. Next, a feature extraction algorithm (such as Harris corner detection or SIFT) is used to analyze the imaging angle features in the image. These features can include the shooting angle, the distortion of the viewing angle, and the perspective effect in the image. Through algorithm processing, the extracted imaging angle data includes the viewing angle information of each image and a detailed description of the shooting angle. These data will be stored in a structured database for subsequent query and analysis. Based on the obtained crop image imaging angle data, the abnormal state features of the crop growth morphology images are identified. This process uses machine learning and deep learning techniques, especially convolutional neural networks (CNNs) for image classification. First, the image data and imaging angle data are combined, and feature extraction and classification are performed through a pre-trained CNN model (such as ResNet or VGG). This model has been trained with a large number of labeled samples and can identify normal and abnormal growth forms. During the processing, the system compares the abnormal state with the normal state and identifies the abnormal features of the crop morphology. Finally, the generated crop morphology abnormal state data includes the type of abnormality, severity and its location in the image. These data will be organized and provided for further analysis steps.

[0086] Preferably, step S2 comprises the following steps:

[0087] Step S21: performing crop deformity morphology recognition on the crop morphology abnormal state data to obtain crop deformity morphology data;

[0088] Step S22: evaluating the nutritional deficiency symptoms according to the crop deformity morphology data to obtain the nutritional deficiency symptom evaluation data of the crop;

[0089] Step S23: performing survival sensitivity analysis on the crop nutrition deficiency symptom assessment data to obtain nutrition deficiency-survival sensitivity data;

[0090] Step S24: simulating crop physiological limitation response of crop nutrient deficiency symptom assessment data according to nutrient deficiency-survival sensitivity data to obtain crop physiological limitation response data.

[0091] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0092] Step S21: performing crop deformity morphology recognition on the crop morphology abnormal state data to obtain crop deformity morphology data;

[0093] In an embodiment of the present invention, after the crop morphology abnormal state data is obtained, the crop deformity morphology is identified. This process relies on advanced image processing and pattern recognition technology. First, the abnormal state data is input into a computer vision system for deformity morphology recognition, which is usually equipped with image analysis and machine learning tools, such as deep learning frameworks in TensorFlow or PyTorch. Use a pre-trained convolutional neural network (CNN), such as Inception or DenseNet, for feature extraction and classification. During the training process, these network models are trained with labeled deformity morphology images so that they can recognize and classify deformity states. In actual operation, the system analyzes each image in the abnormal state data and identifies the characteristics of the deformity, such as morphological distortion, deformation, or abnormal growth pattern. The image processing steps include feature extraction, pattern matching, and anomaly detection. The result will generate crop deformity morphology data, which contains a detailed description of each identified deformity, including the type of deformity, the location of occurrence, and the possible severity. The recognition results will be integrated and stored in a database for further analysis and processing.

[0094] Step S22: evaluating the nutritional deficiency symptoms according to the crop deformity morphology data to obtain the nutritional deficiency symptom evaluation data of the crop;

[0095] In an embodiment of the present invention, after obtaining the crop deformity morphology data, a nutrient deficiency symptom assessment is performed. This step uses data analysis and statistical methods, combined with knowledge in the field of agricultural science to assess the impact of nutrient deficiency. First, a nutrient deficiency symptom assessment model is constructed, which can be based on previous research and the knowledge of agricultural experts, such as through a statistical regression model or an expert system. The specific operation includes comparing the deformity morphology data with a known nutrient deficiency symptom database. Using analysis tools, such as the SciPy library in R or Python, calculate the correlation between various types of nutrient deficiency (such as nitrogen, phosphorus, and potassium deficiency) and the identified deformity morphology. This step may require combining image data with soil and plant sample data collected in the field for comprehensive analysis. During the assessment process, the data analysis software will apply statistical analysis methods (such as principal component analysis or factor analysis) to determine the relationship between deformity morphology and nutrient deficiency. The crop nutrient deficiency symptom assessment data generated by the analysis results records in detail the specific symptoms, severity, and association with deformity morphology of each nutrient deficiency. The assessment data will be integrated and stored to support subsequent research or decision-making processes.

