Crop growth situation analysis and yield evaluation method

Abstract

The invention belongs to the technical field of agricultural informatization, and discloses a crop growth situation analysis and yield evaluation method, which comprises the steps of capturing crop growth monitoring data in real time through an Internet of Things edge node cluster; the crop growth monitoring data comprises crop growth data, growth environment data and crop growth image data; an environment weight mechanism is introduced, the environment weight of each time point is calculated according to the growth environment data of the time sequence, and the adjustment coefficient of the weight influence degree is dynamically adjusted through a deviation feedback mechanism; smoothing the crop growth data by using an improved moving average algorithm in combination with the growth environment data, and automatically adjusting the size of a window according to environment fluctuation to obtain smoothed crop growth data; performing noise filtering on the smoothed crop growth data by using a Kalman filtering algorithm, and refining to generate crop growth analysis data; healthy growth of crops is ensured, and yield loss is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural informatization. More specifically, the present invention relates to a method for analyzing the growth trend and evaluating the yield of crops. Background Art

[0002] A patent with the publication number CN114997535A discloses a method and system platform for intelligent analysis of big data in the whole process of smart agriculture. The method includes: collecting in real time the image data of the growth trend of crops and the whole production process; preprocessing the image data collected in real time; constructing and training a prediction model for the growth trend of crops and the whole production process, including a first prediction sub-model and a second prediction sub-model; inputting the preprocessed image data into the first prediction sub-model and the second prediction sub-model respectively to obtain a first prediction result for the growth trend of crops and the whole production process and a second prediction result for the growth trend of crops and the whole production process; and obtaining the final prediction result for the growth trend of crops and the whole production process by weighted summation. The present invention constructs an agricultural big data platform based on big data and neural network technologies, and can fully adapt to the development direction of the fields of agricultural big data and information intelligent analysis technologies.

[0003] The existing methods for analyzing the growth trend and evaluating the yield of crops have the following defects:

[0004] The crop growth data will have drastic fluctuations under the influence of environmental changes and observation errors, which will cause the model to be unable to accurately reflect the long-term growth trend and lead to inaccurate prediction results; the growth state of crops is usually affected by environmental conditions (such as climate, soil, pests and diseases, etc.), and the model may not be able to adapt to the rapid changes in the environment, resulting in too large prediction errors, ignoring the influence of environmental fluctuations on growth data, and the model cannot effectively adapt to the changes in environmental conditions, thus limiting the adaptability of the system when facing different types of environmental fluctuations and affecting its stability and accuracy; the crop growth data may contain a large amount of noise caused by random fluctuations such as short-term climate changes and measurement errors, resulting in the lack of continuity of the data and affecting the accuracy of subsequent analysis tasks such as pest early warning; the size of the sliding window is fixed, and the environmental conditions change frequently, and the system lacks an adaptive mechanism; the environmental weight and difference adjustment coefficient in the growth environment data are not combined, which will underestimate or over-average the importance of environmental conditions, thus affecting the response ability of the model to environmental changes;

[0005] Considering only a single or a few environmental factors may overlook other important factors such as soil quality, air quality, pests and diseases. The growth of crops is affected by the interaction of multiple environmental factors. Ignoring these factors may lead to an incomplete assessment of crop growth and affect the accuracy of growth prediction. Different types of environmental fluctuations have different impacts on crop growth. Static smoothing strategies may not be able to adapt to these changes. Without adjusting the sensitivity to environmental changes, it may lead to insufficient sensitivity or overreaction to certain environmental factors. Without an effective dynamic smoothing mechanism, it may lead to the failure to eliminate the short-term fluctuations in data caused by environmental fluctuations in a timely manner, resulting in unstable input data for the model and affecting the accuracy of subsequent analysis.

[0006] When the measurement error is large or the environmental fluctuations are severe, the model may overly rely on predicted values or observed values, resulting in inaccurate prediction of the growth state of crops. Without flexible adjustment, the model may not be able to timely adjust its response to sudden environmental changes, thus missing the key changes in crop growth and affecting the reliability and accuracy of long-term trends.

[0007] In view of this, the present invention proposes a method for analyzing the growth trend and evaluating the yield of crops to solve the above problems. Summary of the Invention

[0008] In order to overcome the above defects of the prior art and to achieve the above object, the present invention provides the following technical solution: A method for analyzing the growth trend and evaluating the yield of crops, comprising:

[0009] S1. Real-time capture crop growth monitoring data through an IoT edge node cluster; the crop growth monitoring data includes crop growth data, growth environment data, and crop growth image data;

[0010] S2. Introduce an environmental weight mechanism, calculate the environmental weight at each time point according to the growth environment data in the time series, and dynamically adjust the adjustment coefficient of the weight influence degree through a deviation feedback mechanism; use an improved moving average algorithm to smooth the crop growth data in combination with the growth environment data, and automatically adjust the window size according to environmental fluctuations;

[0011] Use the Kalman filter algorithm to filter out the noise of the smoothed crop growth data and refine the crop growth analysis data; process the growth environment data and the crop growth image data to obtain growth environment variable data and growth image analysis data;

[0012] S3. Integrate the crop growth analysis data, the growth environment variable data, and the growth image analysis data to obtain a crop growth state aggregate;

[0013] S4. Input the crop growth state aggregate into the pre-trained growth trend evaluation model to predict the growth trend of the crops in the next n periods of time;

[0014] S5. Obtain historical yield data, train and obtain a crop yield prediction model based on the historical yield data and the growth trend of the crops in the next n periods of time, and predict the yield of the crops based on the crop yield prediction model;

[0015] S6. Evaluate whether the crops are in the best growth trend according to the predicted yield of the crops;

[0016] S7. If the crops are not in the best growth trend, feedback to the staff through the crop growth intelligent monitoring terminal for timely intervention.

[0017] Preferably, the crop growth data includes growth form data and growth function data; the growth environment data includes meteorological data, soil data, air quality data and pest and disease data; the crop growth image data includes crop vegetation index images, chlorophyll content images, plant growth height images, crop canopy coverage images and leaf area index images.

