Disease and pest identification, prevention and early warning system and method

Abstract

The invention relates to the technical field of disease and pest identification and control, in particular to a disease and pest identification and control early warning system and method, and the system comprises a data collection module, a polarization data analysis module, an image super-resolution processing module, a feature extraction module, a disease prediction module and a decision support module. According to the method, early recognition and accurate prediction of plant diseases and insect pests are achieved by integrating capture and analysis of spectrum polarization data, image super-resolution processing and time sequence analysis, fine structural changes of plant leaves are accurately captured through the polarized light technology, fine lesion features can be revealed by deeply analyzing the polarization data, and the accuracy of the plant diseases and insect pests is improved. The local spectrum enhancement technology further refines the image quality, it is ensured that the change of the cellular level can be accurately recognized, deep learning analysis is carried out in combination with environmental data, the disease development trend can be effectively predicted, and the processing speed and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pest identification and prevention, and in particular to a pest identification and prevention early warning system and method. Background Art

[0002] The field of pest and disease identification and control technology involves the monitoring, identification, early warning and control strategies of agricultural pests and plant diseases, mainly including the use of chemical, biological or physical methods to prevent and control pests and diseases, as well as the use of modern information technology and data analysis to improve the efficiency and accuracy of pest and disease management. With the development of artificial intelligence, machine learning and image recognition technology, pest and disease identification and control technology has been able to realize automated and intelligent monitoring and processing systems, capture images of crops through cameras or sensors, analyze image data to identify specific types of pests and diseases, and automatically trigger corresponding early warning and control measures based on the identification results. In addition, it is widely used in precision agriculture to improve the sustainability and environmental friendliness of agricultural production by optimizing resource allocation and reducing the use of pesticides.

[0003] Among them, automatic identification and management of agricultural pests and diseases through technical means is mainly used for pest and disease monitoring during the growth process of crops. By real-time analysis of farmland images or data, potential pests and diseases can be quickly and accurately identified. Once pests and diseases are identified, the system will immediately notify farmers and make prevention and control suggestions or directly initiate prevention and control measures, such as automatic spraying of pesticides. This not only greatly reduces labor costs, but also improves the timeliness and accuracy of prevention and control, helps protect crops from losses, and maintains the balance of the agricultural ecology.

[0004] Current pest and disease identification and control technologies mainly rely on traditional monitoring methods, such as direct visual observation and standardized image analysis, which have limitations in early disease identification and detailed analysis. Especially in the early stages of the disease, traditional methods have difficulty capturing subtle changes in plants, resulting in delayed early warning and control measures. In terms of data processing, traditional technologies often lack real-time and automation and rely on periodic manual inspections, which not only increases labor intensity but may also miss the best time for prevention and control due to human negligence. In addition, the lack of effective data integration and analysis methods makes predictions inaccurate and unable to provide customized prevention and control recommendations, which is particularly evident when faced with diverse and complex pest and disease situations. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a pest identification and control early warning system and method.

[0006] In order to achieve the above-mentioned object, the present invention adopts the following technical scheme: a pest identification and control early warning system, including a data acquisition module, a polarization data analysis module, an image super-resolution processing module, a feature extraction module, a disease prediction module and a decision support module; The data acquisition module is used to receive the spectral polarization data of crop leaves through the polarization spectrum sensor, record the scattering and polarization state of light, and generate original polarization data results; The polarization data analysis module is used to extract spectral and polarization features from the original polarization data results, compare the feature differences between normal and diseased leaves, calculate statistical indicators of polarization data, and perform difference mapping on the statistical indicators to obtain polarization difference mapping results; The image super-resolution processing module uses the polarization difference mapping result to reconstruct the image through local spectral enhancement, analyzes the cell structure deformation and identifies pigmentation in the image, and obtains an enhanced disease image; The feature extraction module is used to apply edge detection and texture analysis to the enhanced disease image, extract key morphological and spectral features, including edge clarity, color distribution and texture uniformity, and establish a feature description set; The disease prediction module is used to input the feature description set into the deep learning network, combine the temperature and humidity data to perform time series analysis, predict the disease development trend, evaluate the disease possibility, and simulate the future state to obtain the offset dynamic prediction analysis results; The decision support module is used to analyze the deviation dynamic prediction analysis results, compare the disease data with the current model prediction, determine the probability of disease development, formulate corresponding prevention and control measures and suggestions, and output the disease and pest control plan.

