Agricultural assistance data evaluation method and device, electronic equipment, medium and program

Abstract

The embodiment of the invention discloses an agricultural assistance data evaluation method and device, electronic equipment, a medium and a program, and relates to the technical field of big data, artificial intelligence and financial science and technology, and the method comprises the steps: obtaining evaluation association reference data of agricultural assistance data in a target evaluation agricultural region; wherein the evaluation associated reference data comprises yield prediction data, agricultural production data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data and climate risk data; and inputting the evaluation associated reference data into a multivariable linear regression model, and evaluating the agricultural assistance data of the target evaluation agricultural area through the multivariable linear regression model according to the evaluation associated reference data and the weight of the evaluation associated reference data. According to the technical scheme of the embodiment of the invention, the efficiency, accuracy, scientificity and rationality of agricultural assistance data evaluation can be improved.

Description

Technical Field

[0001] Embodiments of the present invention relate to the fields of big data, artificial intelligence and financial technology, and in particular to an agricultural assistance data evaluation method, device, electronic device, medium and program. Background Art

[0002] As an important support for agricultural development, agricultural credit and agricultural subsidies and other agricultural assistance policies play an irreplaceable role in promoting agricultural modernization, increasing farmers' income and ensuring food security. The demand for various agricultural assistance is also growing. For agricultural assistance policies, there are often three significant characteristics: flexibility, risk and cyclicality. Taking agricultural credit as an example, agricultural credit institutions need to provide farmers with stable financial support under the premise of controllable risks to cope with natural disasters and market fluctuations. At the same time, agricultural credit institutions also need to reasonably arrange loan terms and repayment methods according to the cyclical characteristics of agricultural production to ensure that farmers can repay on time. Therefore, before implementing agricultural assistance policies, it is often necessary to conduct preliminary inspections and assessments on the assistance targets in order to provide scientific and reasonable assistance policies for the assistance targets, so as to better support agricultural development and help farmers improve production efficiency and quality.

[0003] Preliminary inspection and evaluation of the aid recipients often requires prediction of their production values, etc., and evaluation must be combined with other related influencing factors. However, the current inspection and evaluation methods for aid recipients are relatively simple, usually using manual surveys and sampling to estimate production, that is, selecting individuals from a group with a certain probability, calculating the predicted production value, and then using mathematical statistical probability methods to infer the overall production.

[0004] However, there are both advantages and disadvantages to using manual surveys and sampling to estimate yields. It has the advantages of real data and high reliability, but the disadvantages of being time-consuming, labor-intensive and ineffective are more prominent. At the same time, traditional ground measurement methods are difficult to provide efficient and accurate data in complex terrain and irregular farmland. Summary of the invention

[0005] The embodiments of the present invention provide an agricultural assistance data evaluation method, device, electronic device, medium and program, which can improve the efficiency, accuracy, scientificity and rationality of agricultural assistance data evaluation.

[0006] According to one aspect of the present invention, there is provided an agricultural assistance data evaluation method, comprising:

[0007] Acquire assessment-related reference data of agricultural assistance data in the target assessment agricultural area; wherein the assessment-related reference data includes yield forecast data, agricultural production data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data and climate risk data;

[0008] The evaluation-related reference data are input into a multivariate linear regression model, and the agricultural assistance data of the target evaluation agricultural area is evaluated through the multivariate linear regression model based on the evaluation-related reference data and the weights of the evaluation-related reference data.

[0009] According to another aspect of the present invention, there is provided an agricultural assistance data evaluation device, comprising:

[0010] An evaluation-related reference data acquisition module is used to acquire evaluation-related reference data of agricultural assistance data in a target evaluation agricultural area; wherein the evaluation-related reference data includes yield forecast data, historical credit data and climate risk data;

[0011] The agricultural assistance data evaluation module is used to input the evaluation-related reference data into a multivariate linear regression model, and evaluate the agricultural assistance data of the target evaluation agricultural area through the multivariate linear regression model based on the evaluation-related reference data and the weights of the evaluation-related reference data.

[0012] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0013] at least one processor; and

[0014] a memory communicatively connected to the at least one processor; wherein,

[0015] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the agricultural assistance data evaluation method described in any embodiment of the present invention.

[0016] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the agricultural assistance data evaluation method described in any embodiment of the present invention when executed.

[0017] According to another aspect of the present invention, there is also provided a computer program product, comprising a computer program, which, when executed by a processor, implements the agricultural assistance data evaluation method described in any embodiment of the present invention.

[0018] The embodiment of the present invention obtains evaluation-related reference data such as yield forecast data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data and climate risk data of agricultural assistance data in the target evaluated agricultural area, so as to input the multi-dimensional evaluation-related reference data into a multivariate linear regression model, thereby evaluating the agricultural assistance data of the target evaluated agricultural area according to the multi-dimensional evaluation-related reference data and the weights of the evaluation-related reference data through the multivariate linear regression model, thereby solving the problems of low efficiency and accuracy of existing agricultural assistance data evaluation, and being able to improve the efficiency, accuracy, scientificity and rationality of agricultural assistance data evaluation.