[0096] Step S23: performing survival sensitivity analysis on the crop nutrition deficiency symptom assessment data to obtain nutrition deficiency-survival sensitivity data;

[0097] In an embodiment of the present invention, the survival sensitivity analysis of the crop nutrition deficiency symptom assessment data is performed. First, a survival sensitivity analysis model is constructed, which usually involves a combination of ecology and statistics. Statistical analysis tools, such as Pandas and Statsmodels libraries in R or Python, are used to process the assessment data. The survival sensitivity index is calculated by comparing the crop growth data under different nutrition deficiency conditions. The specific operation includes inputting the nutrition deficiency symptom assessment data into the survival analysis model, and using methods such as the Cox proportional hazard model or the Kaplan-Meier survival curve to analyze the impact of different nutrition deficiency conditions on the crop survival rate. The input data of the model includes various types of nutrition deficiency symptoms and their influencing factors on crop growth. Through these model calculations, the sensitivity of nutrition deficiency to crop survival is identified. The generated nutrition deficiency-survival sensitivity data contains the relationship between the degree of nutrition deficiency and the survival ability of the crop, as well as the vulnerability of crop growth under various nutrition deficiency conditions. These data help to further understand and explain the survival performance of crops under different nutrition deficiency conditions.

[0098] Step S24: simulating crop physiological limitation response of crop nutrient deficiency symptom assessment data according to nutrient deficiency-survival sensitivity data to obtain crop physiological limitation response data.

[0099] In an embodiment of the present invention, after obtaining the nutrient deficiency-survival sensitive data, a crop physiological restricted response simulation is performed. This step uses system biology and computational models to simulate the physiological response of crops under nutrient deficiency conditions. First, a physiological restricted response model is established, which is usually based on a mathematical description of biological mechanisms. A modeling tool, such as Simulink of MATLAB or the SciPy library in Python, is selected to build and simulate the model. Model building includes setting input parameters, such as nutrient deficiency-survival sensitive data, and defining variables of physiological responses (such as growth rate, leaf area, root development, etc.). Through the constructed model, dynamic simulation is performed to simulate the physiological response of crops under different nutrient deficiency conditions. The simulation process involves adjusting model parameters and running multiple simulation scenarios to observe the response changes of crops under various nutrient deficiency conditions. The simulation results generate crop physiological restricted response data, which record in detail the changes in physiological indicators of crops under different conditions. These data include the intensity, change trend and possible limiting factors of the physiological response, providing a scientific basis for subsequent optimization measures.

[0100] Preferably, step S23 includes the following steps:

[0101] Step S231: performing nutrient element deficiency identification on the crop nutrient element deficiency symptom assessment data to obtain crop nutrient element deficiency data;

[0102] Step S232: performing an analysis of the environmental adaptability impairment of crops based on the crop nutrient element deficiency data to obtain element deficiency environmental adaptability impairment data;

[0103] Step S233: calculating the impairment difference of the element-missing environment adaptability impairment data to obtain the environment adaptability impairment difference data;

[0104] Step S234: Perform survival sensitivity analysis based on the environmental adaptability impairment difference data to obtain nutritional deficiency-environmental survival sensitivity data.