[0007] As a further scheme of the present invention, the original polarization data results include spectral range, polarization mode, and light intensity; the polarization difference mapping results include polarization angle difference diagram, spectral difference diagram, and data comparison index; the enhanced disease image includes detail enhancement diagram, color contrast diagram, and structure clarity diagram; the feature description set includes morphological feature indicators, spectral feature indicators, and texture analysis results; the offset dynamic prediction analysis results include trend prediction diagrams, probability evaluation values, and simulation output data results; the pest and disease control plan includes a list of preventive measures, potential risk assessment results, and emergency response strategies.

[0008] As a further solution of the present invention, the data acquisition module includes a sensor configuration submodule, a data receiving submodule, and a data recording submodule; The sensor configuration submodule installs polarization spectral sensors in the field environment, adjusts the angle and sensitivity to match the target crop type and growth stage, captures the spectral polarization data of crop leaves, and generates sensor configuration records; The data receiving submodule adopts the sensor configuration record to receive the spectral polarization data of the polarization spectrum sensor in real time, monitor the stability and integrity of data transmission, and obtain the real-time spectral data set; The data recording submodule captures and records the scattering characteristics of light and its polarization state according to the real-time spectral data set, performs data structured processing, and generates original polarization data results.

[0009] As a further solution of the present invention, the polarization data analysis module includes a feature extraction submodule, a feature comparison submodule, and a difference mapping submodule; The feature extraction submodule performs spectrum analysis based on the original polarization data results, separates the scattered and polarized components of light, identifies the spectral intensity of all frequency bands, and generates a feature data set; The feature comparison submodule extracts feature data of normal and diseased leaves from the feature data set, calculates feature difference values, and obtains feature difference data; The difference mapping submodule uses the characteristic difference data to perform difference mapping on the difference data, highlighting the healthy and diseased areas of the leaf, and obtaining a polarization difference mapping result.

[0010] As a further solution of the present invention, the image super-resolution processing module includes a local enhancement submodule, a structure analysis submodule, and an image output submodule; The local enhancer module analyzes and adjusts the frequency band of the spectral data based on the polarization difference mapping results, dynamically adjusts the spectral intensity of the target area, and generates a local enhanced image; The structural analysis submodule uses the local enhanced image to perform image segmentation to distinguish healthy and diseased areas in the image, and uses texture analysis to evaluate and compare the color and texture differences between the healthy and diseased areas to obtain morphological analysis results; The image output submodule integrates and processes the image data based on the morphological analysis results, analyzes the image for rendering, highlights the diseased area, and generates an enhanced diseased image.

[0011] As a further solution of the present invention, the feature extraction module includes an edge detection submodule, a texture analysis submodule, and a feature integration submodule; The edge detection submodule uses edge detection to analyze the enhanced disease image based on the enhanced disease image, and performs grayscale conversion to identify edge clarity, highlight the contour of the diseased area, and generate an edge feature map; The texture analysis submodule uses the edge feature map to apply a texture filter and a gray level co-occurrence matrix to analyze the image texture of the edge feature map, calculates the statistical parameters of the texture, and obtains the texture analysis result; The feature integration submodule performs edge detection on the image based on the texture analysis result, integrates the image edge detection data and texture data, and generates a feature description set.

[0012] As a further solution of the present invention, the disease prediction module includes a data integration submodule, a model training submodule, and a simulation output submodule; The data integration submodule integrates external environmental data, including temperature and humidity, based on the feature description set, standardizes the data, merges the data sets for time series analysis, and generates a comprehensive feature data set; The model training submodule adopts the comprehensive feature data set and uses a deep learning network to train a time series model to obtain a lesion trend prediction model; The simulation output submodule predicts the future state based on the lesion trend prediction model, evaluates the development trend of the disease, and generates the deviation dynamic prediction analysis results.

[0013] As a further solution of the present invention, the decision support module includes a dynamic analysis submodule, a risk assessment submodule, and a strategy formulation submodule; The dynamic analysis submodule compares the lesion trend prediction model with the disease data based on the deviation dynamic prediction analysis results, performs data matching and trend comparison, quantitatively analyzes the consistency and deviation of the data, and generates a trend comparison analysis report; The risk assessment submodule uses the trend comparison analysis report to assess the probability and risk of disease development, calculate the possibility of disease outbreak, analyze the impact on crop health, and obtain risk assessment results; The strategy formulation submodule formulates customized prevention and control measures based on the risk assessment results and the target disease risk factors, and outputs a pest and disease control plan in combination with the actual operating environment requirements.