[0019] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 is a flow chart of an agricultural assistance data evaluation method provided by Embodiment 1 of the present invention;

[0022] Figure 2 is a flow chart of an agricultural assistance data evaluation method provided by Embodiment 2 of the present invention;

[0023] Figure 3 is a schematic diagram of an agricultural assistance data evaluation device provided by Embodiment 3 of the present invention;

[0024] Figure 4 A schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "include", "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatus.

[0027] Embodiment 1

[0028] Figure 1 This is a flowchart of an agricultural assistance data evaluation method provided in the first embodiment of the present invention. This embodiment is applicable to the situation where agricultural assistance data is evaluated based on multi-dimensional evaluation-related reference data combined with a multivariate linear regression model. The method can be executed by an agricultural assistance data evaluation device, which can be implemented by software and / or hardware, and can generally be integrated in an electronic device. The electronic device can be a terminal device or a server device. As long as it can execute the agricultural assistance data evaluation method, the embodiment of the present invention does not limit the specific device type of the electronic device. Accordingly, Figure 1 As shown, the method includes the following operations:

[0029] S110. Obtain assessment-related reference data for agricultural assistance data in the target assessed agricultural area; wherein the assessment-related reference data includes yield forecast data, agricultural production data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data and climate risk data.

[0030] Among them, the target agricultural area for evaluation can be an agricultural area that needs to evaluate agricultural assistance data, for example, it can be a designated farmland, planting industry, pasture area, dairy animal industry area, etc. As long as it meets the evaluation needs of agricultural assistance data, the embodiment of the present invention does not limit the type and regional scope of the target agricultural area for evaluation.

[0031] Among them, agricultural assistance data can be data designed for agricultural assistance projects provided for target assessment agricultural areas. So-called agricultural assistance refers to a series of projects and measures implemented by relevant institutions to help farmers increase their income, improve their lives, and improve agricultural production levels. These assistance measures cover many aspects and aim to promote the development of the agricultural industry and increase farmers' income. Exemplarily, agricultural assistance measures may include but are not limited to agricultural loans, agricultural subsidies, technical training, and agricultural support. Correspondingly, agricultural assistance data may include but are not limited to agricultural loan amounts or agricultural loan risk data, agricultural subsidy amounts, technical training scope and duration, and types and quantities of materials provided. The embodiments of the present invention do not limit the specific project types of agricultural assistance and the types of agricultural assistance data.

[0032] Among them, the evaluation-related reference data may be the relevant data used for reference when evaluating the agricultural assistance data in the target evaluation agricultural area. The yield prediction data may be the data for predicting the crop yield in the target evaluation agricultural area. Agricultural production data is the most basic and important type of data, and such data may include but are not limited to crop types, planting areas, yields, climate data, and planting management measures. Historical credit data may be the credit data generated historically by the assistance object in the target evaluation agricultural area. It is understandable that the assistance object is generally a farmer. Exemplarily, the historical credit data may include but are not limited to the credit information, credit score, historical loan records (including but not limited to the number of repayments on time, the number of overdue payments, and the number of defaults, etc.) of the assistance object, the total amount of historical loans and the repayment ratio, and the number of overdue days (including the longest overdue days and the average overdue days, etc.). Optionally, the historical credit data may also include the economic data of the assistance object, for example, it may include the proportion of the agricultural income of the assistance object to the total income, other sources of income, and the value of mortgaged assets (such as real estate, land, and machinery, etc.). Agricultural technology cost data may be data on the agricultural technology level and agricultural production costs involved in the target evaluation agricultural area. Among them, agricultural technology level data may include, but are not limited to, different levels of technology such as manual farming, semi-mechanized and mechanized. Agricultural production cost data may include, but are not limited to, various cost-related data such as the cost of agricultural materials such as seeds, fertilizers and pesticides, labor costs and mechanization costs. Soil condition data may be data reflecting the soil and land conditions in the target agricultural area, such as soil fertility and the degree of land pollution and degradation. Crop market price sales data may be sales data related to crops in the target agricultural area, such as including but not limited to the sales price, supply and demand, and sales channels of crops in the target agricultural area, which can be used to analyze the stability of the income of the aid recipient. Climate risk data may be data related to climate and natural disaster risks in the target agricultural area, such as including but not limited to rainfall forecast data for key agricultural periods such as sowing and harvesting periods, and may also include the frequency of extreme weather such as drought, floods, hail and storms, and may also include forecast data on natural disaster risks.

[0033] In an embodiment of the present invention, in order to achieve accurate evaluation of agricultural assistance data, multi-dimensional evaluation-related reference data of agricultural assistance data in the target evaluation agricultural area can be obtained. Among them, the multi-dimensional evaluation-related reference data may include but is not limited to yield forecast data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data and climate risk data, and then from the full dimensions related to agriculture such as yield, credit, technology, cost, means of production, market and climate, the full evaluation-related reference data related to the target evaluation agricultural area is obtained as the basic data source for evaluating the agricultural assistance data required for the target evaluation agricultural area.