[0105] In an embodiment of the present invention, the nutrient deficiency identification is performed on the crop nutrient deficiency symptom assessment data. This process involves the application of data analysis technology to determine the deficiency of various nutrient elements. First, the nutrient deficiency symptom assessment data is input into the data analysis software, such as the Pandas library in R or Python. Using predefined nutrient deficiency standards and algorithms (such as logistic regression or decision trees), the relationship between the symptoms in the assessment data and the deficiency of specific nutrient elements is analyzed. This analysis step includes feature selection and classification model training, and the model is used to identify which nutrient element deficiency may cause symptoms. The final crop nutrient element deficiency data records the deficiency of each element and its specific impact on the crop, providing basic data for subsequent environmental adaptation ability analysis. Environmental adaptation ability impairment analysis is performed based on crop nutrient element deficiency data. This step mainly involves evaluating the impact of nutrient element deficiency on the ability of crops to adapt to the environment. First, the nutrient element deficiency data is input into the environmental adaptation analysis model, which may include a statistical regression model or a multivariate analysis model. Use the Statsmodels library in R or Python to analyze the relationship between the adaptability of crops under different environmental conditions and the missing elements. The model calculates the degree of impact of nutrient deficiency on the adaptability of crops under specific environmental factors (such as soil quality and climatic conditions). The resulting element deficiency environmental adaptability impairment data records the specific impairment of crop environmental adaptability under various nutrient deficiency conditions, providing data support for the next step of impairment difference calculation. The element deficiency environmental adaptability impairment data is used to calculate the impairment difference. This step uses data analysis and statistical calculation methods to compare the differences in crop environmental adaptability under different nutrient deficiency conditions. Use data analysis tools, such as NumPy or SciPy libraries in Python, to calculate the differences between different environmental adaptability impairment data. Specific operations include calculating the mean, standard deviation, and other statistical indicators of the impairment to evaluate the differences in the impact of different element deficiencies on environmental adaptability. This process may also include using hypothesis testing methods (such as t-tests or variance analysis) to determine whether the impairment difference is significant. The final environmental adaptability impairment difference data records in detail the differences in adaptability under different nutrient deficiency conditions, providing a detailed basis for the subsequent survival sensitivity analysis. Survival sensitivity analysis is performed based on the environmental adaptability impairment difference data. This step involves applying the impairment difference data to the survival sensitivity model to assess the survival of crops under the condition of environmental adaptation impairment. First, a survival sensitivity analysis model is constructed, which may include a multivariate regression model or a survival analysis model, using tools such as the survival package in R or the lifelines library in Python. The environmental adaptation impairment difference data is used as input to analyze the survival sensitivity of crops under different impairment levels.The model calculates and evaluates the impact of reduced environmental adaptability on crop survival and derives specific survival sensitivity indicators. The resulting nutrient deficiency-environmental survival sensitivity data records in detail the survival sensitivity of crops under different conditions of reduced environmental adaptability, providing an in-depth understanding of the survival risks of crops under various nutrient deficiency conditions.

[0106] Preferably, step S24 comprises the following steps:

[0107] Step S241: identifying key factors of the crop nutrition deficiency symptom assessment data according to the nutrition deficiency-survival sensitivity data to obtain nutrition deficiency key factor data;

[0108] Step S242: performing factor association analysis on the key factor data of nutritional deficiency to obtain nutritional deficiency factor association data;

[0109] Step S243: performing multi-level data decomposition according to the nutritional deficiency factor correlation data to obtain nutritional deficiency multi-level factor correlation data;

[0110] Step S244: simulating crop physiological restriction response according to the multi-level factor correlation data of nutrient deficiency to obtain crop physiological restriction response data.

[0111] In an embodiment of the present invention, the key factors of the crop nutrient deficiency symptom assessment data are identified based on the nutrient deficiency-survival sensitive data. This step involves applying data mining and feature selection techniques to identify nutrient deficiency factors that have a significant impact on crop physiological responses. First, the nutrient deficiency-survival sensitive data and the nutrient deficiency symptom assessment data are input into data analysis software, such as Scikit-learn in Python or the Caret package in R. Using feature selection algorithms, such as feature importance scores or LASSO regression in random forests, key factors that play a major role in crop physiological responses are identified. These algorithms determine key factors by calculating the contribution of each factor to model prediction. The nutrient deficiency key factor data finally obtained lists the main nutrients and their deficiency conditions that affect crop physiological restricted responses, providing basic data for subsequent analysis. Factor association analysis is performed on the identified nutrient deficiency key factor data. This step is intended to reveal the mutual relationship between key factors and the comprehensive impact on crop physiological responses. Use association rule mining techniques, such as the Apriori algorithm or the FP-Growth algorithm, to analyze the relationship between key factors. Data analysis tools such as the mlxtend library in Python or the arules package in R can be used to implement these algorithms. By calculating the support, confidence and lift between key factors, significant association rules are identified. The resulting nutrient deficiency factor association data show the relationship patterns between key factors, such as which factors are often missing together and their combined effects on crop physiological responses. These data help understand the interaction of nutrient deficiency factors and their impact on crop health. Multi-level data decomposition is performed based on the nutrient deficiency factor association data. This process involves decomposing the association data into multiple levels to more deeply analyze the factor relationships at each level. First, the association data is reduced in dimension and the main components are extracted using data decomposition methods such as principal component analysis (PCA) or factor analysis. PCA can be implemented using the Scikit-learn library in Python or the prcomp function in R. The decomposed data will be further refined to form a multi-level factor association data structure. Each level contains different degrees of factor association information, which helps identify the performance and impact of key factors in different scenarios. The resulting nutrient deficiency multi-level factor association data shows the hierarchical relationship between factors and reveals the impact of key factors on crop physiological responses at different levels. Crop physiological restricted response simulation is performed based on the nutrient deficiency multi-level factor association data. This step uses biological models and computational tools to simulate the physiological responses of crops under different factor association conditions. First, select a suitable simulation tool, such as Simulink in MATLAB or SciPy library in Python, to build a physiological response model. The model needs to set input variables and parameters based on multi-level factor association data.The simulation process includes defining the initial conditions of the model, running different simulation scenarios, and adjusting the missing conditions of key factors. Through simulation calculations, the effects of different combinations of factors on crop physiological responses are analyzed, and the generated results include changes in various physiological indicators. The final crop physiological restricted response data records the physiological state of crops under different nutrient deficiency scenarios and provides detailed physiological response simulation results to support further research and decision-making.