[0014] As a further embodiment of the present invention, the probability of the disease outbreak is calculated according to the formula Calculate, where P represents the probability of disease outbreak, Represents the frequency of recent disease occurrence, represents the historical outbreak intensity of the disease, Represents the current health status index of the crop, represents the ambient temperature, Represents the ambient humidity, Represents the pH value of soil.

[0015] A method for early warning of disease and insect pest identification and prevention, comprising the following steps: S1: Capture the spectral polarization data of crop leaves, monitor the stability and integrity of data transmission, perform data structured processing and storage, and generate raw polarization data results; S2: Based on the original polarization data result, spectrum analysis is performed, spectrum intensity of all frequency bands is identified and recorded through spectrum analysis, difference analysis is performed to compare the difference values ​​of features, difference mapping analysis is performed, and polarization difference mapping results are generated; S3: Based on the polarization difference mapping result, analyzing and adjusting the frequency band of the spectral data, distinguishing the healthy and diseased areas in the image, analyzing and rendering the image, and generating an enhanced disease image; S4: Based on the enhanced disease image, identifying the edge clarity of the image, highlighting the outline of the diseased area, calculating the statistical parameters of the texture, integrating the edge and texture data of the image, and generating a feature description set; S5: Based on the feature description set, integrate the external environment data, merge the data sets to perform time series analysis, use the deep learning network to train the time series model, and obtain the model training results; S6: Based on the model training results, predict the future status, evaluate the development trend of the disease, quantitatively analyze the consistency and deviation of historical and predicted data, and generate a trend comparison analysis report; S7: Based on the trend comparison and analysis report, evaluate the probability and risk of disease development, analyze the potential impact on crop health, formulate targeted prevention and control measures and suggestions, formulate customized prevention and control measures for target disease risk factors, and output pest and disease control plans.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by capturing and analyzing comprehensive spectral polarization data, image super-resolution processing and time series analysis, earlier identification and accurate prediction of pests and diseases are achieved. Polarized light technology is used to accurately capture subtle structural changes in plant leaves. Through in-depth analysis of polarization data, subtle pathological characteristics can be revealed. Local spectral enhancement technology further refines image quality, ensuring that changes at the cellular level can be accurately identified. Combined with environmental data for deep learning analysis, it can effectively predict the development trend of diseases and improve processing speed and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a system flow chart of the present invention; Figure 2 It is the overall flow chart of the modules of the present invention; Figure 3 It is a flow chart of the data acquisition module of the present invention; Figure 4 is a flow chart of the polarization data analysis module of the present invention; Figure 5 This is a flow chart of the image super-resolution processing module of the present invention; Figure 6It is a flow chart of the feature extraction module of the present invention; Figure 7 It is a flow chart of the disease prediction module of the present invention; Figure 8 It is a flow chart of the decision support module of the present invention; Fig. 9 The figure is a flow chart of the steps of the method of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0020] See also Figure 1 and Figure 2 ,A pest and disease identification and prevention early warning system includes: a data acquisition module, a polarization data analysis module, an image super-resolution processing module, a feature extraction module, a disease prediction module and a decision support module; The data acquisition module is used to receive the spectral polarization data of crop leaves through the polarization spectrum sensor, record the scattering and polarization state of light, and generate original polarization data results; The polarization data analysis module is used to extract spectral and polarization features from the original polarization data results, compare the feature differences between normal and diseased leaves, calculate the statistical indicators of the polarization data, and perform difference mapping on the statistical indicators to obtain polarization difference mapping results; The image super-resolution processing module is used to use the polarization difference mapping results to reconstruct the image through local spectral enhancement, analyze the cell structure deformation and identify pigmentation in the image, and obtain enhanced disease images; The feature extraction module is used to apply edge detection and texture analysis to the enhanced disease images, extract key morphological and spectral features, including edge clarity, color distribution, and texture uniformity, and establish a feature description set; The disease prediction module is used to input the feature description set into the deep learning network, combine the temperature and humidity data to perform time series analysis, predict the disease development trend, evaluate the possibility of disease, and simulate the future state to obtain the offset dynamic prediction analysis results; The decision support module is used to analyze the results of the offset dynamic prediction analysis, compare the disease data with the current model prediction, determine the probability of disease development, formulate corresponding prevention and control measures and suggestions, and output the pest and disease control plan.