[0034] S120, inputting the evaluation-related reference data into a multivariate linear regression model, and evaluating the agricultural assistance data of the target evaluation agricultural area through the multivariate linear regression model according to the evaluation-related reference data and the weights of the evaluation-related reference data.

[0035] The multivariate linear regression model may be a model that uses multi-dimensional evaluation-related reference data to evaluate agricultural assistance data of a target evaluation agricultural region. Optionally, the multivariate linear regression model may be a linear regression model.

[0036] Optionally, in an embodiment of the present invention, a multivariate linear regression model can be used to evaluate the agricultural assistance data of the target evaluated agricultural area. Specifically, the multivariate linear regression model can use multi-dimensional evaluation-related reference data as variables, and configure corresponding weights for each variable. After obtaining the evaluation-related reference data of the target evaluated agricultural area, the obtained evaluation-related reference data of the target evaluated agricultural area can be input into the multivariate linear regression model as input data. The multivariate linear regression model can calculate the agricultural assistance data of the target evaluated agricultural area by weighting the weights configured for each variable according to the input data and the model.

[0037] For example, when it is necessary to evaluate the agricultural loans needed for the target assessment agricultural area, the yield forecast data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data and climate risk data of the target assessment agricultural area can be used as input and input into a multivariate linear regression model to evaluate the loan amount of the target assessment agricultural area based on the above-mentioned assessment-related reference data and the weights of each assessment-related reference data through the multivariate linear regression model.

[0038] For example, when it is necessary to evaluate the cost of agricultural inputs required for a target agricultural area, the yield forecast data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data and climate risk data of the target agricultural area can be used as inputs and input into a multivariate linear regression model to evaluate the cost of agricultural inputs for the target agricultural area based on the above-mentioned evaluation-related reference data and the weights of each evaluation-related reference data through the multivariate linear regression model.

[0039] Scientific and comprehensive agricultural assistance data evaluation is inseparable from the collection and analysis of multi-dimensional data. The above technical solution evaluates the agricultural assistance data of the target agricultural area from multiple perspectives such as agricultural production, economy and social policies as variables, which can improve the efficiency, accuracy, scientificity and rationality of agricultural assistance data.

[0040] The embodiment of the present invention obtains evaluation-related reference data such as yield forecast data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data and climate risk data of agricultural assistance data in the target evaluated agricultural area, so as to input the multi-dimensional evaluation-related reference data into a multivariate linear regression model, thereby evaluating the agricultural assistance data of the target evaluated agricultural area according to the multi-dimensional evaluation-related reference data and the weights of the evaluation-related reference data through the multivariate linear regression model, thereby solving the problems of low efficiency and accuracy of existing agricultural assistance data evaluation, and being able to improve the efficiency, accuracy, scientificity and rationality of agricultural assistance data evaluation.

[0041] Embodiment 2

[0042] Figure 2 is a flow chart of an agricultural assistance data evaluation method provided by Example 2 of the present invention. This embodiment is specific based on the above embodiment. In this embodiment, multiple specific optional implementation methods for obtaining yield prediction data, evaluating agricultural assistance data through a multivariate linear regression model, and optimizing and adjusting agricultural assistance data are provided. Figure 2 As shown, the method of this embodiment may include:

[0043] S210: Acquire UAV remote sensing data of the target agricultural area for assessment.

[0044] Among them, the drone remote sensing data can be remote sensing data collected by a drone.

[0045] S220. Input the UAV remote sensing data into a target network model that combines convolution and attention mechanisms, so as to predict the crop yield of the target assessed agricultural area according to the UAV remote sensing data through the target network model to obtain the yield prediction data.

[0046] Among them, the convolutional neural network imitates human neurons and is composed of many neurons with weights and activation functions. As a computational model for machines to imitate biological structures and functions, it has broad application prospects in fields such as finance and medicine. The core idea of ​​the attention mechanism is to highlight certain important features of the object. In fact, it is an allocation mechanism that allocates weights of different neurons according to their importance. It is used to capture key features in the field of image processing and to locate key sequences in the field of natural language processing. The target network model can be a network model that combines the advantages of convolution operations and attention mechanisms. The target network model uses the two key ideas of relative attention and vertical stacking to make the model's feature expression ability better and the prediction effect of graphs better.

