Risk assessment method and device based on artificial intelligence, computer equipment and medium

Abstract

The invention belongs to the field of artificial intelligence and the field of financial science and technology, and relates to an artificial intelligence-based risk assessment method, which comprises the steps of obtaining live pig image data uploaded by a farmer; preprocessing the live pig image data to obtain target image data; performing evaluation processing on the target image data based on a health evaluation model to generate live pig health condition prediction data; estimating the target image data based on a weight estimation model to generate live pig weight prediction data; performing statistical processing on the live pig health condition prediction data and the live pig weight prediction data based on a risk assessment model to obtain a risk score; generating a target risk level based on the risk score; and returning the target risk level to the farmer. The invention further provides a risk assessment device based on artificial intelligence, computer equipment and a storage medium. In addition, the target risk level may be stored in a blockchain. According to the invention, based on the use of multiple models, the processing efficiency and evaluation accuracy of risk evaluation of live pig breeding are improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence development technology and financial technology, and in particular to artificial intelligence-based risk assessment methods, devices, computer equipment and storage media. Background Art

[0002] In the traditional pig farming insurance industry, the insurance assessment process is a crucial link, which is directly related to the risk protection level of farmers and the risk management ability of insurance companies. However, for a long time, this link has mainly relied on manual on-site inspections and experience judgments, which have many limitations and shortcomings.

[0003] First, manual on-site inspections are time-consuming, laborious, and inefficient. Since pig farms are usually large in scale and have a large number of pigs, inspecting each pig one by one is not only time-consuming and laborious, but also easily affected by external factors such as weather and traffic, which hinders the progress of the inspection. In addition, with the expansion of breeding scale and the increase of breeding density, the difficulty and cost of manual inspections have also increased accordingly, which has brought considerable burdens to insurance companies and farmers.

[0004] Secondly, experience-based judgment is highly subjective, and it is difficult to guarantee the accuracy of the evaluation results. Traditional evaluation methods often rely on the professional knowledge and experience accumulation of the evaluators, but there are differences in experience and skill levels between different evaluators, which may lead to greater subjectivity and inconsistency in the evaluation results. Summary of the invention

[0005] The purpose of the embodiments of the present application is to propose a risk assessment method, device, computer equipment and storage medium based on artificial intelligence, so as to solve the technical problems that the existing risk assessment method of pig farming mainly relies on manual on-site investigation and experience judgment, which is inefficient and difficult to ensure the accuracy of the assessment results.

[0006] In order to solve the above technical problems, the embodiment of the present application provides a risk assessment method based on artificial intelligence, which adopts the following technical solutions:

[0007] Obtain pig image data uploaded by farmers;

[0008] Preprocessing the pig image data to obtain corresponding target image data;

[0009] The target image data is evaluated and processed based on a pre-trained health assessment model to generate corresponding pig health status prediction data;

[0010] The target image data is estimated based on a pre-trained weight estimation model to generate corresponding pig weight prediction data;

[0011] Performing statistical processing on the predicted data of the health status of the pigs and the predicted data of the weight of the pigs based on a preset risk assessment model to obtain corresponding risk scores;

[0012] generating a corresponding target risk level based on the risk score;

[0013] The target risk level is returned to the farmer.

[0014] Furthermore, the step of performing statistical processing on the predicted data of the health status of the pigs and the predicted data of the weight of the pigs based on the preset risk assessment model to obtain corresponding risk scores specifically includes:

[0015] The pig health status prediction data is converted based on a preset conversion strategy to obtain a corresponding health score;

[0016] Obtaining a first weight corresponding to the health score;

[0017] Obtaining a second weight corresponding to the pig weight prediction data;

[0018] Based on the risk assessment model, the health score, the pig weight prediction data, the first weight and the second weight are calculated and processed using a preset risk score calculation formula to obtain a corresponding calculation result;

[0019] The calculation result is used as the risk score.

[0020] Furthermore, the step of generating a corresponding target risk level based on the risk score specifically includes:

[0021] Obtaining a first risk score range corresponding to a first preset risk level, obtaining a second risk score range corresponding to a second preset risk level, and obtaining a third risk score range corresponding to a third preset risk level;

[0022] If the risk score is within the first risk score range, setting the target risk level corresponding to the risk score to the first preset risk level;

[0023] If the risk score is within the second risk score range, setting the target risk level corresponding to the risk score to the second preset risk level;

[0024] If the risk score is within the third risk score range, the target risk level corresponding to the risk score is set to the third preset risk level.