[0021] The original polarization data results include spectral range, polarization mode, and light intensity. The polarization difference mapping results include polarization angle difference map, spectral difference map, and data comparison index. The enhanced disease image includes detail enhancement map, color contrast map, and structure clarity map. The feature description set includes morphological feature indicators, spectral feature indicators, and texture analysis results. The offset dynamic prediction analysis results include trend prediction map, probability assessment value, and simulation output data results. The pest and disease control plan includes a list of preventive measures, potential risk assessment results, and emergency response strategy.

[0022] See also Figure 3 ,The data acquisition module includes a sensor configuration submodule, a data receiving submodule, and a data recording submodule; The sensor configuration submodule installs polarization spectral sensors in the field environment, adjusts the angle and sensitivity to match the target crop type and growth stage, captures the spectral polarization data of crop leaves, and generates sensor configuration records; Ensure that the data it captures can accurately reflect the actual spectral state of the target crop type and growth stage. The specific steps involved in this process include selecting a suitable sensor installation location, adjusting the sensor angle according to the growth characteristics of the target crop, and determining the sensor sensitivity settings through experimental data. These settings must take into account environmental factors such as light intensity and the reflective characteristics of crop leaves. Compare the acquired experimental data with the preset parameters to generate accurate sensor configuration records. These records provide the necessary configuration information for the subsequent data receiving submodule.

[0023] The data receiving submodule adopts the sensor configuration record to receive the spectral polarization data from the polarization spectrum sensor in real time, monitor the stability and integrity of data transmission, and obtain the real-time spectral data set; According to the formula The integrity and stability of the calculation data. Represents the spectral intensity at different angles, represents the polarization rate at the corresponding angle, Represents the measurement angle. Detailed explanation of the formula and the process of formula calculation: In order to determine the integrity and stability of the data, first obtain spectral data from a series of angles from the sensor , the data at each angle has a corresponding polarization rate , the overall data integrity measure is obtained by integral calculation When calculating specifically, we can measure the spectral intensity and polarization rate at several key angles, for example , , and the corresponding , , , the integral value is estimated using the numerical integration method. Such a calculation method can ensure the stability and integrity evaluation of data transmission under different environments and conditions.

[0024] The data recording submodule captures and records the scattering characteristics of light and its polarization state based on the real-time spectral data set, performs data structured processing, and generates raw polarization data results; Data is captured and recorded through a refined data structuring process. The steps involved include data preprocessing, such as noise removal and outlier processing, data classification and storage, and specific analysis of the scattering characteristics of light and its polarization state. These processes ensure the quality and availability of the data. Through these structured data processing, the original polarization data results with detailed timestamps and environmental parameters are finally generated, providing a basis for further data analysis and application.

[0025] See also Figure 4 ,The polarization data analysis module includes a feature extraction submodule, a feature comparison submodule, and a difference mapping submodule; The feature extraction submodule performs spectrum analysis based on the original polarization data results, separates the scattered and polarized components of light, identifies the spectral intensity of all frequency bands, and generates a feature data set; According to the formula Calculate the spectral intensity of each frequency band. The representative wavelength is The spectral intensity, Representative wavelength and scattering angle The light intensity distribution function, is a small angle increment. Assuming that under certain experimental conditions, The measurement is (Unit: mW / cm 2 / degree), for the integral of the scattering angle range from 0° to 180°, calculate mW / cm 2 , which shows that the spectral intensity at a specific wavelength characterizes the optical properties of the material.

[0026] The feature comparison submodule extracts feature data of normal and diseased leaves from the feature data set, calculates feature difference values, and obtains feature difference data; By setting thresholds and difference measurement methods, the spectral intensity differences between normal and diseased leaves in the same frequency band are compared. The difference measurement method can be absolute difference or ratio, etc. At the same time, statistical methods are introduced to evaluate the significance of the differences to determine which frequency bands' changes can best reflect the pathological status. Based on the results of these difference data analysis, the researchers can more accurately identify and classify the health status of leaves. The characteristic difference data finally obtained provides a key basis for further analysis.

[0027] The difference mapping submodule uses the feature difference data to perform difference mapping on the difference data, highlighting the healthy and diseased areas of the leaf and obtaining the polarization difference mapping result; The data is converted into visual images through algorithms. The depth and range of colors indicate the degree and area of ​​lesions. Each color corresponds to a certain range of polarization data differences. Image segmentation and edge detection technology are implemented to enhance the accuracy and clarity of the mapping, so that the difference mapping can intuitively show the health status of plant leaves, providing a practical tool for agricultural experts and plant pathologists.