[0018] Preferably, the method for obtaining the smoothed crop growth data includes:

[0019] S31. Preset the growth form data as: Wherein, is the growth form data at time point t; x q is the q-th form index in the growth form data at time point t; q is the index of the form index, q = 1,..., N; N is the total number of form indices;

[0020] Preset the growth function data as: Wherein, is the growth function data at time point t; y j is the j-th function index in the growth function data at time point t; j is the index of the function index, j = 1,..., M; M is the total number of function indices;

[0021] S32. Concatenate the growth form data at time point t with the growth function data at time point t to form crop growth data, and the crop growth data is expressed as: x t = {x1, x2,..., x q ,..., x N , y1, y2,..., y j ,..., y M}; wherein, x tThe crop growth data at time point t;

[0022] The preset growth environment data at time point t is: Among them, is the meteorological data at time point t; is the soil data at time point t; is the air quality data at time point t; is the pest and disease data at time point t;

[0023] S33. According to the growth environment data E at time point t t , calculate the environmental weight of each time point t through the environmental weight calculation formula; the environmental weight calculation formula is: Among them, ω i (E t ) is the weight corresponding to the growth environment data E at time point t t corresponding to the i-th time step; E max is the ideal growth environment data; ||E t -E max || is the difference between the growth environment data corresponding to time point t and the ideal growth environment data E max ; ΔE t represents the change rate of the growth environment data from time point t-1 to time point t; α is the adjustment coefficient that controls the influence degree of the difference between the growth environment data and the ideal growth environment data E max on the weight; β is the adjustment coefficient that controls the influence degree of the change rate of the growth environment data on the weight; i is the index of the time step;

[0024] S34. Dynamically adjust the adjustment coefficient α that controls the influence degree of the difference between the growth environment data and the ideal growth environment data E max on the weight through the difference adjustment coefficient self-adaptive formula. The difference adjustment coefficient self-adaptive formula is: Among them, m is the number of types of growth environment data; is the change rate of the c-th type of growth environment data at time point t; b is the constant that adjusts the influence strength of the change rate on the adjustment coefficient; p is the exponential factor that controls the influence degree of the change rate accumulation on the adjustment coefficient; c is the index of the growth environment data type;

[0025] S35. Smooth the crop growth data using the weighted moving average formula; the weighted moving average formula is: Among them, WA t is the smoothed crop growth data at time point t; x t-a is the crop growth data at time point t-a; γ is the size of the moving window; a is the index of the moving window;

[0026] S36. Dynamically adjust the size γ of the sliding window through the sliding window adjustment formula, and the sliding window adjustment formula is: where γ max is the maximum value of the sliding window; ω avg is the average weight of the growth environment data; d is a constant that controls the influence of environmental fluctuations on the size of the sliding window; is the floor function.

[0027] Preferably, use the Kalman filter algorithm to filter out the noise of the smoothed crop growth data. The method for refining and generating crop growth analysis data includes:

[0028] Define the crop growth state transition model, reflect the change process of the crop growth state over time through the crop growth state transition model, and predict the crop growth state at the current time point based on the crop growth state and control input at the previous time point;

[0029] By updating the predicted value of the crop growth state at the current time point t, calculate the Kalman gain, and balance the weights between the predicted value of the crop growth state and the actual observed value of the crop growth state through the Kalman gain; and update the error covariance matrix in the Kalman gain to reflect the uncertainty of the updated crop growth state; output the crop growth data after denoising by the Kalman filter as the final result to obtain the crop growth analysis data.

[0030] Preferably, the method for processing the growth environment data and the crop growth image data to obtain the growth environment variable data and the growth image analysis data includes:

[0031] Identify and remove the outliers in the growth environment data and the crop growth image data included in the crop growth monitoring data through the density clustering algorithm to obtain the growth environment variable data and the growth image analysis data.

[0032] Preferably, the method for obtaining the crop growth state aggregate includes:

[0033] Perform standard deviation normalization on the crop growth analysis data, the growth environment variable data, and the growth image analysis data, and convert them into a standard normal distribution with a mean of 0 and a standard deviation of 1 to obtain the normalized crop growth analysis data, the growth environment variable data, and the growth image analysis data; perform weighted fusion on the normalized crop growth analysis data, the growth environment variable data, and the growth image analysis data to obtain the crop growth state aggregate.

[0034] Preferably, the training method of the growth trend evaluation model includes:

[0035] Divide the dataset into a training set, a validation set, and a test set, train the model and evaluate the model performance; use the deep learning library TensorFlow to build a growth trend evaluation model; the growth trend evaluation model includes an input layer, an LSTM layer, and an output layer; the input layer of the model is used to input the historical crop growth state aggregate; the output layer of the model is used to output the growth trend of the crops in the next n periods; the growth trend evaluation model is an LSTM model;

[0036] Define the loss function of the model, and use the L2-regularized mean squared error loss function to measure the difference between the predicted value and the true value of the model; use the training set to train the growth trend evaluation model, and update the model parameters through the backpropagation algorithm to minimize the loss function; use the validation set to evaluate the performance of the growth trend evaluation model by calculating the accuracy metric;

[0037] Select the Adam optimization algorithm as the optimizer, tune the model according to the performance feedback of the validation set, and stop adjusting the model parameters until the performance no longer improves or reaches the preset number of iterations; use the test set to evaluate the performance of the model in the prediction task, and use the pre-trained growth trend evaluation model to predict the current crop growth state aggregate to obtain the growth trend of the crops in the next n periods.

[0038] Preferably, the historical yield data includes the historical crop yield per unit area, the historical total crop yield, the historical crop yields under different environments, and the historical yield fluctuation data.

[0039] Preferably, the training method of the crop yield prediction model includes:

[0040] Divide the dataset into a training set, a validation set, and a test set, and build a crop yield prediction model; the crop yield prediction model includes an input layer, a GRU layer, a fully connected layer, and an output layer; the input layer of the crop yield prediction model is used to input the historical yield data and the growth trend of the crops in the next n periods; use the GRU layer to process the historical yield data and the growth trend of the crops in the next n periods, adjust the number of GRU layers and the number of neurons according to the task complexity, and provide an additional non-linear transformation through the fully connected layer; the output layer of the model is used to output the crop yield, and a single neuron is used to output the predicted value, and the identity function is used as the activation function; the crop yield prediction model is a gated recurrent unit model;

[0041] The mean absolute error is used as the loss function to measure the error between the predicted value and the actual value of the model; the training set data is used for model training, and the Adam optimizer is used to minimize the loss function; the validation set is used to evaluate the performance of the model, and the hyperparameters of the model are tuned until the model performance no longer improves or reaches the preset stop condition; the test set is used to evaluate the performance of the model in the prediction task, and the current yield data and the growth trend of the crops in the next n periods are input into the pre-trained crop yield prediction model to obtain the crop yield.

[0042] Preferably, the method for evaluating whether the crop is in the optimal growth trend according to the predicted crop yield includes:

[0043] A crop yield threshold is preset, and the predicted crop yield is compared with the preset crop yield threshold;

[0044] If the predicted crop yield is less than the preset crop yield threshold, it is determined that the crop is in the optimal growth trend;

[0045] If the predicted crop yield is greater than or equal to the preset crop yield threshold, it is determined that the crop is not in the optimal growth trend.

