Crop underwriting method, device and equipment and storage medium

Abstract

The embodiment of the invention provides a crop underwriting method and device, equipment and a storage medium. The method comprises the steps that a growth image of a crop and target planting record information of the crop are acquired through a sensor, target remote sensing data of the crop are acquired through a remote sensing technology, and the target remote sensing data comprise a multispectral image; the multispectral image of the crop is analyzed to obtain growth environment information of the crop, and the growth environment information comprises soil humidity and vegetation indexes; analyzing the growth image of the crop, the target planting record information of the crop and the growth environment information of the crop through a target deep learning model to obtain target risk level information of the plot where the crop is located; and matching corresponding insurance terms for the crops according to the target risk level information. The embodiment of the invention aims to provide the underwriting method for the crops, so as to optimize the underwriting process of the crops, realize personalized risk assessment and improve insurance guarantee.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a crop insurance method, a crop insurance device, a computer device, and a computer-readable storage medium. Background Art

[0002] Agricultural production is often affected by factors such as climate and natural disasters, which lead to frequent agricultural disasters. As an important means of dispersing agricultural production and management risks, agricultural insurance plays a very important role in promoting high-quality development of agricultural production, protecting farmers' income, consolidating the basic position of agriculture and rural revitalization.

[0003] As the scale of agricultural insurance business continues to expand, the phenomenon of insured areas and crop types not matching the actual planting conditions has become increasingly prominent. The traditional on-site inspection method not only consumes a lot of manpower and material resources, but also fails to obtain farmland information in a timely, large-scale and accurate manner, resulting in slow development of agricultural insurance institutions in the direction of accurate underwriting and precise claims. Summary of the invention

[0004] The present application provides a crop insurance method, a crop insurance device, a computer device and a computer-readable storage medium, which optimize the crop insurance process, realize personalized risk assessment and improve insurance protection.

[0005] To achieve the above objectives, the present application provides a crop insurance method, the method comprising:

[0006] Acquire crop growth images and target planting record information of the crop through sensors, and acquire target remote sensing data of the crop through remote sensing technology, wherein the target remote sensing data includes multispectral images;

[0007] Analyzing the multispectral image of the crop to obtain growth environment information of the crop, wherein the growth environment information includes soil moisture and vegetation index;

[0008] Analyzing the growth image of the crop, the target planting record information of the crop, and the growth environment information of the crop through a target deep learning model to obtain target risk level information of the plot where the crop is located;

[0009] The corresponding insurance terms are matched for the crops according to the target risk level information.

[0010] To achieve the above purpose, the present application also provides an insurance device for claiming crops, comprising:

[0011] An acquisition module, used to acquire crop growth images and target planting record information of the crop through sensors, and to acquire target remote sensing data of the crop through remote sensing technology, wherein the target remote sensing data includes multispectral images;

[0012] An analysis module, used for analyzing the multispectral image of the crop to obtain the growth environment information of the crop, wherein the growth environment information includes soil moisture and vegetation index;

[0013] The analysis module is further used to analyze the growth image of the crop, the target planting record information of the crop, and the growth environment information of the crop through a target deep learning model to obtain target risk level information of the plot where the crop is located;

[0014] A matching module is used to match corresponding insurance clauses for the crop according to the target risk level information.

[0015] In addition, to achieve the above-mentioned purpose, the present application also provides a computer device, which includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and implement the steps of any one of the crop insurance methods provided in the embodiments of the present application when executing the computer program.

[0016] In addition, to achieve the above-mentioned purpose, the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor implements the steps of the crop insurance method described in any one of the embodiments of the present application.