[0028] See also Figure 5 ,The image super-resolution processing module includes a local enhancement submodule, a structural analysis submodule, and an image output submodule; The local enhancer module analyzes and adjusts the frequency band of the spectral data based on the polarization difference mapping results, dynamically adjusts the spectral intensity of the target area, and generates a local enhanced image; To ensure that the details of the local area are better displayed, the process involves dynamic adjustment of the spectral intensity, using high-pass filters and local contrast enhancement technology to focus on the significant areas of healthy and diseased leaves, enhance the local contrast and color saturation of the image, make the details clearer, and provide high-quality image data for subsequent structural analysis. The generated local enhanced image can more effectively assist in the identification and analysis of the specific pathological conditions of the leaves.

[0029] The structural analysis submodule uses the local enhanced image to perform image segmentation to distinguish the healthy and diseased areas in the image, and uses texture analysis to evaluate and compare the color and texture differences between the healthy and diseased areas to obtain the morphological analysis results; According to the formula Calculate the image texture difference. In the formula, A measure of texture difference, Represents the intensity of the i-th pixel in the image, Represents the average value of all pixel intensities. is the total number of pixels. Assume that a set of pixel intensities is [100,102,98,101,103] ,average value ,but ; This calculation reveals the degree of difference in pixel intensity in the image, assisting in texture analysis and more accurate identification of lesion areas.

[0030] The image output submodule integrates and processes the image data based on the morphological analysis results, analyzes the image for rendering, highlights the diseased area, and generates an enhanced diseased image; Image data is integrated through a series of image processing techniques, including image rendering and color adjustment, to highlight the diseased area. This process involves analyzing the various components of the image and enhancing or suppressing these components, especially visually highlighting the diseased area to make it more conspicuous in the final output image. In this way, the image output module is not just a simple image display, but through advanced image processing technology, it generates enhanced disease images with analytical value. These images provide important visual information for subsequent medical diagnosis or agricultural monitoring.

[0031] See also Figure 6 ,The feature extraction module includes an edge detection submodule, a texture analysis submodule, and a feature integration submodule; The edge detection submodule uses edge detection to analyze the enhanced disease image based on the enhanced disease image, and performs grayscale conversion to identify edge clarity, highlight the contour of the diseased area, and generate an edge feature map; By applying operators such as Canny or Sobel, which determine the location of the boundary by calculating the gradient change of image intensity, the edge detection process not only improves the contrast of the image, but also highlights the details of the diseased area, making these areas easier to identify in subsequent analysis. In addition, after edge detection, the enhanced diseased image provides the necessary pre-processing steps for the next step of texture analysis and feature integration.

[0032] The texture analysis submodule uses the edge feature map, applies texture filters and gray-level co-occurrence matrix to analyze the image texture of the edge feature map, calculates the statistical parameters of the texture, and obtains the texture analysis results; According to the formula Calculate the texture contrast. Where, represents texture contrast, Represents pixel The gray value of Represents the average grayscale value of all pixels. is the total number of pixels. Assume that the pixel grayscale value of a small area is [100,105,100,95,90] ,average value ,but ; This calculation result reflects the light and dark contrast of the image texture, which is crucial for texture analysis. It helps to reveal subtle differences between different textures and provides a means to accurately evaluate the quality of image texture.

[0033] The feature integration submodule performs edge detection on the image based on the texture analysis results, integrates the image edge detection data and texture data, and generates a feature description set; It involves data fusion technology, including multi-scale analysis and feature selection methods, to ensure that the most useful information is extracted from various data sources. Edge features and texture features are combined through a unified framework to improve the accuracy and efficiency of disease detection. The feature description set contains detailed edge and texture information, which provides rich input data for subsequent machine learning models and promotes the implementation of advanced pattern recognition tasks such as classification and diagnosis, thereby achieving fast and accurate lesion identification in practical applications.

[0034] See also Figure 7 ,The disease prediction module includes a data integration submodule, a model training submodule, and a simulation output submodule; The data integration submodule integrates external environmental data, including temperature and humidity, based on the feature description set, standardizes the data, merges the data sets for time series analysis, and generates a comprehensive feature data set; The process involves a variety of data preprocessing techniques, including data normalization and time window segmentation to meet the needs of time series analysis. In this way, the module not only integrates data from different sources, but also performs necessary transformations on them to ensure the consistency and comparability of data input. The final comprehensive feature dataset provides rich environmental and biological features, providing comprehensive input for model training.