[0112] Preferably, step S3 comprises the following steps:

[0113] Step S31: performing spatial cross-scale analysis on crop physiological restricted response data to obtain restricted response spatial cross-scale characteristic data;

[0114] Step S32: performing spatial response interpolation processing on the restricted response space cross-scale characteristic data to obtain restricted response space interpolation data;

[0115] Step S33: performing trend cluster analysis on the crop physiological restricted response data according to the restricted response space interpolation data to obtain restricted response trend cluster data.

[0116] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0117] Step S31: performing spatial cross-scale analysis on crop physiological restricted response data to obtain restricted response spatial cross-scale characteristic data;

[0118] In an embodiment of the present invention, spatial cross-scale analysis is performed on crop physiological restricted response data. This step involves analyzing the changing characteristics of crop physiological responses at different spatial scales. First, the physiological restricted response data is imported into a geographic information system (GIS) or spatial data analysis software, such as ArcGIS or QGIS. Spatial analysis tools, such as spatial statistical analysis functions or geographically weighted regression (GWR), are used to analyze the changing characteristics of the data at different spatial scales. Spatial cross-scale analysis includes the following steps: visualization of the spatial distribution of data, selection of scales (such as regional scale, local scale, etc.), and evaluation of the impact of spatial scales on physiological response data. By refining and integrating spatial scales, the characteristics and trends of crop physiological responses at different scales are determined. The generated restricted response spatial cross-scale characteristic data describes the characteristics and patterns of crop physiological responses at different spatial scales, providing basic data for further analysis.

[0119] Step S32: performing spatial response interpolation processing on the restricted response space cross-scale characteristic data to obtain restricted response space interpolation data;

[0120] In an embodiment of the present invention, spatial response interpolation processing is performed on the restricted response spatial cross-scale characteristic data. This step is intended to perform spatial filling of the spatial cross-scale characteristic data through interpolation technology to generate more detailed spatial response data. Use interpolation tools and algorithms, such as Kriging interpolation (Kriging) or inverse distance weighted interpolation (IDW), to process spatial characteristic data. Select appropriate tools, such as the Spatial Analyst tool in ArcGIS or the SciPy library in Python, to perform interpolation calculations. The specific operations include: selecting an interpolation method, setting interpolation parameters (such as range and weight), and performing interpolation processing to generate continuous spatial response data. The restricted response space interpolation data generated by the interpolated data provides more detailed spatial distribution information, fills the blank areas in the original data, and makes the analysis of spatial response characteristics more comprehensive.

[0121] Step S33: performing trend cluster analysis on the crop physiological restricted response data according to the restricted response space interpolation data to obtain restricted response trend cluster data.

[0122] In an embodiment of the present invention, trend cluster analysis is performed on crop physiological restricted response data according to restricted response spatial interpolation data. This step is intended to identify the trend pattern of crop physiological response. First, the interpolated data is input into data analysis software, such as the Scikit-learn library in MATLAB or Python. Select a suitable trend clustering algorithm, such as K-means clustering or hierarchical clustering, for analysis. Set the parameters of the algorithm, such as the number of clusters or the distance measurement method, to perform model training and clustering calculations. During the analysis process, the algorithm will group similar response data according to the trend pattern identified in the spatial interpolation data. Different clustering results and their spatial distribution are displayed through data visualization tools (such as scatter plots, heat maps or three-dimensional graphs). The finally obtained restricted response trend clustering data includes different trend patterns and spatial distributions of crop physiological responses, providing a basis for further decision-making and optimization.