[0017] The embodiments of the present application disclose a crop insurance method, a crop insurance device, a computer device, and a computer-readable storage medium. Specifically, the growth image of the crop and the target planting record information of the crop are obtained by a sensor, and the target remote sensing data of the crop are obtained by remote sensing technology, wherein the target remote sensing data includes a multispectral image. Further, the multispectral image of the crop can be analyzed to obtain the growth environment information of the crop, wherein the growth environment information includes soil moisture and vegetation index. Thus, the growth image of the crop, the target planting record information of the crop, and the growth environment information of the crop can be analyzed by a target deep learning model to obtain the target risk level information of the plot where the crop is located. Finally, the corresponding insurance terms can be matched to the crop according to the target risk level information. The present application aims to obtain the growth image and target planting record information of the crop by a sensor, and combine the multispectral image data collected by remote sensing technology to fully grasp the growth status and environmental information of the crop, so as to make the assessment of the health status and environmental risks of the crop more comprehensive and accurate, thereby improving the accuracy of the insurance decision. Furthermore, by analyzing multispectral image data (such as soil moisture, vegetation index and other growth environment information), combined with the historical planting records of crops and the target deep learning model, the target risk level of the crop plot can be automatically generated, avoiding the subjectivity of traditional manual judgment and improving the efficiency and objectivity of underwriting. In addition, according to the risk level of the plot where the crop is located, the corresponding insurance terms can be automatically matched to the crop to ensure that crops with different risk levels receive appropriate protection. This personalized underwriting plan can better meet the needs of users and reduce the insurance company's compensation risk. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 This is a schematic diagram of a scenario of a crop insurance method provided in an embodiment of the present application;

[0020] Figure 2 It is a schematic diagram of the steps of a crop insurance method provided in an embodiment of the present application;

[0021] Figure 3 It is a schematic diagram of a process for obtaining a target deep learning model provided in an embodiment of the present application;

[0022] Figure 4 is a schematic block diagram of a crop insurance device provided in an embodiment of the present application;

[0023] Figure 5 It is a schematic block diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0025] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor do they have to be executed in the order described. For example, some operations / steps can also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions. In addition, although the functional modules are divided in the device schematic, in some cases, the module division can be different from that in the device schematic.

[0026] The term "and / or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0027] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0028] like Figure 1 As shown, the crop insurance method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. The application environment includes a terminal device 110 and a server 120, wherein the terminal device 110 can obtain the growth image of the crop, the target planting record information of the crop through a sensor, and obtain the target remote sensing data of the crop through remote sensing technology, wherein the target remote sensing data includes a multispectral image; analyze the multispectral image of the crop to obtain the growth environment information of the crop, wherein the growth environment information includes soil moisture and vegetation index; analyze the growth image of the crop, the target planting record information of the crop and the growth environment information of the crop through a target deep learning model to obtain the target risk level information of the plot where the crop is located; match the corresponding insurance clause for the crop according to the target risk level information, and send the insurance clause to the server 120. Among them, the terminal device 110 and the server 120 can be independent servers, or cloud servers that provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms, and this application is not limited here.

[0029] See also Figure 2 , Figure 2 Schematic diagram of the steps of a crop insurance method provided in an embodiment of the present application. Figure 2 As shown, the crop insurance method includes steps S11 to S14.

[0030] Step S11: Acquire crop growth images and target crop planting record information through sensors, and acquire target remote sensing data of crops through remote sensing technology.

[0031] Among them, the target remote sensing data includes multispectral images.

[0032] For example, a camera or other image acquisition equipment (such as a drone, ground monitoring equipment, etc.) can be used to capture real-time growth images of crops. Through the growth images, the growth status, pest and disease conditions of the crops can be intuitively determined, so as to better monitor and manage the health of the crops. Among them, crops include but are not limited to soybeans, corn, etc., and this application does not limit this.

[0033] For example, the target planting record information includes basic planting data of crops, such as crop varieties, planting time, fertilization records, irrigation records, pest and disease control measures, etc. The target planting record information can be collected through IoT sensors, agricultural management systems or manual records, and can provide data support for analyzing crop growth conditions and future predictions.

[0034] It should be noted that remote sensing technology can use satellites, drones or high-altitude equipment combined with optical sensors to collect remote sensing data of large-scale farmland. These data can reflect environmental information such as crop growth status, surface temperature, soil moisture, etc.

[0035] Furthermore, multispectral images are a type of remote sensing data that captures farmland information through light of different wavelengths (such as visible light, near infrared light, short-wave infrared light, etc.). Light of different wavelengths can reveal different responses of crops under specific conditions, such as crop health, moisture content, signs of pests and diseases, etc. By analyzing this spectral information, the growth of crops can be evaluated more accurately, thus providing strong technical support for precision agriculture.