[0047] With the development of science and technology, prediction methods such as remote sensing technology and machine learning have also been used in the field of agricultural yield estimation. The yield estimation algorithms currently used are mostly machine learning algorithms based on satellite remote sensing. Although machine learning algorithms based on satellite remote sensing can extract wide-area data, their resolution is low and it is difficult to accurately identify specific crops and growth conditions, which has a great impact on the accuracy of yield prediction data. For the processing of remote sensing images, regression analysis, support vector machines, random forests and deep neural networks are currently used. Although these models can provide accurate predictions for agricultural assistance data based on image data to a certain extent, they also have strong data dependence and difficulty in identifying complex spatial and nonlinear features in farmland images. Especially in the case of changing terrain, overlapping crops or dense planting, the model's performance is unstable and the computational efficiency is low, making it difficult to meet real-time prediction needs.

[0048] In order to solve the above problems, the embodiment of the present invention collects and obtains drone remote sensing data from the target evaluated agricultural area through drone equipment as a data source for yield prediction of the target evaluated agricultural area. Drone remote sensing data can quickly cover a large range of farmland and provide high-resolution real-time images, significantly improving the accuracy and efficiency of data acquisition. The high spatial resolution of drone remote sensing data enables it to accurately capture farmland details, such as crop types, density and health conditions, and provide a more accurate basis for yield estimation. Compared with satellite remote sensing data, drone remote sensing data has more advantages in the application of complex terrain and densely planted areas. Based on the characteristics of different electromagnetic waves reflected by different objects, drone remote sensing image data is effectively processed to obtain crop growth and other information. By combining some agronomy and meteorology related parameters, drone remote sensing data is effectively processed, and crop growth and yield information can be efficiently estimated and obtained on a large scale. Therefore, using drone remote sensing data to predict yield can provide efficient and accurate data support, thereby significantly improving the accuracy of yield prediction.

[0049] At the same time, the embodiment of the present invention introduces a selective attention mechanism into a convolutional neural network to obtain a target network model that combines convolution and attention mechanisms. After obtaining the drone remote sensing data of the target agricultural area for evaluation, the drone remote sensing data can be input into the target network model that combines convolution and attention mechanisms, and the target network model predicts the crop yield of the target agricultural area for evaluation based on the drone remote sensing data to obtain yield prediction data. Specifically, the target network model predicts the importance of each part of the drone remote sensing data to the optimization target, obtains the corresponding attention weight, and then enhances the features with large weights and weakens the data features with small weights according to the size of the attention weights. Since the target network model combines the dual advantages of convolutional neural networks and self-attention mechanisms, by using the convolution layer to extract local features of the image and using the self-attention mechanism to capture the long-distance dependencies between features, a better trade-off between accuracy and efficiency is achieved, thereby improving the simulation effect of the neural network.

[0050] The above technical solution adopts a target network model that combines convolution and attention mechanisms. Through its excellent image feature extraction capabilities, high accuracy, computational efficiency and advantages in modeling complex nonlinear relationships, it has shown stronger advantages than traditional models in agricultural yield estimation, especially when processing drone remote sensing images and crop classification, it can provide more accurate and efficient solutions.

[0051] It is understandable that the target network model first needs to undergo a model training process before being put into use to ensure that the target network model is trained to sufficient accuracy. Therefore, after obtaining the original UAV remote sensing data, the original UAV remote sensing data can be preliminarily processed first, and preprocessing work such as radiation correction, geometric correction, atmospheric correction and noise removal can be performed to ensure the accuracy and reliability of the UAV remote sensing data. Furthermore, the UAV remote sensing data can also be subjected to image cropping, image scaling and normalization processing to ensure the consistency of the data input into the target network model. At the same time, the processed UAV remote sensing data is divided into a training set, a test set and a validation set. During training, the processed UAV remote sensing data is used as the input sample data of the target network model in the form of a control group.

[0052] The core network structure of the target network model includes multiple convolutional layers, attention mechanism modules and pooling layers to enhance the image feature extraction capability. At the same time, the target network model combines convolution operations and attention mechanisms to capture subtle features and long-distance dependencies in drone remote sensing data, thereby achieving accurate image classification and feature extraction.

[0053] S230. Obtain historical credit data, agricultural technology cost data, soil condition data, crop market price sales data, climate risk data and other assessment-related reference data for the target agricultural area.

[0054] S240, calculating the product value between each evaluation-related reference data and the corresponding weight through the multivariate linear regression model.

[0055] S250, calculating the sum of each product value and the bias term to obtain agricultural assistance data for the target agricultural area.

[0056] The bias term can be a constant term in a multivariate linear regression model.

[0057] Considering that in practical applications, the evaluation of agricultural assistance data is not only directly affected by the input characteristics, but may also be affected by implicit systemic factors, such as policy support and other factors that cannot be effectively quantified. These factors can be solved by setting bias terms. Therefore, in addition to setting corresponding weights for the variables of each evaluation-related reference data, a bias term can also be configured in the multivariate linear regression model to replace the impact of factors that cannot be effectively quantified on the evaluation results of agricultural assistance data through the bias term, thereby further improving the accuracy of the multivariate linear regression model.