[0123] Preferably, step S32 includes the following steps:

[0124] Step S321: performing spatial response hysteresis evaluation on the restricted response space cross-scale characteristic data to obtain spatial response hysteresis data;

[0125] Step S322: performing response progressive increment calculation on the spatial response hysteresis data to obtain response hysteresis progressive increment data;

[0126] Step S323: performing spatial response interpolation processing according to the response lag progressive increment data to obtain limited response spatial interpolation data.

[0127] In an embodiment of the present invention, the spatial response hysteresis evaluation is performed on the restricted response spatial cross-scale characteristic data. This step is intended to analyze the hysteresis effect of the spatial response characteristics changing with time or space. First, the spatial cross-scale characteristic data is imported into the spatial statistical analysis software, such as ArcGIS or the Geopandas library in Python. The data is evaluated for hysteresis using spatial lag analysis tools, such as the spatial lag regression model (Spatial Lag Model) or the spatial error model (Spatial Error Model). The specific operation includes: selecting an appropriate lag analysis model, setting lag parameters (such as lag distance or time interval), and performing spatial lag analysis. The lag evaluation result generates spatial response lag data, which describes the lag characteristics of the spatial response at different time or spatial scales, such as the intensity and range of the lag effect, and provides data support for subsequent progressive increment calculations. The spatial response lag data is subjected to response progressive increment calculation. This step involves calculating the progressive change amount of the lag data at different time or spatial scales. First, import the hysteresis data into data analysis software, such as the NumPy library in Python or the dplyr package in R. Use incremental calculation methods, such as differential analysis or incremental regression, to calculate the change in each spatial unit on the lag effect. The specific operations include: setting the parameters of incremental calculation, such as time interval or spatial distance, and performing incremental calculation. The calculation results generate response lag incremental data, which records the change and incremental trend of spatial response under different lag effects. These data help to understand the gradual change of lag effect and provide detailed spatial change information for subsequent interpolation processing. Perform spatial response interpolation processing based on response lag incremental data. This step uses interpolation technology to fill the incremental data in space to generate more detailed spatial response data. First, select a suitable interpolation tool, such as Kriging interpolation (Kriging) or inverse distance weighted interpolation (IDW), and perform interpolation processing in spatial analysis software. Tools such as ArcGIS's Spatial Analyst module or the SciPy library in Python provide the necessary interpolation algorithms. The specific operations include: importing response lag progressive incremental data, setting interpolation parameters (such as interpolation method, range and weight), and performing spatial interpolation calculations. The restricted response spatial interpolation data generated by the interpolation result is the refined spatial response feature data, which fills the blank areas in the original data, making the spatial response features more continuous and comprehensive, and providing support for further data analysis and decision-making.

[0128] Preferably, step S33 includes the following steps:

[0129] Step S331: performing spatial interpolation point similarity measurement on crop physiological restricted response data according to restricted response spatial interpolation data to obtain spatial interpolation point similarity data;

[0130] Step S332: performing non-uniformity analysis on the spatial interpolation point similarity data to obtain similarity non-uniformity data;

[0131] Step S333: performing anisotropic analysis on the spatial interpolation point similarity data according to the similarity non-uniformity data to obtain similarity anisotropic data;

[0132] Step S334: performing trend cluster analysis based on the similarity anisotropic data to obtain restricted response trend cluster data.