[0036] In the embodiments of the present application, multi-source information (images, records, remote sensing data) can be acquired and analyzed to fully understand the growth of crops and provide a data basis for subsequent decision-making and management.

[0037] Step S12: Analyze the multispectral image of the crop to obtain the growth environment information of the crop.

[0038] Among them, the growth environment information includes soil moisture and vegetation index.

[0039] Furthermore, growth environment information refers to key indicators in the environment surrounding crops, which have a direct impact on the healthy growth of crops. By analyzing multispectral images, growth environment information such as soil moisture, vegetation coverage, and light intensity can be derived.

[0040] It should be understood that multispectral images contain reflectance data of crops in different spectral bands. By analyzing these data, it is possible to identify the characteristic responses of crops in different bands and extract key parameters for measuring crop growth conditions. Commonly used methods include spectral reflectance analysis, image segmentation, and index calculation.

[0041] In the embodiment of the present application, by analyzing multispectral images, the soil moisture, vegetation index and other growth environment information of crops are accurately obtained to provide data support for subsequent decision-making.

[0042] Step S13: Analyze the crop growth image, the target planting record information of the crop, and the growth environment information of the crop through the target deep learning model to obtain the target risk level information of the plot where the crop is located.

[0043] Among them, the deep learning model is a machine learning method that simulates the neural network structure of the human brain, and is particularly good at processing complex information such as images and time series data. In agricultural applications, the deep learning model can be trained with a large amount of historical data to learn the impact of different factors on crop health and yield. The embodiment of the present application can use the target deep learning model to integrate and analyze data from multiple sources to predict the risk level of crops.

[0044] For example, images of crop growth can be analyzed by a convolutional neural network in the target deep learning model. The convolutional neural network can automatically identify subtle differences in crop growth, such as leaf color changes, early signs of pests and diseases, and abnormalities in stem and leaf structure, to help the model assess the current health of the crop.

[0045] For example, the target deep learning model analyzes crop growth images, crop target planting record information, and crop growth environment information for multi-level analysis to find potential risk factors related to crop growth, such as weather fluctuations, soil moisture imbalance, latent period of pests and diseases, improper fertilization, etc. The target deep learning model can automatically identify and calculate the comprehensive impact of these factors on crops through complex nonlinear functions.

[0046] In other embodiments, the target deep learning model can not only analyze each data source separately, but also associate images, records, and environmental information through data fusion technology (such as multimodal learning). For example, the target deep learning model can combine yellowing leaves (image features) and continuous high temperature and low humidity (environmental information) to identify possible water stress risks to improve the accuracy of risk prediction.

[0047] For example, through comprehensive analysis of all data, the target deep learning model can output a target risk level to indicate the potential risk faced by the plot where the crop is located. The risk level can be divided into high, medium, and low levels, corresponding to different management urgency. For example, low risk means that the current crop growth is good, the environmental conditions are suitable, and no additional intervention is required; medium risk indicates that there are some potential problems with the crop, but they can be alleviated through appropriate management measures, such as adjusting irrigation or fertilizer application; high risk means that the crop may face major problems, such as serious pest and disease risks or extreme weather impacts, and immediate intervention measures are required.

[0048] The above implementation methods can combine crop growth images, planting record information and growth environment information through a target deep learning model, accurately assess the potential risks of crop growth, and output corresponding risk level information.

[0049] Step S14: matching corresponding insurance clauses for crops according to the target risk level information.

[0050] For example, an insurance product library can be pre-set, which contains different agricultural insurance products from multiple insurance companies. The terms of each product will list the compensation scope, premium structure, compensation conditions and other information in detail. Therefore, a matching algorithm can be used to automatically filter out the most suitable insurance products from the insurance product library based on the target risk level information. For example, if the target risk level of a crop indicates a high flood risk, insurance clauses covering flood damage can be recommended first, and the recommendations can be further refined based on the crop type (such as field crops or fruit trees).