[0046] The technical effects and advantages of a method for analyzing the growth trend and evaluating the yield of crops according to the present invention are as follows:

[0047] Through weighted moving average and sliding window adjustment, the present invention can effectively remove short-term fluctuations caused by factors such as environmental changes and observation errors, making the crop growth data more stable and facilitating subsequent analysis; the smoothing process helps to suppress data random fluctuations caused by accidental factors (such as short-term climate changes, instantaneous environmental fluctuations, etc.), making the data better reflect long-term trends and laws; the size of the sliding window is dynamically adjusted according to the changes in the growth environment data, which helps to make a flexible response to different situations of environmental fluctuations; the adaptive window adjustment mechanism enables the smoothing process to be closely combined with the actual environmental changes, improving the adaptability of the filter and the stability of the model; by combining the environmental weight and the difference adjustment coefficient, the growth data of the crops can be weighted and adjusted according to the current environmental conditions, further enhancing the rationality and accuracy of data smoothing; through the smoothing process, unnecessary interference can be reduced, enabling the model to more accurately reflect the true growth trend of the crops. This is crucial for subsequent tasks such as growth prediction and pest warning; accurate crop growth data can provide strong support for agricultural decision-making, helping to formulate more scientific planting strategies, climate response measures and other optimization decisions;

[0048] Combining multiple growth environment factors such as meteorological data, soil data, air quality, and pests and diseases helps to comprehensively evaluate the growth of crops under different environmental conditions, and dynamically adjust the smoothing strategy according to the changes of these factors, enhancing the adaptability of the system; through the constant that controls the influence of environmental fluctuations on the sliding window size and the adaptive mechanism of the adjustment coefficient, the sensitivity to different types of environmental changes can be finely adjusted to better cope with complex growth environments; the smoothed data can be used for more stable and reliable analysis, avoiding misleading caused by data fluctuations, thereby enhancing the prediction effect and reliability of the model;

[0049] Kalman filtering can effectively reduce the noise introduced by factors such as measurement errors and environmental fluctuations, generating more accurate and reliable crop growth data. This makes the trends and laws of crop growth more obvious, which is beneficial for subsequent analysis; by dynamically adjusting the Kalman gain and updating the feedback and covariance matrix, the balance between prediction and actual observation can be optimized in real time, so that the prediction of crop growth status is more in line with the actual situation; as the growth environment data changes, the Kalman gain can be dynamically adjusted to optimize the response of the prediction model to environmental changes. During the process of crop growth monitoring, the adaptive adjustment of the Kalman gain helps the system better cope with sudden environmental changes (such as climate change, soil humidity, etc.) and provide real-time and accurate crop growth assessment. Brief Description of the Drawings

[0050] Figure 1 It is a schematic flow diagram of a method for analyzing the growth trend and evaluating the yield of crops;

[0051] Figure 2 It is a schematic structural diagram of a system for analyzing the growth trend and evaluating the yield of crops. Detailed Embodiments

[0052] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0053] Embodiment 1

[0054] Please refer to Figure 1 As shown, the method for analyzing the growth trend and evaluating the yield of crops in this embodiment includes:

[0055] S1. Real-time capture of crop growth monitoring data through the Internet of Things edge node cluster; the crop growth monitoring data includes crop growth data, growth environment data, and crop growth image data;

[0056] S2. Introduce an environmental weight mechanism to calculate the environmental weight at each time point based on the growth environment data of the time series, and dynamically adjust the adjustment coefficient of the weight influence degree through a deviation feedback mechanism; use an improved moving average algorithm to smooth the crop growth data in combination with the growth environment data, automatically adjust the window size according to environmental fluctuations, and obtain the smoothed crop growth data;

[0057] Use the Kalman filter algorithm to filter out the noise in the smoothed crop growth data and refine and generate the crop growth analysis data; process the growth environment data and the crop growth image data to obtain the growth environment variable data and the growth image analysis data;

[0058] S3. Integrate the crop growth analysis data, the growth environment variable data, and the growth image analysis data to obtain a crop growth state aggregate;

[0059] S4. Input the crop growth state aggregate into a pre-trained growth trend evaluation model to predict the growth trend of the crops in the next n time periods;

[0060] S5. Obtain the historical yield data, train and obtain a crop yield prediction model based on the historical yield data and the growth trend of the crops in the next n time periods, and predict the yield of the crops based on the crop yield prediction model;

[0061] S6. Evaluate whether the crops are in the best growth trend according to the predicted yield of the crops;

[0062] S7. If the crops are not in the best growth trend, feedback to the staff through the crop growth intelligent monitoring terminal for timely intervention.

[0063] The crop growth data includes growth form data and growth function data; the growth environment data includes meteorological data, soil data, air quality data, and pest and disease data; the crop growth image data includes crop vegetation index images, chlorophyll content images, plant growth height images, crop canopy coverage images, and leaf area index images.

[0064] The growth form data includes the growth height of crops, the stem diameter of crops, and the root depth; the stem diameter and root depth of crops include the leaf area index, photosynthesis efficiency, and canopy coverage of crops; the meteorological data includes growth temperature, precipitation, growth humidity, light intensity, and sunshine duration; the soil data includes soil moisture content, soil temperature, soil nutrient content, soil pH value, soil salt concentration; the air quality data includes atmospheric carbon dioxide concentration and pollutant gas concentration; the pest and disease data includes disease data and pest data; the disease data includes disease types, disease distribution areas, and crop incidence; the pest data includes pest types, pest density, and pest distribution areas; the soil nutrient content includes the concentrations of elements such as nitrogen, phosphorus, and potassium; the Internet of Things edge node cluster includes distributed soil sensors, leaf surface microenvironment probes, airborne hyperspectral imagers, and satellite remote sensing terminals.

[0065] The method for obtaining the smoothed crop growth data includes:

[0066] S31. Preset the growth form data as: where is the growth form data at time point t; x q is the q-th morphological index in the growth form data at time point t (such as the growth height of crops, leaf area index, root depth, etc.); q is the index of the morphological index, q = 1,..., N; N is the total number of morphological indices;

[0067] Preset the growth function data as: where is the growth function data at time point t; y j is the j-th functional index in the growth function data at time point t; j is the index of the functional index, j = 1,..., M; M is the total number of functional indices;

[0068] S32. Concatenate the growth form data at time point t with the growth function data at time point t to form the crop growth data, and the crop growth data is expressed as: x t ={x1, x2,..., x q ,..., x N , y1, y2,..., y j ,..., y M}; where x t is the crop growth data at time point t;

[0069] Preset the growth environment data at time point t as: where is the meteorological data at time point t; Soil data at time point t; Air quality data at time point t; Pest and disease data at time point t;