[0133] In an embodiment of the present invention, spatial interpolation point similarity measurement is performed on crop physiological restricted response data according to restricted response spatial interpolation data. This step is intended to quantify the similarity between different spatial interpolation points to evaluate the consistency of data in space. First, the spatial interpolation data is imported into data analysis software, such as the SciPy library in Python or the sp package in R. The similarity between each pair of spatial interpolation points is calculated using a similarity measurement method, such as cosine similarity or Euclidean distance. The specific operations include: defining the standards and parameters for similarity measurement, performing similarity calculations and generating a similarity matrix. The similarity matrix records the similarity between each spatial interpolation point, and the generated spatial interpolation point similarity data shows the similarity of spatial response characteristics at different locations, providing basic data for subsequent analysis. The spatial interpolation point similarity data is subjected to non-uniformity analysis. This step is intended to identify non-uniform distribution features in spatial similarity data. First, the similarity data is input into a statistical analysis tool, such as the Statsmodels library in Python or the ggplot2 package in R. The distribution characteristics of spatial similarity data are evaluated using heterogeneity analysis methods, such as local autocorrelation analysis (LISA) or coefficient of variation calculation. The specific operations include: selecting an analysis method, setting parameters (such as analysis window size or calculation range), and performing heterogeneity analysis. The similarity heterogeneity data generated by the analysis results record the heterogeneous distribution characteristics in the spatial similarity data, such as hot spots or cold spots, which helps to understand the distribution characteristics of spatial responses. Anisotropic analysis is performed on the spatial interpolation point similarity data based on the similarity heterogeneity data. This step is used to evaluate the variation differences of spatial similarity data in different directions. First, the similarity heterogeneity data is imported into spatial analysis software, such as ArcGIS or the Geopandas library in Python. Anisotropic analysis tools, such as directional variability analysis (Directional Variability Analysis) or structural equation modeling, are used to calculate the response differences of the data in different directions. The specific operations include: selecting an appropriate anisotropic analysis method, setting direction and scale parameters, and performing anisotropic analysis. The similarity anisotropy data generated by the analysis results describe the variation characteristics of spatial similarity in different directions, such as in which directions the similarity changes more significantly, and provide detailed spatial feature data for trend clustering analysis. Trend clustering analysis is performed based on similarity anisotropy data. This step aims to identify and classify trend patterns of different spatial interpolation points. First, input the similarity anisotropy data into a clustering analysis tool, such as the Statistics and Machine Learning Toolbox in MATLAB or the Scikit-learn library in Python. Select an appropriate clustering algorithm, such as K-means clustering or DBSCAN (density-based spatial clustering), to perform trend clustering analysis. Set algorithm parameters (such as the number of clusters or density threshold) and perform clustering calculations.The clustering results are displayed through data visualization tools (such as scatter plots and heat maps) to show different trend patterns. The resulting restricted response trend clustering data records the trend patterns and spatial distribution of different spatial interpolation points, providing detailed spatial response pattern information for further crop disease risk assessment.

[0134] Preferably, step S4 comprises the following steps:

[0135] Step S41: performing clustering trend learning on the restricted response trend clustering data based on a multi-layer perceptron to obtain clustering trend learning data;

[0136] Step S42: normalizing the clustering trend learning data to obtain clustering trend learning normalized data;

[0137] Step S43: construct a crop disease assessment model based on cluster trend learning and normalized data to obtain a crop disease assessment model, and send the crop disease assessment model to a data processing cloud platform to perform crop disease risk assessment.

[0138] In an embodiment of the present invention, clustering trend learning is performed on restricted response trend clustering data based on a multi-layer perceptron to obtain clustering trend learning data. First, the restricted response trend clustering data is imported into a deep learning framework, such as TensorFlow or PyTorch. A multi-layer perceptron (MLP) model is constructed, which includes an input layer, multiple hidden layers, and an output layer. The input layer accepts clustering trend data, the hidden layer is used to extract complex patterns in the data, and the output layer generates prediction results of clustering trends. The training process of the model includes defining a loss function (such as mean square error), selecting an optimization algorithm (such as Adam or SGD), and training to optimize model parameters. A cross-validation method is used during the training process to ensure the generalization ability of the model. After the training is completed, the obtained clustering trend learning data describes the patterns and characteristics of different clustering trends, which provides basic data for the construction of a crop disease assessment model. The clustering trend learning data is normalized to obtain clustering trend learning normalized data. This step is intended to convert the data into a unified scale for subsequent model construction. First, select an appropriate normalization method, such as Min-Max normalization or Z-score normalization. Use data processing tools, such as the Scikit-learn library in Python or the caret package in R, to apply the normalization algorithm. The specific operations include: loading the clustering trend learning data, selecting a normalization method (such as scaling the data to the [0,1] interval), and applying the normalization function. The normalized data ensures that each feature is on the same scale, eliminating the impact caused by data scale differences. The generated clustering trend learning normalized data provides consistent data input for subsequent model building. Build a crop disease assessment model based on the clustering trend learning normalized data. This step includes building an assessment model using the normalized data and deploying the model to a data processing cloud platform. First, select an appropriate model building tool, such as the Scikit-learn library in Python or the Statistics and Machine Learning Toolbox in MATLAB. Build a disease assessment model, which may include a regression model (such as linear regression, support vector regression) or a classification model (such as logistic regression, decision tree). Use clustering trend learning to normalize data to train the model, set model parameters and optimize performance (such as using grid search to tune parameters). After training, save the final model in a standard format (such as a .pickle file or a .h5 file) and upload it to the data processing cloud platform. Integrate the model with the cloud platform through the API interface, and perform crop disease risk assessment on the cloud platform. The output of the model includes prediction results and assessment reports on crop disease risks to help with risk management and decision-making.