[0051] In other embodiments, insurance can be customized based on the specific conditions of the crop and the farmer's preferences. For example, for certain particularly high-risk crops or regions, higher compensation limits or additional terms (such as early warning systems, damage assessment services, etc.) can be recommended to enhance insurance coverage.

[0052] In the embodiment of the present application, by analyzing the target risk level information of the crops, the automatic matching system is used to select the most appropriate insurance terms for the crops. This can help farmers or agricultural managers reduce risks and ensure that they can obtain timely financial compensation when natural disasters or other adverse conditions occur.

[0053] The crop underwriting method disclosed in the embodiment of the present application obtains crop growth images and target planting record information of crops through sensors, and obtains target remote sensing data of crops through remote sensing technology, wherein the target remote sensing data includes multispectral images. Furthermore, the multispectral images of crops can be analyzed to obtain crop growth environment information, wherein the growth environment information includes soil moisture and vegetation index. Thus, the growth images of crops, the target planting record information of crops, and the growth environment information of crops can be analyzed through the target deep learning model to obtain the target risk level information of the plot where the crops are located. Finally, the corresponding insurance terms can be matched for the crops according to the target risk level information. The present application aims to obtain crop growth images and target planting record information through sensors, combined with multispectral image data collected by remote sensing technology, so as to fully grasp the growth status and environmental information of crops, so as to make the assessment of crop health status and environmental risks more comprehensive and accurate, thereby improving the accuracy of underwriting decisions. Furthermore, by analyzing multispectral image data (such as soil moisture, vegetation index and other growth environment information), combined with the historical planting records of crops and the target deep learning model, the target risk level of the crop plot can be automatically generated, avoiding the subjectivity of traditional manual judgment and improving the efficiency and objectivity of underwriting. In addition, according to the risk level of the plot where the crop is located, the corresponding insurance terms can be automatically matched to the crop to ensure that crops with different risk levels receive appropriate protection. This personalized underwriting plan can better meet the needs of users and reduce the insurance company's compensation risk.

[0054] Optionally, the multispectral image of the crop is analyzed to obtain the growth environment information of the crop, wherein the growth environment information includes soil moisture and vegetation index, and further includes: obtaining real-time meteorological information of the environment where the crop is located; analyzing the growth image of the crop, the target planting record information of the crop and the growth environment information of the crop through a target deep learning model to obtain risk level information of the plot where the crop is located, and further includes: analyzing the growth image of the crop, the target planting record information, the growth environment information and the real-time meteorological information through a target deep learning model to obtain risk level information.

[0055] For example, through the comprehensive analysis of crop growth images, target planting records, growth environment information (such as soil moisture, vegetation index) and real-time meteorological information by the targeted deep learning model, it is possible to more comprehensively assess the health of crops and the potential risks they face.

[0056] The above implementation methods expand the data sources and analysis scope, making risk assessment more comprehensive and accurate, especially for real-time risk prediction of crops in dynamic environments.

[0057] See also Figure 3 , Figure 3 Schematic diagram of a process for obtaining a target deep learning model provided in an embodiment of the present application. Figure 3 As shown, the target deep learning model can be obtained through steps S21 to S24.

[0058] Step S21: Obtain training data.

[0059] The training data includes initial remote sensing data of crops, initial planting record information and historical record information, and the historical record information at least includes historical insurance claim records and crop growth status information.

[0060] Step S22: Screen the training data to obtain crop planting risk data.

[0061] Among them, the planting risk data includes at least one of the crop growth cycle, meteorological information, and historical disaster frequency.

[0062] Step S23: labeling the planting risk data to obtain labeling results.

[0063] Among them, the labeling results include the initial risk level information of the plot where the crops are located.

[0064] Step S24: Train the initial deep learning model using the planting risk data and the annotation results to obtain a target deep learning model.

[0065] Specifically, at least one training data including crop growth cycle, meteorological information, and historical disaster frequency can be collected for training. For example, the training data set can be obtained by manual collection, web crawler, or public data set, and this application does not limit this.