[0070] S33. According to the growth environment data E at time point t t , calculate the environmental weight for each time point t through the environmental weight calculation formula; the environmental weight calculation formula is: where ω i (E t ) is the weight corresponding to the growth environment data E at time point t t corresponding to the i-th time step, and this weight determines the importance of the environmental factors at this moment in the weighted moving average. The greater the weight, the greater the impact of the current environmental factors on the final calculation result; E max is the ideal growth environment data. The growth environment data includes growth temperature, humidity, light intensity, soil moisture content, etc. For example, at time point t, the ideal growth temperature of the growth environment data is 25°C and the humidity is 80, then E max ={25, 80}; ||E t -E max || is the difference between the growth environment data corresponding to time point t and the ideal growth environment data E max . This difference measures the deviation between the current environmental conditions and the ideal conditions. The greater the deviation, the greater the factor affecting the weight; ΔE t represents the change rate of the growth environment data from time point t - 1 to time point t; α is the adjustment coefficient that controls the influence degree of the difference between the growth environment data and the ideal growth environment data E max on the weight; β is the adjustment coefficient that controls the influence degree of the change rate of the growth environment data on the weight; i is the index of the time step;

[0071] It should be noted that the design basis of the environmental weight calculation formula includes: considering the difference between the environment and the ideal state: the growth of crops is significantly affected by the environment. The ideal growth environment is conducive to crop growth, while there are differences between the actual environment and it. ||E t -E max || is introduced in the formula. By calculating the difference between the actual growth environment data E t and the ideal growth environment data E max , the degree of deviation of the current environment from the ideal state is measured. For example, the ideal temperature is conducive to photosynthesis and nutrient absorption. The greater the deviation of the actual temperature from it, the greater the potential negative impact on crop growth. This difference should be reflected when calculating the environmental weight.

[0072] Pay attention to the environmental change rate: The environmental change has a great impact on crop growth. A stable environment is conducive to growth, and an environment with large fluctuations may bring stress responses. ΔEt It represents the change rate of the growth environment data from time t - 1 to time t, reflecting the environmental changes. Such as sudden temperature changes, large fluctuations in precipitation, etc., which may affect the physiological processes of crops. Incorporating this parameter into the environmental weight calculation formula can adjust the weight in a timely manner according to environmental changes.

[0073] Introduction of the adjustment coefficient: Different crops have different sensitivities to environmental differences and changes. By adjusting the two coefficients α and β, it can adapt to the growth characteristics and actual needs of different crops.

[0074] By comprehensively considering the difference between the environment and the ideal state and the environmental change rate, the environmental weight is accurately calculated, enabling the weight to reflect the suitability and potential impact of the current environment on crop growth. A high weight means that the environment is closer to the ideal state or the change is more stable, which is beneficial to crop growth; otherwise, it is not. When smoothing the crop growth data, combined with the environmental weight, different weights can be assigned to the data at different time points according to the environmental situation. When the environment is suitable and stable, the corresponding data weight is high and has a great impact on the result in the smoothing process; when the environment is poor or changes greatly, the data weight is low, reducing its interference with the smoothing result and making the smoothed data better reflect the true growth trend of the crops. The environmental weight calculation formula comprehensively considers the difference between the environment and the ideal state and the environmental change rate, comprehensively reflecting the impact of the environment on crop growth. Compared with measuring the environmental impact by a single factor, it is more in line with the actual situation and can provide a more accurate basis for crop growth analysis.

[0075] Compared with the technical effects of the prior art:

[0076] More accurate environmental impact assessment: The prior art may not fully consider the difference between the environment and the ideal state and the environmental change rate, or only consider a single factor. The environmental weight calculation formula comprehensively integrates these factors, can more accurately evaluate the impact of the environment on crop growth, and provide more accurate information for agricultural decision-making. For example, when judging whether measures need to be taken to improve the environment, the assessment based on this formula is more valuable. Optimize the data processing effect: Combining the environmental weight for data smoothing can more effectively remove noise and interference, highlighting the true trend of crop growth. When predicting the growth trend and yield of crops, the model trained with the data processed based on this formula is more accurate, improving the reliability of the prediction. For example, when predicting the crop yield, it can more accurately reflect the impact of environmental factors on the yield and reduce the prediction error. S34. The adjustment coefficient α that regulates the influence degree of the difference between the controlled growth environment data and the ideal growth environment data E max is dynamically adjusted through the difference adjustment coefficient self-adaptive formula. The difference adjustment coefficient self-adaptive formula is: where m is the number of types of growth environment data; is the change rate of the c-th type of growth environment data at time point t; b is a constant that adjusts the influence strength of the change rate on the adjustment coefficient; p is an exponential factor that controls the influence degree of the cumulative change rate on the adjustment coefficient; c is the index of the types of growth environment data, and its value range is from 1 to m;

[0077] Dynamically adjust the adjustment coefficient α according to the change rate of the environmental data to ensure that in a rapidly changing environment, the adjustment coefficient can automatically decrease, thus avoiding over-reliance on environmental changes. On the contrary, when the environment is stable, the adjustment coefficient can be maintained at a high level to enhance the influence on the weights; through the constants b and p, the differential adjustment coefficient adaptive formula can be flexibly adjusted according to the actual environmental needs, adapt to different types and change rates of growth environment data, be well adaptable in different agricultural scenarios, effectively take into account the types and change situations of environmental factors, and achieve reasonable smoothing processing of crop growth data through the adjustment coefficient.

[0078] Different degrees of environmental change require different degrees of adjustment coefficient α to balance the influence of environmental differences on the weights of crop growth data. When the environmental change is small, it is hoped that α remains at a high level so that the environmental difference has a greater influence on the weights; when the environmental change is drastic, α should decrease to avoid over-reliance on environmental change data. The formula realizes flexible adjustment of α through two parameters b and p. By dynamically adjusting α, the environmental weight calculation is improved, the weighted moving average process is optimized, and the quality of crop growth data smoothing processing is improved. The smoothed data can more accurately reflect the true growth trend of the crops, providing reliable data support for subsequent growth trend analysis and yield prediction.

[0079] Compared with the technical effects of the prior art:

[0080] Enhance environmental adaptability: The prior art may adopt a fixed adjustment coefficient or a simple adjustment method, which cannot adapt to complex and changeable environments. This formula can dynamically adjust α in real time according to environmental changes, making the system more sensitive and accurate in responding to environmental changes, and enhancing the adaptability of the crop growth trend analysis and yield evaluation system to different environmental conditions.

[0081] Improve data processing accuracy: In terms of data processing, this formula optimizes the environmental weight calculation, makes the weighted moving average more accurate, effectively reduces the interference of environmental fluctuations on the data, and improves the data smoothing processing effect. Based on the processed data for growth trend prediction and yield assessment, the accuracy of prediction and assessment can be significantly improved.