[0058] Specifically, after the evaluation-related reference data is input into the multivariate linear regression model, the multivariate linear regression model can calculate the product value between each evaluation-related reference data and the corresponding weight. Further, the sum of each product value and the bias term is calculated to reduce the deviation of the evaluation result of the multivariate linear regression model by introducing the bias term, thereby obtaining accurate agricultural assistance data for the target evaluation agricultural area.

[0059] In an optional embodiment of the present invention, before obtaining the evaluation-related reference data of the agricultural assistance data in the target evaluation agricultural area, it may also include: obtaining evaluation-related reference sample data; optimizing and updating the initial weights and initial bias items in the multivariate linear regression model using a regularized regression method and a gradient descent method based on the evaluation-related reference sample data.

[0060] The evaluation-associated reference sample data may be used to optimize and update the initial weights and initial bias items in the multivariable linear regression model, and the sample data is also obtained based on the evaluation-associated reference data. The initial weights may be the weights initially set in the multivariable linear regression model. The initial bias item may be the bias item initially set in the multivariable linear regression model.

[0061] It is understandable that the multivariate linear regression model initially created based on the initial weights and initial bias items may have a low evaluation accuracy for agricultural assistance data, so it is necessary to optimize and update the initial weights and initial bias items in the multivariate linear regression model. Exemplarily, the formula of the multivariate linear regression model can be: R pred =w1·X1+w2·X2+...+w n ·X n +b. Among them, R pred is the evaluation value of agricultural assistance data, w n is the weight value, X n is one of the evaluation related reference data, and b is the bias term. It can be seen that the multivariate linear regression model establishes a linear relationship between various input features and agricultural assistance data. Among them, the input feature X n That is, the features such as yield forecast data, agricultural production data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data and climate risk data. In order to obtain the optimal weight (i.e. coefficient) to fit the relationship between the input features and the agricultural assistance data, the evaluation-related reference sample data can be used as the input features, such as (X1, X2, X3...), etc., and the agricultural assistance data can be used as the target evaluation value to train and optimize the initial weights and initial bias terms in the multivariable linear regression model. Since the regularization term can avoid overfitting and constrain the weight value to a certain extent, it is more suitable for optimizing the weights and bias terms in the training multivariable linear regression model. Therefore, the regularized regression model can be used to train the weights and bias terms in the multivariable linear regression model. Specifically, the regularized regression method and the gradient descent method can be used to calculate the gradient of the loss function of the multivariable linear regression model, so as to gradually update the weights of the multivariable linear regression model. The update formula is as follows.

[0062]

[0063] Where η is the learning rate, L(w) is the loss function, is the gradient of the loss function with respect to the weight w. During the weight value update, the bias term is also updated synchronously. After the weight and bias term training optimization of the multivariate linear regression model is completed, it can be tested on the validation set generated based on the evaluation association reference sample data.

[0064] The regularized regression method is a technique used to prevent overfitting. It limits the model by adding a regularization term in the loss function. It has the advantages of preventing overfitting and feature selection. It combines the gradient descent method to train and optimize the multivariable linear regression model, which can improve the predictive ability and generalization ability of the multivariable linear regression model.

[0065] In an optional embodiment of the present invention, the target assessment agricultural area includes a plurality of different types of crops. After obtaining the UAV remote sensing data of the target assessment agricultural area, it may also include: identifying the plurality of different types of crops included in the target assessment agricultural area according to the UAV remote sensing data; segmenting the same type of crop areas in the target assessment agricultural area to obtain a plurality of local crop area remote sensing data; inputting the UAV remote sensing data into a target network model combining convolution and attention mechanisms, so as to predict the crop yield of the target assessment agricultural area according to the UAV remote sensing data through the target network model to obtain the yield prediction data, which may include: inputting each of the local crop area remote sensing data into the target network model respectively, so as to predict the yield of a plurality of crops in the target assessment agricultural area according to the local crop area remote sensing data through the target network model to obtain the yield prediction data.

[0066] The local crop area remote sensing data may be remote sensing data of a crop area including only one type of crop.

[0067] It is understandable that a target assessment agricultural area can usually include a variety of different types of crops. For example, a target assessment agricultural area can include different types of crops such as wheat, corn and rice, and different types of crops usually have different support policies. Therefore, when the target assessment agricultural area includes a variety of different types of crops, it is necessary to use the crop type as a benchmark and predict the corresponding yield prediction data according to different types of crops. Specifically, the various types of crops included in the target assessment agricultural area can be first identified based on the drone remote sensing data, and the same type of crop areas in the target assessment agricultural area can be segmented to obtain multiple local crop area remote sensing data. Exemplarily, all wheat areas in the target assessment agricultural area can be segmented into a local crop area remote sensing data, and all corn areas in the target assessment agricultural area can be segmented into a local crop area remote sensing data, etc. Correspondingly, when predicting crop yields, the remote sensing data of each local crop area can be input into the target network model separately based on the crop. The target network model predicts the yields of various crops in the target assessed agricultural area based on the remote sensing data of local crop areas. Each crop can be predicted to obtain a corresponding yield prediction data.