[0066] Furthermore, the training data can be screened to obtain the planting risk data of the crop. Each planting risk data is labeled to obtain the labeling result including the initial risk level information of the plot where the crop is located. Then, each set of planting risk data with standard results is input into the initial deep learning model for supervised learning. When the training end conditions are met, such as when the number of training times reaches the number threshold or the output accuracy of the model reaches the accuracy threshold, the training ends and the training target deep learning model is obtained.

[0067] In an embodiment of the present application, the training data and the annotation results can be input into the initial deep learning model for supervised learning, and then the target deep learning model can be trained.

[0068] Optionally, after the initial deep learning model is trained by planting risk data and annotation results to obtain the target deep learning model, the following steps include: iteratively training the target deep learning model based on the planting risk data and annotation results to extract data features, and calculating the loss function; iteratively training the loss function using a preset method with the purpose of reducing the value of the loss function until the expected threshold requirement is met; and obtaining the iterative target deep learning model based on the loss function after iterative training.

[0069] It can be understood that in order to train a target deep learning model with higher accuracy, the target deep learning model can be repeatedly iterated to continuously reduce the loss function until the loss function meets the expected threshold requirement.

[0070] It should be noted that the present application does not limit the above-mentioned preset method and expected threshold. For example, the preset method may be a gradient descent algorithm, a batch gradient descent algorithm, a stochastic gradient descent algorithm, etc. The present application takes the gradient descent algorithm as an example for explanation.

[0071] The purpose of the gradient descent algorithm is to find the minimum value of the loss function by iteration, or to converge to the minimum value. In a geometric sense, the gradient descent algorithm means that the gradient decreases fastest in the direction opposite to the vector where the function changes the fastest, so it is easier to find the minimum value of the function. Based on this, in an embodiment of the present application, the target deep learning model can be repeatedly iterated and trained by the gradient descent algorithm so that the loss function is continuously reduced, thereby reducing the error of the calculation result.

[0072] In an embodiment of the present application, the target deep learning model is repeatedly iteratively trained using a gradient descent algorithm so that the loss function is continuously reduced to obtain an iterative target deep learning model, and thus a more accurate result can be obtained based on the iterative target deep learning model.

[0073] Optionally, after obtaining the training data, the method includes: performing preprocessing operations on the training data to obtain preprocessed training data, wherein the preprocessing operations include at least one of missing value processing, normalization processing, and feature selection; screening the training data to obtain crop planting risk data, including: screening the preprocessed training data to obtain crop planting risk data.

[0074] By screening the preprocessed training data, high-quality data sets related to crop planting risks can be extracted. These data include various factors that may affect the success of crop planting, such as: crop growth status (based on the analysis results of multispectral images), soil and climate conditions (such as soil moisture, temperature, precipitation, etc.) and pest and disease risks (if there is relevant historical data or forecast information).

[0075] The screened risk data will serve as an important input for training models, not only for predicting crop planting risks, but also for providing a basis for subsequent model optimization and planting risk assessment.

[0076] The above implementation methods extract high-quality data closely related to planting risks through data preprocessing and screening, making crop risk assessment more accurate and effective.

[0077] Optionally, after matching corresponding insurance clauses for crops according to the target risk level information, it also includes: obtaining an updated growth image of the crop, updated target planting record information and updated growth environment information; and obtaining user feedback information based on the insurance clauses, wherein the feedback information at least includes user satisfaction information with the insurance clauses; analyzing the updated growth image, updated target planting record information and updated growth environment information through an updated target deep learning model to obtain updated target risk level information; wherein the updated target deep learning model is trained based on training data and feedback information; and matching updated insurance clauses for crops according to the updated target risk level information.

[0078] For example, after the initial risk assessment, the updated growth images of crops, target planting record information, and growth environment information can be continuously obtained to reflect the changes of crops over time and monitor their health and environmental conditions in real time. After the user uses the insurance terms, user feedback information is also collected, which includes user satisfaction information with the current insurance terms, which helps to understand whether the insurance terms meet user needs or need to be improved.

[0079] Furthermore, the target deep learning model can be retrained and optimized based on the updated crop data and user feedback information. By introducing user feedback information, the model can better adjust the risk assessment strategy and improve the accuracy of prediction and user experience.