[0082] For example, the initial adjustment coefficient α is 0.8, the number of types of growth environment data m is 3, and the change rates of each type of growth environment data are respectively The constant b that adjusts the influence strength of the change rate on the adjustment coefficient is 2; the exponential factor p that controls the cumulative influence degree of the change rate on the adjustment coefficient is 1; the cumulative sum of the calculated environmental change rate is Then, after dynamic adjustment

[0083] S35. Use the weighted moving average formula to smooth the crop growth data to obtain the smoothed crop growth data; the weighted moving average formula is: Among them, WA t is the smoothed crop growth data at time point t; x t-a is the crop growth data at time point t - a, which is the unsmoothed growth data and is used to generate the smoothed result; γ is the size of the moving window; a is the index of the moving window, and its value range is from 0 to γ - 1;

[0084] S36. Dynamically adjust the size γ of the moving window through the moving window adjustment formula. The moving window adjustment formula is: Among them, γ max is the maximum value of the moving window; ω avg is the average weight of the growth environment data; d is the constant that controls the influence of environmental fluctuations on the size of the moving window; is the floor function;

[0085] The moving window adjustment formula dynamically adjusts the size of the moving window according to the changes in the environment to better cope with external changes. When the environmental fluctuations are large, the window becomes smaller, which helps to capture subtle changes; while when the environment is relatively stable, the window is larger, which helps to smooth out the noise in the data; by adjusting the constant d that controls the influence of environmental fluctuations on the size of the moving window, the window is reduced in the case of large environmental fluctuations to adapt to rapidly changing data, and the window is increased when the environment is relatively stable to ensure data smoothing;

[0086] By dynamically adjusting the size of the moving window, the data processing process can better adapt to different environmental fluctuation situations. When the environment changes violently, the window size is reduced to enhance the ability to capture data details; when the environment is stable, the window size is increased to improve the smoothing effect of the data, so as to more accurately reflect the characteristics of the crop growth data and provide more reliable data support for subsequent analysis and prediction.

[0087] Optimize the weighted moving average effect: The reasonable adjustment of the moving window size helps to optimize the effect of the weighted moving average. An appropriate window size can make the weighted moving average remove noise while retaining the useful information of the data to the greatest extent, avoid data distortion or trend misjudgment caused by inappropriate window size, and improve the quality of the smoothing process of crop growth data.

[0088] Technical effects compared with the prior art:

[0089] Enhance data processing flexibility: In the prior art, the size of the sliding window is usually fixed and cannot be adaptively adjusted according to environmental changes. The sliding window adjustment formula of the present invention can dynamically change the window size and can more effectively process crop growth data under different environmental conditions, improving the flexibility and adaptability of data processing.

[0090] Improve prediction accuracy: The dynamically adjusted sliding window can better process data, making the subsequent prediction of crop growth trends and yield evaluation based on these data more accurate. By more precisely capturing data changes and trends, the prediction errors caused by improper data processing are reduced, providing a more reliable basis for agricultural production decisions and helping to improve the efficiency and benefits of agricultural production.

[0091] Adapt to complex environmental changes: When facing the complex and changeable crop growth environment, this formula can respond to environmental changes in a timely manner and adjust the window size. In contrast, the fixed window of the prior art is difficult to adapt to this complex environment and easily leads to data processing deviations. The formula of the present invention can better adapt to complex environmental changes, ensuring the stability and reliability of the crop growth analysis and yield evaluation system.

[0092] For example, the maximum value γ of the sliding window max is 10, the growth environment data E at time point t t corresponds to the weight ω at the i-th time step i (E t ) is 7; the average weight of the growth environment data is 5; the constant d that controls the influence of environmental fluctuations on the sliding window size is 0.2, then the sliding window size is

[0093] A method for filtering noise from the smoothed crop growth data using the Kalman filter algorithm and refining and generating crop growth analysis data includes:

[0094] Define a crop growth state transition model, and reflect the change process of the crop growth state over time through the crop growth state transition model. Predict the crop growth state at the current time point based on the crop growth state at the previous time point and the control input.

[0095] The crop growth state transition model is: Wherein, is the predicted value of the crop growth state at the current time point t; A is the state transition matrix, indicating the change relationship of the crop growth state from time point t - 1 to time point t; is the predicted value of the crop growth state at time point t-1; B is the control input matrix, representing the external control input vector u t the impact on the crop growth state; u t is the external control input vector;

[0096] Crop growth has specific laws and biological characteristics, and there are internal connections in the changes of various indicators at different growth stages. Taking corn growth as an example, the changes of growth indicators such as plant height, number of leaves, and stem diameter follow certain laws at different growth stages. Through long-term observation and research on the corn growth process, researchers have grasped the change relationships between these indicators at adjacent time points. For example, during the jointing stage of corn, there is a certain correlation between the current plant height and the plant height and number of leaves at the previous time point, because the photosynthesis of leaves provides energy and substances for plant height growth. By mathematically modeling this internal connection and representing the influence relationships between various growth indicators in the form of matrix elements, the state transition matrix A is constructed. In agricultural production, external control measures such as irrigation, fertilization, and pest control have important impacts on crop growth. Through long-term practical experience, agricultural experts understand the reasonable dosages and implementation times of different control measures at different crop growth stages, as well as the impact degrees of these measures on the crop growth state. For example, during the tillering stage of rice growth, appropriate application of nitrogen fertilizer can promote tillering and increase the number of effective panicles. Based on this experience, the external control inputs (such as fertilization amount, irrigation amount, etc.) are associated with the crop growth state indicators, and the influence coefficients of different control inputs on each growth state indicator are determined. These coefficients form the elements of the control input matrix B.

[0097] The predicted value of the crop growth state at the current time point t is updated through the state update formula. The state update formula is: where, is the predicted value of the updated crop growth state; K t is the Kalman gain, which controls the balance between the predicted value of the crop growth state and the actual observed value of the crop growth state; z t is the actual observed value of the crop growth state; H is the observation matrix, representing the mapping from the predicted value of the crop growth state to the actual observed value of the crop growth state;

[0098] The Kalman gain is calculated through the Kalman gain calculation formula, and the Kalman gain is used to balance the weights between the predicted value of the crop growth state and the actual observed value of the crop growth state. The Kalman gain calculation formula is: where, P t is the predicted error covariance matrix; R t (ΔE t ) is the measurement noise covariance matrix dynamically adjusted according to the change rate of the growth environment data; H Tis the transpose of the observation matrix;

[0099] The error covariance matrix is updated through the covariance matrix update formula, reflecting the uncertainty of the updated crop growth state; the covariance matrix update formula is: where is the updated error covariance matrix; I is the identity matrix; the identity matrix I is a special square matrix. Its main diagonal elements are all 1, and the rest of the elements are all 0.

[0100] The crop growth data denoised by Kalman filtering is output as the final result to obtain the crop growth analysis data.

[0101] The method for processing the growth environment data and the crop growth image data to obtain the growth environment variable data and the growth image analysis data includes:

[0102] The density clustering algorithm is used to identify and remove the outliers in the growth environment data and the crop growth image data included in the crop growth monitoring data, obtaining the growth environment variable data and the growth image analysis data.