[0035] The model training submodule uses a comprehensive feature data set and a deep learning network to train a time series model to obtain a lesion trend prediction model; According to the formula Calculate the model loss. In the formula, represents the model loss, Representative The actual observed values, represents the model prediction value, is the total number of samples. Assume there are 5 observations [3,5,2,6,4] and the corresponding model prediction is [2,5,3,5,4] ,but ; This calculation result evaluates the model's fitting effect on the time series, guides subsequent model optimization and parameter adjustment, and improves the accuracy of the prediction.

[0036] The simulation output submodule predicts the future state based on the lesion trend prediction model, evaluates the development trend of the disease, and generates the deviation dynamic prediction analysis results; The process uses a trained deep learning model to predict future data and uses the model output to evaluate the possible development path of plant diseases. In addition, the prediction results use dynamic simulation to show the possible changes of diseases in the future, providing decision support for agricultural management. These dynamic prediction results help to adjust crop management strategies in a timely manner, prevent or reduce the impact of diseases, and thus improve the overall health and yield of crops.

[0037] See also Figure 8 ,The decision support module includes a dynamic analysis submodule, a risk assessment submodule, and a strategy formulation submodule; The dynamic analysis submodule compares the lesion trend prediction model with the disease data based on the deviation dynamic prediction analysis results, performs data matching and trend comparison, quantitatively analyzes the consistency and deviation of the data, and generates a trend comparison analysis report; This process evaluates the accuracy and reliability of forecast data through statistical methods such as correlation coefficient and deviation analysis. The key execution steps include data alignment, time window selection and difference calculation, which ensure the accuracy of analysis and the systematic nature of operation. The results of trend comparison help identify the performance of the model under specific conditions and provide a basis for further model adjustment and optimization.

[0038] The risk assessment submodule uses trend comparison analysis reports to assess the probability and risk of disease development, calculate the possibility of disease outbreaks, analyze the impact on crop health, and obtain risk assessment results; The probability of a disease outbreak is calculated according to the formula: Calculate, where P represents the probability of disease outbreak, Represents the recent frequency of disease occurrence, represents the number of times the disease occurs within a week, Represents the historical outbreak intensity of the disease, indicating the percentage of crops affected in the most severe outbreak in the past year. Represents the current health status index of the crop, obtained from the crop health monitoring system, Represents the ambient temperature, the average temperature of the most recent day provided by the weather station. Represents the ambient humidity, the average humidity of the most recent day provided by the weather station. Represents soil pH value, obtained through soil analysis instrument.

[0039] Represents the recent frequency of disease occurrence, which means the number of times the disease occurs within a week, for example, 7 times.

[0040] Represents the historical outbreak intensity of the disease, indicating the percentage of crops affected by the most severe outbreak in the past year, for example 30%.

[0041] Represents the current health status of the crop, obtained from the crop health monitoring system, assumed to be 85 (out of 100).

[0042] Represents the ambient temperature, which is the average temperature of the most recent day provided by the weather station, assumed to be 23°C.

[0043] Represents the ambient humidity, which is the average humidity of the most recent day provided by the weather station, assumed to be 75%.

[0044] Represents soil pH, obtained through a soil analysis instrument, and the assumed value is 7.0.

[0045] Specific calculation process: calculate ; calculate ; calculate ; calculate ; calculate ; calculate ; calculate ; calculate ; calculate ; calculate ; The results show that based on the current environmental conditions and crop status, the calculated disease risk value is 1.726, which expresses the probability that crops will be affected by diseases under current conditions. A value exceeding 1 indicates a high disease risk and that appropriate prevention and control measures need to be taken.

[0046] The strategy formulation submodule formulates customized prevention and control measures based on the risk assessment results and the target disease risk factors, and outputs the pest control plan in combination with the actual operating environment requirements; It involves analyzing the specific disease types of crops, combining them with the actual operating environment needs, and designing prevention and control plans suitable for different crops and regions. This includes selecting appropriate chemical and biological control products, setting the time and method of application, considering environmental protection and cost-effectiveness, and the output of pest and disease control plans aims to minimize the impact of diseases and ensure crop health and yield. Through the implementation of this strategy, farm managers can effectively deal with possible disease risks and optimize crop management.