[0080] For example, the updated risk level is more in line with the current condition of the crop and the latest environmental changes. Based on the updated risk level information, the updated insurance terms can be matched for the crop, ensuring that the terms are more accurate and dynamic, meeting the actual needs of the crop and the user's expectations for insurance.

[0081] The above implementation methods can not only continuously acquire and analyze new crop data, but also combine users' actual feedback information to iterate the model, thereby continuously improving the effect of matching risk assessment with insurance terms. This approach makes the insurance plan more personalized and timely.

[0082] Optionally, after matching corresponding insurance clauses for crops according to target risk level information, the method further includes: displaying the insurance clauses through a display interface, so that a user can confirm the insurance clauses through the display interface and complete the insurance of the crops.

[0083] The above implementation method displays insurance terms through a display interface, providing users with a convenient and intuitive way to confirm and complete insurance. It not only improves the user's sense of participation, but also simplifies the insurance process, allowing users to make more effective decisions based on the system's suggestions. This step ensures the user's initiative and transparency in the entire risk assessment and insurance matching process, and ultimately achieves a seamless connection from risk assessment to insurance.

[0084] See also Figure 4 , Figure 4 It is a schematic block diagram of a crop insurance device provided in an embodiment of the present application. The crop insurance device can be configured in a server to execute the steps of the aforementioned crop insurance method.

[0085] like Figure 4 As shown, the crop insurance device 200 includes: an acquisition module 201 , an analysis module 202 and a matching module 203 .

[0086] An acquisition module 201 is used to acquire a crop growth image and target planting record information of the crop through a sensor, and to acquire target remote sensing data of the crop through remote sensing technology, wherein the target remote sensing data includes a multispectral image;

[0087] An analysis module 202 is used to analyze the multispectral image of the crop to obtain the growth environment information of the crop, wherein the growth environment information includes soil moisture and vegetation index;

[0088] The analysis module 202 is further used to analyze the growth image of the crop, the target planting record information of the crop, and the growth environment information of the crop through a target deep learning model to obtain target risk level information of the plot where the crop is located;

[0089] The matching module 203 is used to match corresponding insurance clauses for the crop according to the target risk level information.

[0090] The analysis module 202 is also used to obtain real-time meteorological information of the environment where the crop is located; the growth image of the crop, the target planting record information, the growth environment information and the real-time meteorological information are analyzed through the target deep learning model to obtain the risk level information.

[0091] The acquisition module 201 is also used to acquire training data, wherein the training data includes initial remote sensing data, initial planting record information and historical record information of the crop, and the historical record information includes at least historical insurance claim records and crop growth status information; the training data is screened to obtain planting risk data of the crop, wherein the planting risk data includes at least at least one of crop growth cycle, meteorological information, and historical disaster frequency; the planting risk data is labeled to obtain a labeling result, wherein the labeling result includes initial risk level information of the plot where the crop is located; an initial deep learning model is trained using the planting risk data and the labeling result to obtain the target deep learning model.

[0092] The acquisition module 201 is also used to iteratively train the target deep learning model based on the planting risk data and the annotation results to extract data features and calculate the loss function; iteratively train the loss function using a preset method for the purpose of reducing the value of the loss function until the expected threshold requirement is met; based on the loss function after iterative training, obtain the iterative target deep learning model.

[0093] The acquisition module 201 is also used to perform a preprocessing operation on the training data to obtain preprocessed training data, wherein the preprocessing operation includes at least one of missing value processing, normalization processing and feature selection; and screen the preprocessed training data to obtain the planting risk data of the crop.

[0094] The acquisition module 201 is also used to obtain an updated growth image of the crop, updated target planting record information and updated growth environment information; and to obtain user feedback information based on the insurance terms, wherein the feedback information at least includes the user's satisfaction information with the insurance terms; the updated growth image, the updated target planting record information and the updated growth environment information are analyzed by an updated target deep learning model to obtain updated target risk level information; wherein the updated target deep learning model is trained based on training data and the feedback information; and updated insurance terms are matched for the crop according to the updated target risk level information.

[0095] The matching module 203 is further used to display the insurance terms through a display interface, so that the user can confirm the insurance terms through the display interface and complete the insurance of the crop.