[0139] Preferably, the present invention further provides a crop disease risk assessment device, comprising a data processing cloud platform, for executing the crop disease risk assessment method according to claim 1, the crop disease risk assessment device comprising:

[0140] The abnormal characteristic state recognition module is used to collect crop growth morphology image data from the crop planting area through electronic monitoring equipment to obtain crop growth morphology images; perform abnormal state feature recognition on the crop growth morphology images to obtain crop morphology abnormal state data;

[0141] The crop physiological restriction response module is used to identify crop deformity morphology based on crop morphological abnormality data to obtain crop deformity morphology data; perform survival sensitivity analysis based on crop deformity morphology data to obtain nutrient deficiency-survival sensitivity data; perform crop physiological restriction response simulation based on nutrient deficiency-survival sensitivity data to obtain crop physiological restriction response data;

[0142] A trend cluster analysis module is used to perform trend cluster analysis on crop physiological restricted response data to obtain restricted response trend cluster data;

[0143] The assessment model building module is used to build a crop disease assessment model based on the restricted response trend clustering data using a multi-layer perceptron to obtain a crop disease assessment model; and send the crop disease assessment model to the data processing cloud platform to perform crop disease risk assessment.

[0144] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0145] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for assessing crop disease risk, It is characterized in that The following steps are involved: Step S1: collecting crop growth morphology image data in the crop planting area through electronic monitoring equipment to obtain crop growth morphology images; performing abnormal state feature recognition on the crop growth morphology images to obtain crop morphology abnormal state data; Step S2: performing crop deformity morphology recognition on the crop morphology abnormal state data to obtain crop deformity morphology data; performing survival sensitivity analysis based on the crop deformity morphology data to obtain nutrient deficiency-survival sensitivity data; Simulate crop physiological restriction response based on nutrient deficiency-survival sensitivity data to obtain crop physiological restriction response data; Step S3: performing trend cluster analysis on the crop physiological restricted response data to obtain restricted response trend cluster data; Step S4: constructing a crop disease assessment model based on the restricted response trend clustering data based on the multi-layer perceptron to obtain a crop disease assessment model; and sending the crop disease assessment model to the data processing cloud platform to perform crop disease risk assessment.

2. The crop disease risk assessment method according to claim 1, It is characterized in that Step S1 includes the following steps: Step S11: collecting crop growth morphology image data in the crop planting area through electronic monitoring equipment to obtain crop growth morphology images; Step S12: performing imaging angle feature analysis on the crop growth morphology image to obtain crop image imaging angle data; Step S13: performing abnormal state feature recognition on the crop growth morphology image according to the crop image imaging angle data to obtain crop morphology abnormal state data.

3. The crop disease risk assessment method according to claim 2, It is characterized in that Step S2 includes the following steps: Step S21: performing crop deformity morphology recognition on the crop morphology abnormal state data to obtain crop deformity morphology data; Step S22: evaluating the nutritional deficiency symptoms according to the crop deformity morphology data to obtain the nutritional deficiency symptom evaluation data of the crop; Step S23: performing survival sensitivity analysis on the crop nutrition deficiency symptom assessment data to obtain nutrition deficiency-survival sensitivity data; Step S24: simulating crop physiological limitation response of crop nutrient deficiency symptom assessment data according to nutrient deficiency-survival sensitivity data to obtain crop physiological limitation response data.