[0103] The method for obtaining the crop growth state aggregate includes:

[0104] The normalized crop growth analysis data, the growth environment variable data, and the growth image analysis data are fused through a weighted model to obtain the crop growth state aggregate;

[0105] The crop growth analysis data is denoted as O1, the growth environment variable data is denoted as O2, and the growth image analysis data is denoted as O3; the weighted model is: QZ = O1·δ1 + O2·δ2 + O3·δ3; where QZ is the crop growth state aggregate; δ1 is the weight coefficient of the crop growth analysis data; δ2 is the weight coefficient of the growth environment variable data; δ3 is the weight coefficient of the growth image analysis data.

[0106] According to the importance of the normalized crop growth analysis data in the crop growth state aggregate, the initial value range of the weight factor δ1 is set to 0.2 - 0.4, and the initial default value is 0.2; using the historical crop growth state aggregate, analyze the influence degree of the normalized crop growth analysis data on the growth trend of the crops; if the normalized crop growth analysis data has a greater influence on the growth trend of the crops (such as the proportion of the influence on the evaluation result exceeds 50%), then increase the value of δ1; otherwise, decrease its value;

[0107] According to the importance of the normalized growth environment variable data in the crop growth state aggregate, set the initial value range of the weight factor δ2 to be 0.4 - 0.6, and the initial default value is 0.5; use the historical crop growth state aggregate to analyze the influence degree of the normalized growth environment variable data on the growth trend of crops; if the normalized growth environment variable data has a greater impact on the growth trend of crops (such as the change of growth environment characteristics directly affects the growth trend of crops), then increase the value of δ2; otherwise, decrease its value.

[0108] According to the importance of the normalized growth image analysis data in the crop growth state aggregate, set the initial value range of the weight factor δ3 to be 0.3 - 0.5, and the initial default value is 0.3; determine its influence degree by analyzing the correlation between the normalized growth image analysis data and the growth trend of crops (such as the role of growth image analysis data in identifying the growth trend of crops); if the normalized growth image analysis data plays a key role in the actual growth process of crops, then increase the value of δ3; if its influence is weak, then appropriately decrease it.

[0109] Gradually optimize the values of the weight coefficient δ1, the weight coefficient δ2, and the weight coefficient δ3 according to the real-time feedback of the model to improve the generalization ability of the model.

[0110] The training method of the growth trend evaluation model includes:

[0111] Divide the data set into a training set, a validation set, and a test set, train the model and evaluate the model performance; the sample set is a subset of the data set, and each sample set includes the historical crop growth state aggregate and the corresponding growth trend of crops in the next n time periods; use the deep learning library TensorFlow to construct the growth trend evaluation model; the growth trend evaluation model includes an input layer, an LSTM layer, and an output layer; the input layer of the model is used to input the historical crop growth state aggregate; the output layer of the model is used to output the growth trend of crops in the next n time periods; the growth trend evaluation model is an LSTM model.

[0112] Define the loss function of the model, use the L2 regularization mean square error loss function to measure the difference between the predicted value and the true value of the model; use the training set to train the growth trend evaluation model, update the model parameters through the backpropagation algorithm to minimize the loss function; use the validation set to evaluate the performance of the growth trend evaluation model by calculating the accuracy index.

[0113] The Adam optimization algorithm is selected as the optimizer, and the model is tuned according to the performance feedback of the validation set. The model parameters are adjusted until the performance no longer improves or reaches the preset number of iterations. The test set is used to evaluate the performance of the model in the prediction task, and the pre-trained growth trend evaluation model is used to predict the current crop growth state aggregate to obtain the growth trend of the crops in the next n periods.

[0114] Historical yield data includes the historical crop yield per unit area, the total historical crop yield, the crop yields in different historical environments, and historical yield fluctuation data. The historical yield data is retrieved from the historical crop database through the crop growth intelligent monitoring terminal, and the crop growth intelligent monitoring terminal interacts with the historical crop database through a standardized API interface.

[0115] The construction method of the historical crop database includes: installing sensors in the farmland to collect the crop yield per unit area in real time, combining the planting and harvesting information manually recorded by farmers and the annual agricultural production reports of government statistical departments, and using a mature database management system (such as MySQL, PostgreSQL) to build the historical crop database.

[0116] The historical crop database is used to store the historical crop yield per unit area, the total historical crop yield, the crop yields in different historical environments, and historical yield fluctuation data. Through the API interface and the Internet of Things gateway, new data is synchronized to the database in real time, and the historical data query and call functions are configured. Users can directly access the database information through the crop growth intelligent monitoring terminal.

[0117] The training method of the crop yield prediction model includes:

[0118] The dataset is divided into a training set, a validation set, and a test set to construct a crop yield prediction model. The sample set is a subset of the dataset, and each sample set includes historical yield data, the growth trend of the crops in the next n periods, and the corresponding crop yield. The crop yield prediction model includes an input layer, a GRU layer, a fully connected layer, and an output layer.

[0119] The input layer of the crop yield prediction model is used to input historical yield data and the growth trend of the crops in the next n periods. The GRU layer is used to process the historical yield data and the growth trend of the crops in the next n periods. The number of GRU layers and the number of neurons are adjusted according to the task complexity, and additional non-linear transformations are provided through the fully connected layer. The output layer of the model is used to output the crop yield, and a single neuron is used to output the predicted value, and the identity function is used as the activation function. The crop yield prediction model is a gated recurrent unit model.

[0120] Use the mean absolute error as the loss function to measure the error between the predicted value and the actual value of the model; use the training set data to train the model and minimize the loss function through the Adam optimizer; use the validation set to evaluate the performance of the model and tune the hyperparameters of the model until the model performance no longer improves or reaches the preset stopping condition; use the test set to evaluate the performance of the model in the prediction task, and input the current yield data and the growth trend of the crops in the next n periods into the pre-trained crop yield prediction model to obtain the crop yield.

[0121] The method for evaluating whether the crops are in the best growth trend according to the predicted crop yield includes:

[0122] Preset the crop yield threshold, and compare the predicted crop yield with the preset crop yield threshold;

[0123] If the predicted crop yield is less than the preset crop yield threshold, it is determined that the crops are in the best growth trend;

[0124] If the predicted crop yield is greater than or equal to the preset crop yield threshold, it is determined that the crops are not in the best growth trend.

[0125] The preset crop yield threshold is set by the staff. The yields of different crops are collected through the intelligent crop growth monitoring terminal, and the average value of the yields of multiple crops is taken as the preset crop yield threshold.