[0047] See also Fig. 9 , a pest identification and prevention early warning method, comprising the following steps: S1: Capture the spectral polarization data of crop leaves, monitor the stability and integrity of data transmission, perform data structured processing and storage, and generate raw polarization data results; S2: Based on the original polarization data results, perform spectrum analysis, identify and record the spectral intensity of all frequency bands through spectrum analysis, perform difference analysis to compare the difference values ​​of features, perform difference mapping analysis, and generate polarization difference mapping results; S3: Based on the polarization difference mapping results, analyze and adjust the frequency band of the spectral data, distinguish the healthy and diseased areas in the image, analyze and render the image, and generate an enhanced disease image; S4: Based on the enhanced disease image, identify the edge clarity of the image, highlight the outline of the diseased area, calculate the statistical parameters of the texture, integrate the edge and texture data of the image, and generate a feature description set; S5: Based on the feature description set, integrate the external environment data, merge the data set for time series analysis, use the deep learning network to train the time series model, and obtain the model training results; S6: Based on the model training results, predict the future status, evaluate the development trend of the disease, quantitatively analyze the consistency and deviation of historical and predicted data, and generate a trend comparison analysis report; S7: Based on the trend comparison analysis report, evaluate the probability and risk of disease development, analyze the potential impact on crop health, formulate targeted prevention and control measures and suggestions, formulate customized prevention and control measures for target disease risk factors, and output pest and disease control plans.

[0048] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A pest identification and prevention early warning system, characterized by: The system includes a data acquisition module, a polarization data analysis module, an image super-resolution processing module, a feature extraction module, a disease prediction module and a decision support module; The data acquisition module is used to receive the spectral polarization data of crop leaves through the polarization spectrum sensor, record the scattering and polarization state of light, and generate original polarization data results; The polarization data analysis module is used to extract spectrum and polarization features from the original polarization data results, compare the feature differences between normal and diseased leaves, calculate statistical indicators of polarization data, and perform difference mapping on the statistical indicators to obtain polarization difference mapping results; The image super-resolution processing module is used to use the polarization difference mapping result to reconstruct the image through local spectral enhancement, analyze the cell structure deformation and identify pigmentation in the image, and obtain an enhanced disease image; The feature extraction module is used to apply edge detection and texture analysis to the enhanced disease image, extract key morphological and spectral features, including edge clarity, color distribution and texture uniformity, and establish a feature description set; The disease prediction module is used to input the feature description set into the deep learning network, perform time series analysis in combination with temperature and humidity data, predict disease development trends, assess disease likelihood, and simulate future states to obtain offset dynamic prediction analysis results; The decision support module is used to analyze the offset dynamic prediction analysis results, compare disease data with current model predictions, determine the probability of disease development, formulate corresponding prevention and control measures and suggestions, and output a pest control plan.

2. The pest identification and prevention early warning system according to claim 1, characterized in that: The original polarization data results include spectral range, polarization mode, and light intensity; the polarization difference mapping results include polarization angle difference map, spectral difference map, and data comparison index; the enhanced disease image includes detail enhancement map, color contrast map, and structure clarity map; the feature description set includes morphological feature indicators, spectral feature indicators, and texture analysis results; the offset dynamic prediction analysis results include trend prediction maps, probability evaluation values, and simulation output data results; the pest and disease control plan includes a list of preventive measures, potential risk assessment results, and emergency response strategies.

3. The pest identification and prevention early warning system according to claim 1, characterized in that: The data acquisition module includes a sensor configuration submodule, a data receiving submodule, and a data recording submodule; The sensor configuration submodule installs polarization spectral sensors in the field environment, adjusts the angle and sensitivity to match the target crop type and growth stage, captures the spectral polarization data of crop leaves, and generates sensor configuration records; The data receiving submodule adopts the sensor configuration record to receive the spectral polarization data of the polarization spectrum sensor in real time, monitor the stability and integrity of data transmission, and obtain the real-time spectral data set; The data recording submodule captures and records the scattering characteristics of light and its polarization state according to the real-time spectral data set, performs data structured processing, and generates original polarization data results.

4. The pest identification and prevention early warning system according to claim 1, characterized in that: The polarization data analysis module includes a feature extraction submodule, a feature comparison submodule, and a difference mapping submodule; The feature extraction submodule performs spectrum analysis based on the original polarization data results, separates the scattered and polarized components of light, identifies the spectral intensity of all frequency bands, and generates a feature data set; The feature comparison submodule extracts feature data of normal and diseased leaves from the feature data set, calculates feature difference values, and obtains feature difference data; The difference mapping submodule uses the characteristic difference data to perform difference mapping on the difference data, highlighting the healthy and diseased areas of the leaf, and obtaining a polarization difference mapping result.