[0025] Furthermore, the step of preprocessing the pig image data to obtain corresponding target image data specifically includes:

[0026] Performing noise removal processing on the pig image data to obtain corresponding first image data;

[0027] Performing image resizing processing on the first image data to obtain corresponding second image data;

[0028] performing contrast enhancement processing on the second image data to obtain corresponding third image data;

[0029] performing standardization processing on the third image data to obtain corresponding fourth image data;

[0030] The fourth image data is used as the target image data.

[0031] Furthermore, after the step of generating a corresponding target risk level based on the risk score, the method further includes:

[0032] Calculate the total value of pigs based on the predicted pig weight data;

[0033] Obtaining adjustment demand information corresponding to the breeder;

[0034] Adjust the total value of the live pigs based on the adjustment demand information to obtain a corresponding insurance amount;

[0035] Based on the target risk level, the insurance amount is calculated using a preset premium calculation strategy to obtain a corresponding target premium;

[0036] generating a corresponding insurance proposal based on the insurance amount and the target premium;

[0037] The insurance proposal is sent to the farmer.

[0038] Furthermore, the step of calculating the insurance amount based on the target risk level using a preset premium calculation strategy to obtain a corresponding target premium specifically includes:

[0039] Get the preset rate table;

[0040] Obtaining a target rate that matches the insurance amount and the target risk level from the rate table;

[0041] Obtaining a target credit rating for the farmer;

[0042] determining a premium discount corresponding to the target credit rating;

[0043] Get the preset premium calculation formula;

[0044] The insurance amount, the target rate and the premium discount are calculated based on the premium calculation formula to obtain the corresponding target premium.

[0045] Furthermore, before the step of evaluating the target image data based on the pre-trained health assessment model to generate corresponding pig health status prediction data, the method further includes:

[0046] Obtain pre-collected historical pig image data;

[0047] Performing data preprocessing on the historical pig image data to obtain corresponding sample data;

[0048] Calling a preset initial neural network model; wherein the initial neural network model includes at least a convolutional layer, a pooling layer, a fully connected layer, and an output layer;

[0049] Determine the target loss function and target optimization algorithm;

[0050] Based on the target loss function and the target optimization algorithm, the initial neural network model is trained using the sample data to obtain a trained specified neural network model;

[0051] The designated neural network model is used as the health assessment model.

[0052] In order to solve the above technical problems, the embodiment of the present application also provides a risk assessment device based on artificial intelligence, which adopts the following technical solution:

[0053] The first acquisition module is used to acquire pig image data uploaded by farmers;

[0054] A preprocessing module, used for preprocessing the pig image data to obtain corresponding target image data;

[0055] An evaluation module, used to evaluate and process the target image data based on a pre-trained health evaluation model to generate corresponding pig health status prediction data;

[0056] An estimation module, used to estimate the target image data based on a pre-trained weight estimation model to generate corresponding pig weight prediction data;

[0057] A statistical module, used for performing statistical processing on the predicted data of the health status of the pigs and the predicted data of the weight of the pigs based on a preset risk assessment model to obtain corresponding risk scores;

[0058] A first generating module, configured to generate a corresponding target risk level based on the risk score;

[0059] A return module is used to return the target risk level to the farmer.

[0060] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0061] Obtain pig image data uploaded by farmers;

[0062] Preprocessing the pig image data to obtain corresponding target image data;

[0063] The target image data is evaluated and processed based on a pre-trained health assessment model to generate corresponding pig health status prediction data;

[0064] The target image data is estimated based on a pre-trained weight estimation model to generate corresponding pig weight prediction data;

[0065] Performing statistical processing on the predicted data of the health status of the pigs and the predicted data of the weight of the pigs based on a preset risk assessment model to obtain corresponding risk scores;

[0066] generating a corresponding target risk level based on the risk score;

[0067] The target risk level is returned to the farmer.

[0068] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0069] Obtain pig image data uploaded by farmers;

[0070] Preprocessing the pig image data to obtain corresponding target image data;

[0071] The target image data is evaluated and processed based on a pre-trained health assessment model to generate corresponding pig health status prediction data;

[0072] The target image data is estimated based on a pre-trained weight estimation model to generate corresponding pig weight prediction data;

[0073] Performing statistical processing on the predicted data of the health status of the pigs and the predicted data of the weight of the pigs based on a preset risk assessment model to obtain corresponding risk scores;

[0074] generating a corresponding target risk level based on the risk score;

[0075] The target risk level is returned to the farmer.