[0126] In this embodiment, through weighted moving average and sliding window adjustment, short-term fluctuations caused by factors such as environmental changes and observation errors can be effectively removed, making the crop growth data more stable and facilitating subsequent analysis; the smoothing process helps to suppress the random fluctuations of data caused by accidental factors (such as short-term climate changes, instantaneous environmental fluctuations, etc.), making the data better reflect long-term trends and laws; the size of the sliding window is dynamically adjusted according to the changes in the growth environment data, which helps to make a flexible response to different situations of environmental fluctuations; the adaptive window adjustment mechanism enables the smoothing process to be closely combined with the actual environmental changes, improving the adaptability of the filter and the stability of the model; by combining the environmental weight and the difference adjustment coefficient, the growth data of the crops can be weighted and adjusted according to the current environmental conditions, further enhancing the rationality and accuracy of data smoothing; through the smoothing process, unnecessary interference can be reduced, enabling the model to more accurately reflect the true growth trend of the crops. This is crucial for subsequent growth prediction, pest warning and other tasks; accurate crop growth data can provide strong support for agricultural decision-making, helping to formulate more scientific planting strategies, climate response measures and other optimization decisions;

[0127] Combining multiple growth environment factors such as meteorological data, soil data, air quality, and pests and diseases helps to comprehensively evaluate the growth of crops under different environmental conditions, and dynamically adjust the smoothing strategy according to the changes of these factors to enhance the adaptability of the system; through the constant controlling the influence of environmental fluctuations on the size of the sliding window and the adaptive mechanism of the adjustment coefficient, the sensitivity to different types of environmental changes can be finely adjusted to better cope with complex growth environments; the smoothed data can be used for more stable and reliable analysis, avoiding misleading caused by data fluctuations, thereby enhancing the prediction effect and reliability of the model;

[0128] Kalman filtering can effectively reduce the noise introduced by factors such as measurement errors and environmental fluctuations, and generate more accurate and reliable crop growth data. This makes the trends and laws of crop growth more obvious, which is beneficial to subsequent analysis; by dynamically adjusting the Kalman gain and updating the feedback and covariance matrix, the balance between prediction and actual observation can be optimized in real time, so that the prediction of crop growth status is more in line with the actual situation; as the growth environment data changes, the Kalman gain can be dynamically adjusted to optimize the response of the prediction model to environmental changes. During the process of crop growth monitoring, the adaptive adjustment of the Kalman gain helps the system better cope with sudden environmental changes (such as climate change, soil humidity, etc.) and provide real-time and accurate crop growth assessment.

[0129] Embodiment 2

[0130] Please refer to Figure 2 As shown, for the parts not described in detail in this embodiment, refer to the description content of Embodiment 1. A crop growth trend analysis and yield evaluation system is provided, including:

[0131] A data acquisition unit for capturing crop growth monitoring data in real time through an Internet of Things edge node cluster; the crop growth monitoring data includes crop growth data, growth environment data, and crop growth image data;

[0132] A data processing unit for introducing an environmental weight mechanism, calculating the environmental weight at each time point according to the growth environment data in the time series, and dynamically adjusting the adjustment coefficient of the weight influence degree through a deviation feedback mechanism; using an improved moving average algorithm to smooth the crop growth data in combination with the growth environment data, automatically adjusting the window size according to environmental fluctuations, and obtaining the smoothed crop growth data;

[0133] Using the Kalman filtering algorithm to filter the noise of the smoothed crop growth data and refine the crop growth analysis data; processing the growth environment data and the crop growth image data to obtain growth environment variable data and growth image analysis data;

[0134] A data fusion unit for fusing crop growth analysis data, growth environment variable data, and growth image analysis data to obtain a crop growth state aggregate;

[0135] A growth trend analysis unit for importing the crop growth state aggregate into a pre-trained growth trend evaluation model to predict the growth trend of crops in the next n time periods;

[0136] A yield prediction unit for obtaining historical yield data, training a crop yield prediction model based on the historical yield data and the growth trend of crops in the next n time periods, and predicting the yield of crops based on the crop yield prediction model;

[0137] A growth trend evaluation unit for evaluating whether the crops are in the best growth trend according to the predicted yield of the crops;

[0138] An optimization feedback unit, if the crops are not in the best growth trend, feeds back to the staff through the crop growth intelligent monitoring terminal for timely intervention.

[0139] Since the electronic device introduced in this embodiment is the electronic device used to implement a method for analyzing the growth trend and evaluating the yield of a crop in an embodiment of the present application, based on the method for analyzing the growth trend and evaluating the yield of a crop introduced in an embodiment of the present application, those skilled in the art can understand the specific implementation manner and various variations of the electronic device in this embodiment. Therefore, the implementation of how this electronic device realizes the method in the embodiment of the present application will not be described in detail here. As long as those skilled in the art implement the electronic device used in a method for analyzing the growth trend and evaluating the yield of a crop in an embodiment of the present application, it falls within the protection scope of the present application.

[0140] The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by collecting a large amount of data for software simulation to obtain a formula that is closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0141] The above is only a preferred implementation manner of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be noted that for ordinary technical users in the technical field, several improvements and refinements made without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A method for analyzing crop growth and yield evaluation, characterized in that: include: S1. Real-time capture of crop growth monitoring data through IoT edge node clusters; crop growth monitoring data includes crop growth data, growth environment data, and crop growth image data; S2. Introduce an environmental weight mechanism, calculate the environmental weight of each time point based on the growth environment data of the time series, and dynamically adjust the adjustment coefficient of the weight influence through the deviation feedback mechanism; use the improved sliding average algorithm combined with the growth environment data to smooth the crop growth data, automatically adjust the window size according to the environmental fluctuations, and obtain the smoothed crop growth data; Use the Kalman filter algorithm to filter out noise from the smoothed crop growth data, and refine and generate crop growth analytical data; process the growth environment data and crop growth image data to obtain growth environment variable data and growth image analytical data; S3, fusing the crop growth analysis data, the growth environment variable data and the growth image analysis data to obtain a crop growth state aggregate; S4, inputting the crop growth state aggregate into the pre-trained growth state assessment model to predict the growth state of the crops in the next n periods of time; S5. Obtain historical yield data, train a crop yield prediction model based on the historical yield data and the growth trend of crops in the next n periods of time, and predict the yield of crops based on the crop yield prediction model; S6. Based on the predicted crop yields, assess whether the crops are in the best growth state; S7. If the crops are not in the best growth state, the information will be fed back to the staff through the crop growth intelligent monitoring terminal for timely intervention.

2. A method for analyzing crop growth and yield assessment according to claim 1, characterized in that: The crop growth data includes growth morphology data and growth function data; the growth environment data includes meteorological data, soil data, air quality data and pest and disease data; the crop growth image data includes crop vegetation index images, chlorophyll content images, plant growth height images, crop canopy cover images and leaf area index images.