[0096] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and modules and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0097] The method and apparatus of the present application can be used in many general or special computing system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer terminal devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices.

[0098] Exemplarily, the above method and apparatus may be implemented in the form of a computer program. The computer program may be implemented in the form of a computer program. Figure 5 Runs on the computer device shown.

[0099] See also Figure 5 , Figure 5 1 is a schematic diagram of a computer device provided in an embodiment of the present application. The computer device may be a server.

[0100] like Figure 5As shown, the computer device includes a processor, a memory and a network interface connected through a system bus, wherein the memory may include a volatile storage medium, a non-volatile storage medium and an internal memory.

[0101] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute the steps of any crop insurance method.

[0102] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0103] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute the steps of the insurance method for any crop.

[0104] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art can understand that the structure of the computer device is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0105] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0106] In some embodiments, the processor is used to run a computer program stored in a memory to implement the following steps: obtaining a growth image of a crop and target planting record information of the crop through a sensor, and obtaining target remote sensing data of the crop through remote sensing technology, wherein the target remote sensing data includes a multispectral image; analyzing the multispectral image of the crop to obtain growth environment information of the crop, wherein the growth environment information includes soil moisture and a vegetation index; analyzing the growth image of the crop, the target planting record information of the crop and the growth environment information of the crop through a target deep learning model to obtain target risk level information of the plot where the crop is located; and matching corresponding insurance terms for the crop according to the target risk level information.

[0107] In some embodiments, the processor is also used to obtain real-time meteorological information of the environment where the crop is located; the growth image of the crop, the target planting record information, the growth environment information and the real-time meteorological information are analyzed through the target deep learning model to obtain the risk level information.

[0108] In some embodiments, the processor is also used to obtain training data, wherein the training data includes initial remote sensing data, initial planting record information and historical record information of the crop, and the historical record information includes at least historical insurance claim records and crop growth status information; the training data is screened to obtain planting risk data of the crop, wherein the planting risk data includes at least at least one of crop growth cycle, meteorological information, and historical disaster frequency; the planting risk data is labeled to obtain a labeling result, wherein the labeling result includes initial risk level information of the plot where the crop is located; an initial deep learning model is trained using the planting risk data and the labeling result to obtain the target deep learning model.

[0109] In some embodiments, the processor is also used to iteratively train the target deep learning model based on the planting risk data and the annotation results to extract data features and calculate a loss function; iteratively train the loss function using a preset method for the purpose of reducing the value of the loss function until the expected threshold requirement is met; based on the loss function after iterative training, obtain the iterative target deep learning model.

[0110] In some embodiments, the processor is further used to perform a preprocessing operation on the training data to obtain preprocessed training data, wherein the preprocessing operation includes at least one of missing value processing, normalization processing, and feature selection; and screen the preprocessed training data to obtain the planting risk data of the crop.

[0111] In some embodiments, the processor is also used to obtain an updated growth image of the crop, updated target planting record information, and updated growth environment information; and, obtain user feedback information based on the insurance terms, wherein the feedback information at least includes the user's satisfaction information with the insurance terms; analyze the updated growth image, the updated target planting record information, and the updated growth environment information through an updated target deep learning model to obtain updated target risk level information; wherein the updated target deep learning model is trained based on training data and the feedback information; and match updated insurance terms for the crop according to the updated target risk level information.

[0112] In some embodiments, the processor is further configured to display the insurance terms through a display interface, so that the user can confirm the insurance terms through the display interface and complete the insurance of the crop.

[0113] An embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. The computer program includes program instructions, and when the program instructions are executed, the steps of the crop insurance method provided in the embodiment of the present application are implemented.

[0114] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc., equipped on the computer device.

[0115] Furthermore, the computer-readable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function, and the like.

[0116] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A crop insurance method, characterized in that: The method comprises: Acquire crop growth images and target planting record information of the crop through sensors, and acquire target remote sensing data of the crop through remote sensing technology, wherein the target remote sensing data includes multispectral images; Analyzing the multispectral image of the crop to obtain growth environment information of the crop, wherein the growth environment information includes soil moisture and vegetation index; Analyzing the growth image of the crop, the target planting record information of the crop, and the growth environment information of the crop through a target deep learning model to obtain target risk level information of the plot where the crop is located; The corresponding insurance terms are matched for the crops according to the target risk level information.