[0068] Correspondingly, after the yield prediction data of different crops in the target agricultural area are predicted, the corresponding agricultural production data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data, climate risk data and other dimensional evaluation-related reference data can be obtained for different types of crops, taking crops as units. Furthermore, for each type of crop, the corresponding yield prediction data, agricultural production data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data, climate risk data and other multi-dimensional evaluation-related reference data can be input into the multivariate linear regression model, so as to evaluate the corresponding agricultural assistance data for each crop in the target agricultural area through the multivariate linear regression model.

[0069] The above technical scheme predicts crop yields based on crop types, and then evaluates the agricultural assistance data of each crop based on the prediction results combined with evaluation-related reference data in other dimensions, which can further improve the accuracy, rationality and scientificity of the agricultural assistance data evaluation.

[0070] In an optional embodiment of the present invention, after evaluating the agricultural assistance data of the target evaluated agricultural area, it may also include: obtaining historical agricultural assistance data and geographic location information of the target evaluated agricultural area; generating an agricultural assistance data association map based on at least one of the historical agricultural assistance data, the geographic location information and the evaluated agricultural assistance data; and optimizing and adjusting the agricultural assistance data based on the agricultural assistance data association map.

[0071] The historical agricultural assistance data may be agricultural assistance data generated by the historical assessment of the target agricultural area. The agricultural assistance data association graph may be a graph associated with the agricultural assistance data. Exemplarily, the types of agricultural assistance data association graphs may include but are not limited to scatter plots, line graphs, histograms, heat maps, and other available graph types, and the embodiments of the present invention do not limit the graph types of the agricultural assistance data association graphs.

[0072] In order to fully play the role of agricultural assistance data in the target agricultural area, after obtaining the agricultural assistance data in the target agricultural area, the historical agricultural assistance data and geographical location information such as the corresponding geographical coordinates of the target agricultural area can be further obtained to analyze and organize the current evaluation data and all the related data obtained, generate a variety of required agricultural assistance data association diagrams, and realize the visualization of agricultural assistance data. Correspondingly, after obtaining the corresponding type of agricultural assistance data association diagram, the agricultural assistance data generated by the evaluation can be further optimized and adjusted according to the agricultural assistance data association diagram.

[0073] For example, by taking the obtained reference data associated with each assessment as feature values, and combining the historically generated and currently assessed agricultural assistance data, a scatter plot can be drawn to show the relationship between the reference data associated with each assessment and the agricultural assistance data. For example, agricultural loans with low credit history scores also have poor risk assessment values.

[0074] For example, with the support of more timely data, a time series line graph can be drawn for agricultural assistance data in different time periods to show data fluctuations in different seasons or years. Institutions that provide agricultural assistance policies can identify periods that need to be focused on in advance based on historical data and forecast results to adjust assistance strategies accordingly. For example, taking agricultural credit risk value as an example of agricultural assistance data, it is possible to display credit risk fluctuations in different seasons or years. Credit institutions can identify high-risk periods in advance based on historical data and forecast results to adjust loan strategies. For another example, taking technical training as an example of agricultural assistance data, it is possible to display special skills training data in different seasons or years. Skill training institutions can determine skill training arrangements for assistance targets in target assessment agricultural areas based on visualized data.

[0075] For example, farmers can be used as a benchmark to divide the agricultural assistance data of each farmer into intervals and display the number of farmers or farms in each interval in the form of a histogram, and then the different intervals can be distinguished according to the evaluation results to provide differentiated assistance policies for farmers in different intervals. For example, taking the agricultural credit risk value as an example of agricultural assistance data, the credit risk assessment values ​​of multiple farmers can be divided into several intervals, and the number of farmers or farms in each interval can be displayed in the form of a histogram. Further, the credit risk value is divided into three categories: low risk, medium risk and high risk, and different colors are used to distinguish them. It is convenient for credit institutions to understand the credit risk range of most farmers. If it is shown that the credit risk of most farmers is low and the credit risk of a few farmers is high, differentiated loan strategies can be adopted. For example, the constraints on loans can be appropriately relaxed for farmers with low credit risks, and the constraints on loans can be strengthened for farmers with high credit risks.

[0076] Exemplarily, geographic information can also be combined with agricultural assistance data, and a heat map can be generated using geographic coordinates to display the distribution data of the target agricultural area based on the geographic location. Taking the agricultural credit risk value as an example of agricultural assistance data, geographic information can also be combined with the credit risk assessment value. Use geographic coordinates (such as the location of a farm or region) combined with the credit risk assessment value to generate a heat map showing high-risk areas and low-risk areas. Decision makers can adjust credit policies based on the risk distribution of geographic locations. For example, farmers in certain areas may be at higher risk due to climate factors, poor land, etc., and credit institutions can formulate specific loan conditions.