3. A method for analyzing crop growth and yield assessment according to claim 2, characterized in that: The method for obtaining the smoothed crop growth data includes: S31. The preset growth morphology data is: in, is the growth morphology data at time point t; x q is the qth morphological index in the growth morphological data at time point t; q is the index of the morphological index, q=1,...,N; N is the total number of morphological indexes; The default growth function data is: in, is the growth function data at time point t; y j is the jth functional index in the growth function data at time point t; j is the index of the functional index, j=1,...,M; M is the total number of functional indexes; S32, the growth morphology data at time point t with the growth function data at time point t Spliced ​​together to form crop growth data, crop growth data is expressed as: x t ={x1,x2,...,x q ,...,x N ,y1,y2,...,y j ,...,y M }; where x t is the crop growth data at time point t; The growth environment data preset at time point t is: in, is the meteorological data at time point t; is the soil data at time point t; is the air quality data at time point t; is the pest and disease data at time point t; S33, according to the growth environment data E at time point t t , the environmental weight at each time point t is calculated using the environmental weight calculation formula; the environmental weight calculation formula is: Among them, ω i (E t ) is the growth environment data E at time point t t The weight corresponding to the i-th time step; E max is the ideal growth environment data; ||E t -E max || is the growth environment data corresponding to time point t and the ideal growth environment data E max The difference between t represents the rate of change of the growth environment data from time point t-1 to time point t; α is the difference between the control growth environment data and the ideal growth environment data E max The difference between them is the adjustment coefficient of the influence of the weight; β is the adjustment coefficient of the influence of the change rate of the growth environment data on the weight; i is the index of the time step; S34, the control growth environment data and the ideal growth environment data E are adjusted by the adaptive formula of the difference adjustment coefficient max The difference between them dynamically adjusts the adjustment coefficient α of the weight influence, and the adaptive formula of the difference adjustment coefficient is: Among them, m is the number of types of growth environment data; is the change rate of the cth type of growth environment data at time point t; b is the constant that adjusts the influence of the change rate on the adjustment coefficient; p is the exponential factor that controls the influence of the cumulative change rate on the adjustment coefficient; c is the index of the type of growth environment data; S35. Use a weighted moving average formula to smooth the crop growth data; the weighted moving average formula is: Among them, WA t is the smoothed crop growth data at time point t; x t-a is the crop growth data at time point ta; γ is the sliding window size; a is the index of the sliding window; S36. Dynamically adjust the sliding window size γ through a sliding window adjustment formula. The sliding window adjustment formula is: Among them, γ max is the maximum value of the sliding window; ω avg is the average weight of the growth environment data; d is a constant that controls the effect of environmental fluctuations on the sliding window size; is the floor function.

4. A method for analyzing crop growth and yield assessment according to claim 3, characterized in that: The method of using the Kalman filter algorithm to filter out noise from the smoothed crop growth data and to refine and generate crop growth analytical data includes: Define a crop growth state transfer model, which reflects the change of crop growth state over time, and predicts the crop growth state at the current time point based on the crop growth state and control input at the previous time point; By updating the predicted value of the crop growth state at the current time point t, the Kalman gain is calculated, and the weight between the predicted value of the crop growth state and the actual observed value of the crop growth state is balanced by the Kalman gain; and the error covariance matrix in the Kalman gain is updated to reflect the uncertainty of the updated crop growth state; the crop growth data denoised by the Kalman filter is output as the final result to obtain the crop growth analysis data.

5. A method for analyzing crop growth and yield assessment according to claim 4, characterized in that: The method of processing the growth environment data and the crop growth image data to obtain the growth environment variable data and the growth image analysis data comprises: The density clustering algorithm is used to identify and eliminate outliers in the growth environment data and crop growth image data included in the crop growth monitoring data, and obtain growth environment variable data and growth image analysis data.

6. A method for analyzing crop growth and yield assessment according to claim 5, characterized in that: The method for obtaining the crop growth status aggregate comprises: The crop growth analysis data, growth environment variable data and growth image analysis data are normalized by standard deviation and converted into standard normal distribution with mean 0 and standard deviation 1 to obtain normalized crop growth analysis data, growth environment variable data and growth image analysis data; the normalized crop growth analysis data, growth environment variable data and growth image analysis data are weighted fused to obtain crop growth status aggregate.

7. A method for analyzing crop growth and evaluating yield according to claim 6, characterized in that: The training method of the growth status assessment model includes: The data set is divided into a training set, a validation set and a test set, and the model is trained and the model performance is evaluated; the deep learning library TensorFlow is used to build a growth trend assessment model; the growth trend assessment model includes an input layer, an LSTM layer and an output layer; the input layer of the model is used to input the historical crop growth state aggregate; the output layer of the model is used to output the growth trend of crops in the next n period of time; the growth trend assessment model is an LSTM model; Define the loss function of the model and use the L2 regularized mean square error loss function to measure the difference between the model's predicted value and the true value; use the training set to train the growth trend assessment model and update the model parameters through the back propagation algorithm to minimize the loss function; use the validation set to evaluate the performance of the growth trend assessment model by calculating the accuracy index; The Adam optimization algorithm is selected as the optimizer, and the model is tuned according to the performance feedback of the validation set. The model parameters are adjusted until the performance no longer improves or the preset number of iterations is reached. The test set is used to evaluate the performance of the model in the prediction task, and the pre-trained growth status evaluation model is used to predict the current crop growth status aggregate to obtain the growth status of crops in the next n period of time.

8. A method for analyzing crop growth and yield assessment according to claim 7, characterized in that: The historical yield data include historical crop yield per unit area, historical total crop yield, crop yield under different historical environments and historical yield fluctuation data.

9. A method for analyzing crop growth and yield assessment according to claim 8, characterized in that: The training method of the crop yield prediction model includes: The data set is divided into a training set, a validation set and a test set, and a crop yield prediction model is constructed; the crop yield prediction model includes an input layer, a GRU layer, a fully connected layer and an output layer; the input layer of the crop yield prediction model is used to input historical yield data and the growth trend of crops in the next n period of time; the GRU layer is used to process the historical yield data and the growth trend of crops in the next n period of time, and the number of GRU layers and the number of neurons are adjusted according to the complexity of the task, and the fully connected layer is used to provide additional nonlinear transformation; the model output layer is used to output the yield of crops, and the predicted value is output through a neuron, and the identity function is used as the activation function; the crop yield prediction model is a gated recurrent unit model; The mean absolute error is used as the loss function to measure the error between the model's predicted value and the actual value. The training set data is used to train the model, and the loss function is minimized through the Adam optimizer. The validation set is used to evaluate the performance of the model and the hyperparameters of the model are tuned until the model performance no longer improves or reaches the preset stopping condition. The test set is used to evaluate the performance of the model in the prediction task, and the current yield data and the growth trend of crops in the next n periods of time are input into the pre-trained crop yield prediction model to obtain the crop yield.

10. A method for analyzing crop growth and evaluating yield according to claim 9, characterized in that: The method for evaluating whether crops are in an optimal growth state based on the predicted crop yields includes: A yield threshold of the crop is preset, and the predicted yield of the crop is compared with the preset yield threshold of the crop; If the predicted crop yield is less than the preset crop yield threshold, the crop is determined to be in the best growth state; If the predicted crop yield is greater than or equal to the preset crop yield threshold, it is determined that the crop is not in the optimal growth state.

Images