2. The method according to claim 1, characterized in that: The method further comprises: analyzing the multispectral image of the crop to obtain the growth environment information of the crop, wherein the growth environment information includes soil moisture and vegetation index, and further comprises: Obtaining real-time meteorological information of the environment where the crop is located; The target deep learning model is used to analyze the growth image of the crop, the target planting record information of the crop, and the growth environment information of the crop to obtain the risk level information of the plot where the crop is located, and further includes: The target deep learning model is used to analyze the growth image of the crop, the target planting record information, the growth environment information and the real-time meteorological information to obtain the risk level information.

3. The method according to claim 1, characterized in that The target deep learning model is obtained in the following way: Acquiring training data, wherein the training data includes initial remote sensing data, initial planting record information, and historical record information of the crop, wherein the historical record information includes at least historical insurance claim records and crop growth status information; Screening the training data to obtain planting risk data of the crop, wherein the planting risk data includes at least one of crop growth cycle, meteorological information, and historical disaster frequency; Labeling the planting risk data to obtain a labeling result, wherein the labeling result includes initial risk level information of the plot where the crop is located; The initial deep learning model is trained using the planting risk data and the annotation results to obtain the target deep learning model.

4. The method according to claim 3, characterized in that After the initial deep learning model is trained by the planting risk data and the annotation results to obtain the target deep learning model, the method includes: Iteratively train the target deep learning model based on the planting risk data and the annotation results to extract data features, and calculate a loss function; Iteratively training the loss function using a preset method for the purpose of reducing the value of the loss function until the expected threshold value is met; Based on the loss function after iterative training, the target deep learning model after iteration is obtained.

5. The method according to claim 3, characterized in that: After the training data is obtained, the following steps are included: Performing a preprocessing operation on the training data to obtain preprocessed training data, wherein the preprocessing operation includes at least one of missing value processing, normalization processing, and feature selection; The step of screening the training data to obtain the crop planting risk data includes: The preprocessed training data is screened to obtain the planting risk data of the crop.

6. The method according to claim 1, characterized in that After matching corresponding insurance clauses for the crops according to the target risk level information, the method further includes: Acquiring an updated growth image of the crop, updated target planting record information, and updated growth environment information; and, Acquiring feedback information from the user based on the insurance terms, wherein the feedback information at least includes information on the user's satisfaction with the insurance terms; The updated growth image, the updated target planting record information and the updated growth environment information are analyzed by an updated target deep learning model to obtain updated target risk level information; wherein the updated target deep learning model is trained based on the training data and the feedback information; Updated insurance terms are matched for the crop according to the updated target risk level information.

7. The method according to claim 1, characterized in that After matching corresponding insurance clauses for the crops according to the target risk level information, the method further includes: The insurance terms are displayed through a display interface so that the user can confirm the insurance terms through the display interface and complete the insurance of the crop.

8. A crop insurance device, characterized in that: The crop insurance device comprises: An acquisition module, used to acquire crop growth images and target planting record information of the crop through sensors, and to acquire target remote sensing data of the crop through remote sensing technology, wherein the target remote sensing data includes multispectral images; An analysis module, used for analyzing the multispectral image of the crop to obtain the growth environment information of the crop, wherein the growth environment information includes soil moisture and vegetation index; The analysis module is further used to analyze the growth image of the crop, the target planting record information of the crop, and the growth environment information of the crop through a target deep learning model to obtain target risk level information of the plot where the crop is located; A matching module is used to match corresponding insurance clauses for the crop according to the target risk level information.

9. A computer device, characterized in that: include: A memory and a processor; wherein the memory is connected to the processor and is used to store a program; the processor is used to implement the steps of the crop insurance method according to any one of claims 1 to 7 by running the program stored in the memory.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, enables the processor to implement the steps of the crop insurance method according to any one of claims 1 to 7.

Images