[0077] The above technical solution optimizes and adjusts the agricultural assistance data according to the agricultural assistance data association diagram to realize the visualization of agricultural assistance data, which can not only further improve the accuracy of agricultural assistance data evaluation, but also help decision makers quickly understand the output of the multivariate linear regression model, and provide data-driven decision support for assistance agencies. By deeply analyzing the impact of different factors on agricultural assistance data, it is also appropriate to formulate or adjust corresponding assistance policies.

[0078] Agricultural assistance data evaluation is a typical regression problem based on large amounts of data, which is of great significance for farmers to obtain additional income and agricultural security. At the same time, agricultural assistance data evaluation can also better promote the implementation of agricultural assistance activities, help optimize the rural financial and technological environment, help farmers solve problems in agricultural production and rural development, and promote the healthy development of agricultural production.

[0079] In the technical solution of the present disclosure, the collected information involved is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0080] It should be noted that any arrangement and combination of the technical features in the above embodiments also falls within the protection scope of the present invention.

[0081] Embodiment 3

[0082] Figure 3 is a schematic diagram of an agricultural assistance data evaluation device provided by Embodiment 3 of the present invention, such as Figure 3 As shown, the device includes: an evaluation-related reference data acquisition module 310 and an agricultural assistance data evaluation module 320, wherein:

[0083] An evaluation-related reference data acquisition module 310 is used to acquire evaluation-related reference data of agricultural assistance data in a target evaluation agricultural area; wherein the evaluation-related reference data includes yield forecast data, historical credit data, and climate risk data;

[0084] The agricultural assistance data evaluation module 320 is used to input the evaluation-related reference data into a multivariate linear regression model, and evaluate the agricultural assistance data of the target evaluation agricultural area through the multivariate linear regression model based on the evaluation-related reference data and the weights of the evaluation-related reference data.

[0085] The embodiment of the present invention obtains evaluation-related reference data such as yield forecast data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data and climate risk data of agricultural assistance data in the target evaluated agricultural area, so as to input the multi-dimensional evaluation-related reference data into a multivariate linear regression model, thereby evaluating the agricultural assistance data of the target evaluated agricultural area according to the multi-dimensional evaluation-related reference data and the weights of the evaluation-related reference data through the multivariate linear regression model, thereby solving the problems of low efficiency and accuracy of existing agricultural assistance data evaluation, and being able to improve the efficiency, accuracy, scientificity and rationality of agricultural assistance data evaluation.

[0086] Optionally, if the evaluation-associated reference data includes yield prediction data, the evaluation-associated reference data acquisition module 310 is also used to: obtain UAV remote sensing data of the target evaluation agricultural area; input the UAV remote sensing data into a target network model that combines convolution and attention mechanisms, so as to predict the crop yield of the target evaluation agricultural area based on the UAV remote sensing data through the target network model to obtain the yield prediction data.

[0087] Optionally, the agricultural assistance data evaluation module 320 is also used to: calculate the product value between each evaluation-related reference data and the corresponding weight through the multivariate linear regression model; calculate the sum of each product value and the bias term to obtain the agricultural assistance data of the target evaluation agricultural area.

[0088] Optionally, the above-mentioned device also includes a model training module, which is used to: obtain evaluation-related reference sample data; and optimize and update the initial weights and initial bias terms in the multivariate linear regression model using a regularized regression method and a gradient descent method based on the evaluation-related reference sample data.

[0089] Optionally, the target assessment agricultural area includes multiple different types of crops, and the assessment associated reference data acquisition module 310 is also used to: identify the multiple different types of crops included in the target assessment agricultural area based on the UAV remote sensing data; segment the same type of crop areas in the target assessment agricultural area to obtain multiple local crop area remote sensing data; input each of the local crop area remote sensing data into the target network model, so as to predict the yield of multiple crops in the target assessment agricultural area based on the local crop area remote sensing data through the target network model to obtain the yield prediction data.

[0090] Optionally, the above-mentioned device also includes a data optimization and adjustment module, which is used to: obtain historical agricultural assistance data and geographical location information of the target evaluated agricultural area; generate an agricultural assistance data association map based on at least one of the historical agricultural assistance data, the geographical location information and the evaluated agricultural assistance data; and optimize and adjust the agricultural assistance data according to the agricultural assistance data association map.

[0091] The above agricultural assistance data evaluation device can execute the agricultural assistance data evaluation method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method. For technical details not described in detail in this embodiment, please refer to the agricultural assistance data evaluation method provided by any embodiment of the present invention.

[0092] Since the agricultural assistance data evaluation device introduced above is a device that can execute the agricultural assistance data evaluation method in the embodiment of the present invention, based on the agricultural assistance data evaluation method introduced in the embodiment of the present invention, the technical personnel of the field can understand the specific implementation of the agricultural assistance data evaluation device of the present embodiment and its various variations, so how the agricultural assistance data evaluation device implements the agricultural assistance data evaluation method in the embodiment of the present invention is not described in detail here. As long as the technical personnel of the field implement the device used in the agricultural assistance data evaluation method in the embodiment of the present invention, it belongs to the scope of protection of this application.