5. The pest identification and prevention early warning system according to claim 1, characterized in that: The image super-resolution processing module includes a local enhancement submodule, a structure analysis submodule, and an image output submodule; The local enhancer module analyzes and adjusts the frequency band of the spectral data based on the polarization difference mapping results, dynamically adjusts the spectral intensity of the target area, and generates a local enhanced image; The structural analysis submodule uses the local enhanced image to perform image segmentation to distinguish healthy and diseased areas in the image, and uses texture analysis to evaluate and compare the color and texture differences between the healthy and diseased areas to obtain morphological analysis results; The image output submodule integrates and processes the image data based on the morphological analysis results, analyzes the image for rendering, highlights the diseased area, and generates an enhanced diseased image.

6. The pest identification and prevention early warning system according to claim 1, characterized in that: The feature extraction module includes an edge detection submodule, a texture analysis submodule, and a feature integration submodule; The edge detection submodule uses edge detection to analyze the enhanced disease image based on the enhanced disease image, and performs grayscale conversion to identify edge clarity, highlight the contour of the diseased area, and generate an edge feature map; The texture analysis submodule uses the edge feature map, applies a texture filter and a gray level co-occurrence matrix to analyze the image texture of the edge feature map, calculates the statistical parameters of the texture, and obtains the texture analysis result; The feature integration submodule performs edge detection on the image based on the texture analysis result, integrates the image edge detection data and texture data, and generates a feature description set.

7. The pest identification and prevention early warning system according to claim 1, characterized in that: The disease prediction module includes a data integration submodule, a model training submodule, and a simulation output submodule; The data integration submodule integrates external environmental data, including temperature and humidity, based on the feature description set, standardizes the data, merges the data sets for time series analysis, and generates a comprehensive feature data set; The model training submodule adopts the comprehensive feature data set and uses a deep learning network to train a time series model to obtain a lesion trend prediction model; The simulation output submodule predicts the future state based on the lesion trend prediction model, evaluates the development trend of the disease, and generates the deviation dynamic prediction analysis results.

8. The pest identification and prevention early warning system according to claim 1, characterized in that: The decision support module includes a dynamic analysis submodule, a risk assessment submodule, and a strategy formulation submodule; The dynamic analysis submodule compares the lesion trend prediction model with the disease data based on the deviation dynamic prediction analysis results, performs data matching and trend comparison, quantitatively analyzes the consistency and deviation of the data, and generates a trend comparison analysis report; The risk assessment submodule uses the trend comparison analysis report to assess the probability and risk of disease development, calculate the possibility of disease outbreak, analyze the impact on crop health, and obtain risk assessment results; The strategy formulation submodule formulates customized prevention and control measures based on the risk assessment results and the target disease risk factors, and outputs a pest and disease control plan in combination with the actual operating environment requirements.

9. The pest identification and prevention early warning system according to claim 8, characterized in that: The probability of the disease outbreak is calculated according to the formula: Calculate, where P represents the probability of disease outbreak, Represents the frequency of recent disease occurrence, Represents the historical outbreak intensity of the disease, Represents the current health status index of the crop, represents the ambient temperature, Represents the ambient humidity, Represents the pH value of soil.

10. A method for early warning of disease and insect pest identification and prevention, characterized in that: According to any one of claims 1 to 9, the early warning system for identifying and controlling pests and diseases comprises the following steps: Capture the spectral polarization data of crop leaves, monitor the stability and integrity of data transmission, perform structured data processing and storage, and generate raw polarization data results; Based on the original polarization data results, spectrum analysis is performed, the spectrum intensity of all frequency bands is identified and recorded through spectrum analysis, difference analysis is performed to compare the difference values ​​of features, difference mapping analysis is performed, and polarization difference mapping results are generated; Based on the polarization difference mapping result, analyzing and adjusting the frequency band of the spectral data, distinguishing the healthy and diseased areas in the image, analyzing and rendering the image, and generating an enhanced disease image; Based on the enhanced disease image, identifying the edge clarity of the image, highlighting the outline of the diseased area, calculating the statistical parameters of the texture, integrating the edge and texture data of the image, and generating a feature description set; Based on the feature description set, external environment data is integrated, data sets are merged for time series analysis, a time series model is trained using a deep learning network, and model training results are obtained; Based on the model training results, predict future status, evaluate the development trend of the disease, quantitatively analyze the consistency and deviation of historical and predicted data, and generate a trend comparison analysis report; Based on the trend comparison and analysis report, the probability and risk of disease development are evaluated, the potential impact on crop health is analyzed, targeted prevention and control measures and suggestions are formulated, customized prevention and control measures are formulated for target disease risk factors, and pest and disease control plans are output.

Images