4. The crop disease risk assessment method according to claim 3, It is characterized in that Step S23 includes the following steps: Step S231: performing nutrient element deficiency identification on the crop nutrient element deficiency symptom assessment data to obtain crop nutrient element deficiency data; Step S232: performing an analysis of the environmental adaptability impairment of crops based on the crop nutrient element deficiency data to obtain element deficiency environmental adaptability impairment data; Step S233: calculating the impairment difference of the element-missing environment adaptability impairment data to obtain the environment adaptability impairment difference data; Step S234: Perform survival sensitivity analysis based on the environmental adaptability impairment difference data to obtain nutritional deficiency-environmental survival sensitivity data.

5. The crop disease risk assessment method according to claim 3, It is characterized in that Step S24 includes the following steps: Step S241: identifying key factors of the crop nutrition deficiency symptom assessment data according to the nutrition deficiency-survival sensitivity data to obtain nutrition deficiency key factor data; Step S242: performing factor association analysis on the key factor data of nutritional deficiency to obtain nutritional deficiency factor association data; Step S243: performing multi-level data decomposition according to the nutritional deficiency factor correlation data to obtain nutritional deficiency multi-level factor correlation data; Step S244: simulating crop physiological restriction response according to the multi-level factor correlation data of nutrient deficiency to obtain crop physiological restriction response data.

6. The crop disease risk assessment method according to claim 1, It is characterized in that Step S3 includes the following steps: Step S31: performing spatial cross-scale analysis on crop physiological restricted response data to obtain restricted response spatial cross-scale characteristic data; Step S32: performing spatial response interpolation processing on the restricted response space cross-scale characteristic data to obtain restricted response space interpolation data; Step S33: performing trend cluster analysis on the crop physiological restricted response data according to the restricted response space interpolation data to obtain restricted response trend cluster data.

7. The crop disease risk assessment method according to claim 6, It is characterized in that Step S32 includes the following steps: Step S321: performing spatial response hysteresis evaluation on the restricted response space cross-scale characteristic data to obtain spatial response hysteresis data; Step S322: performing response progressive increment calculation on the spatial response hysteresis data to obtain response hysteresis progressive increment data; Step S323: performing spatial response interpolation processing according to the response lag progressive increment data to obtain limited response spatial interpolation data.

8. The crop disease risk assessment method according to claim 6, It is characterized in that Step S33 includes the following steps: Step S331: performing spatial interpolation point similarity measurement on crop physiological restricted response data according to restricted response spatial interpolation data to obtain spatial interpolation point similarity data; Step S332: performing non-uniformity analysis on the spatial interpolation point similarity data to obtain similarity non-uniformity data; Step S333: performing anisotropic analysis on the spatial interpolation point similarity data according to the similarity non-uniformity data to obtain similarity anisotropic data; Step S334: performing trend cluster analysis based on the similarity anisotropic data to obtain restricted response trend cluster data.

9. The crop disease risk assessment method according to claim 1, It is characterized in that Step S4 includes the following steps: Step S41: performing clustering trend learning on the restricted response trend clustering data based on a multi-layer perceptron to obtain clustering trend learning data; Step S42: normalizing the clustering trend learning data to obtain clustering trend learning normalized data; Step S43: construct a crop disease assessment model based on cluster trend learning and normalized data to obtain a crop disease assessment model, and send the crop disease assessment model to a data processing cloud platform to perform crop disease risk assessment.

10. A device for assessing the risk of crop diseases. It is characterized in that A data processing cloud platform is included for executing the crop disease risk assessment method according to claim 1, wherein the crop disease risk assessment device includes: The abnormal characteristic state recognition module is used to collect crop growth morphology image data from the crop planting area through electronic monitoring equipment to obtain crop growth morphology images; perform abnormal state feature recognition on the crop growth morphology images to obtain crop morphology abnormal state data; The crop physiological restriction response module is used to identify crop deformity morphology based on crop morphological abnormality data to obtain crop deformity morphology data; perform survival sensitivity analysis based on crop deformity morphology data to obtain nutrient deficiency-survival sensitivity data; perform crop physiological restriction response simulation based on nutrient deficiency-survival sensitivity data to obtain crop physiological restriction response data; A trend cluster analysis module is used to perform trend cluster analysis on crop physiological restricted response data to obtain restricted response trend cluster data; The assessment model building module is used to build a crop disease assessment model based on the restricted response trend clustering data using a multi-layer perceptron to obtain a crop disease assessment model; and send the crop disease assessment model to the data processing cloud platform to perform crop disease risk assessment.

Images