[0093] Embodiment 4

[0094] Figure 4A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0095] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0096] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0097] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the agricultural assistance data evaluation method.

[0098] Optionally, agricultural assistance data assessment methods may include:

[0099] Acquire assessment-related reference data of agricultural assistance data in the target assessment agricultural area; wherein the assessment-related reference data includes yield forecast data, agricultural production data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data and climate risk data;

[0100] The evaluation-related reference data are input into a multivariate linear regression model, and the agricultural assistance data of the target evaluation agricultural area is evaluated through the multivariate linear regression model based on the evaluation-related reference data and the weights of the evaluation-related reference data.

[0101] In some embodiments, the agricultural assistance data evaluation method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the agricultural assistance data evaluation method described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to execute the agricultural assistance data evaluation method in any other appropriate manner (for example, by means of firmware).

[0102] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0103] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0104] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0105] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0106] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0107] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0108] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0109] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for evaluating agricultural assistance data, characterized in that: include: Acquire assessment-related reference data of agricultural assistance data in the target assessment agricultural area; wherein the assessment-related reference data includes yield forecast data, agricultural production data, historical credit data, agricultural technology cost data, soil condition data, crop market price sales data and climate risk data; The evaluation-related reference data are input into a multivariate linear regression model, and the agricultural assistance data of the target evaluation agricultural area is evaluated through the multivariate linear regression model based on the evaluation-related reference data and the weights of the evaluation-related reference data.

2. The method according to claim 1, characterized in that If the evaluation-related reference data includes yield prediction data, the evaluation-related reference data for obtaining agricultural assistance data in the target evaluation agricultural area includes: Acquiring UAV remote sensing data of agricultural areas for said objective assessment; The UAV remote sensing data is input into a target network model that combines convolution and attention mechanisms, so as to predict the crop yield of the target assessed agricultural area according to the UAV remote sensing data through the target network model to obtain the yield prediction data.

3. The method according to claim 1, characterized in that The step of evaluating the agricultural assistance data of the target evaluated agricultural area according to the evaluation-related reference data and the weight of the evaluation-related reference data by the multivariate linear regression model includes: Calculate the product value between each of the evaluation-related reference data and the corresponding weight by using the multivariate linear regression model; The sum of each product value and the bias term is calculated to obtain the agricultural assistance data of the target assessment agricultural area.

4. The method according to claim 1, characterized in that: Before obtaining the evaluation-related reference data of the agricultural assistance data in the target evaluation agricultural area, the method further includes: Obtaining evaluation-related reference sample data; The initial weights and initial bias items in the multivariate linear regression model are optimized and updated by using a regularized regression method and a gradient descent method according to the evaluation-associated reference sample data.

5. The method according to claim 2, characterized in that: The target agricultural area for evaluation includes a plurality of different types of crops. After acquiring the drone remote sensing data of the target agricultural area for evaluation, the method further includes: Identifying a plurality of different types of crops included in the target assessment agricultural area based on the drone remote sensing data; Segmenting the same type of crop areas in the target agricultural area to obtain remote sensing data of multiple local crop areas; The step of inputting the UAV remote sensing data into a target network model combining convolution and attention mechanisms, so as to predict the crop yield of the target assessed agricultural area according to the UAV remote sensing data through the target network model to obtain the yield prediction data, comprises: The remote sensing data of each local crop area are respectively input into the target network model, so as to predict the yield of multiple crops in the target assessment agricultural area according to the remote sensing data of each local crop area through the target network model to obtain the yield prediction data.

6. The method according to claim 1, characterized in that After evaluating the agricultural assistance data of the target agricultural area, it also includes: Obtain historical agricultural assistance data and geographic location information for the target assessment agricultural area; generating an agricultural assistance data association graph according to at least one of the historical agricultural assistance data, the geographical location information, and the evaluated agricultural assistance data; The agricultural assistance data is optimized and adjusted according to the agricultural assistance data association diagram.

7. An agricultural assistance data evaluation device, characterized in that: include: An evaluation-related reference data acquisition module is used to acquire evaluation-related reference data of agricultural assistance data in a target evaluation agricultural area; wherein the evaluation-related reference data includes yield forecast data, historical credit data and climate risk data; The agricultural assistance data evaluation module is used to input the evaluation-related reference data into a multivariate linear regression model, and evaluate the agricultural assistance data of the target evaluation agricultural area through the multivariate linear regression model based on the evaluation-related reference data and the weights of the evaluation-related reference data.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the agricultural assistance data evaluation method described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the agricultural assistance data evaluation method described in any one of claims 1-6 when executed.

10. A computer program product comprising a computer program / instructions, wherein: When the computer program / instructions are executed by a processor, the agricultural assistance data evaluation method described in any one of claims 1-6 is implemented.

Images