[0076] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0077] The present application first obtains the pig image data uploaded by the farmer; and pre-processes the pig image data to obtain the corresponding target image data; then evaluates and processes the target image data based on a pre-trained health assessment model to generate corresponding pig health status prediction data; then estimates and processes the target image data based on a pre-trained weight estimation model to generate corresponding pig weight prediction data; subsequently, statistically processes the pig health status prediction data and the pig weight prediction data based on a preset risk assessment model to obtain a corresponding risk score; further generates a corresponding target risk level based on the risk score; and finally returns the target risk level to the farmer. After obtaining the pig image data uploaded by the farmer and pre-processing the pig image data to obtain the target image data, the present application performs risk assessment processing on the target image data based on the use of a pre-trained health assessment model, a weight estimation model, and a risk assessment model, so as to quickly and accurately generate the corresponding risk score, and then automatically generate the corresponding target risk level based on the risk score, effectively improving the processing efficiency and assessment accuracy of the risk assessment of pig farming, and ensuring the accuracy of the obtained target risk level. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. 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.

[0079] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0080] Figure 2 A flowchart of an embodiment of an artificial intelligence-based risk assessment method according to the present application;

[0081] Figure 3 is a schematic structural diagram of an embodiment of an artificial intelligence-based risk assessment device according to the present application;

[0082] Figure 4 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0083] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0084] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0085] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0086] like Figure 1 As shown, the system architecture 100 may include a terminal device 101, a network 102 and a server 103. The terminal device 101 may be a laptop 1011, a tablet computer 1012 or a mobile phone 1013. The network 102 is used to provide a medium for a communication link between the terminal device 101 and the server 103. The network 102 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0087] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0088] The terminal device 101 can be any electronic device with a display screen and supporting web browsing. In addition to a laptop computer 1011, a tablet computer 1012 or a mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV), a laptop computer, a desktop computer, etc.

[0089] The server 103 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal device 101 .

[0090] It should be noted that the artificial intelligence-based risk assessment method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the artificial intelligence-based risk assessment device is generally set in the server / terminal device.

[0091] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0092] Continue to refer Figure 2 , showing a flow chart of an embodiment of an artificial intelligence-based risk assessment method according to the present application. According to different requirements, the order of the steps in the flow chart can be changed, and some steps can be omitted. The artificial intelligence-based risk assessment method provided in the embodiment of the present application can be applied to any scenario that requires risk assessment, and the artificial intelligence-based risk assessment method can be applied to products in these scenarios, for example, pig farming insurance assessment in the field of financial insurance. The artificial intelligence-based risk assessment method comprises the following steps:

[0093] Step S201, obtaining pig image data uploaded by farmers.

[0094] In this embodiment, the electronic device (e.g. Figure 1The server / terminal device shown in the figure) can obtain the pig image data through a wired connection or a wireless connection. It should be noted that the above wireless connection method may include but is not limited to 3G / 4G / 5G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future. The execution subject of this application may specifically be an electronic device equipped with a risk assessment system for pig farming. The farmer pre-installs an intelligent application (App) of the risk assessment system on the electronic device and ensures that the electronic device has the ability to shoot high-quality images and videos. The initial settings of the App can be completed in advance, including account registration, login and basic information entry (such as farm location, scale, etc.). Specifically, the above pig image data can be used by farmers to take images and videos of pigs in the farm by using the shooting function in the App. Among them, all pigs should be covered as much as possible during shooting to ensure sufficient light and clear images so that they can be accurately identified. And the pig image data (including images and videos) captured will be automatically saved on the electronic device.

[0095] Step S202, preprocessing the pig image data to obtain corresponding target image data.

[0096] In this embodiment, the above-mentioned preprocessing of the pig image data to obtain the specific implementation process of the corresponding target image data will be further described in detail in subsequent specific embodiments of the present application, and will not be elaborated in detail here.

[0097] Step S203, evaluating and processing the target image data based on a pre-trained health assessment model to generate corresponding pig health status prediction data.

[0098] In this embodiment, the target image data is input into a pre-trained health assessment model, so that the target image data is evaluated and processed by the health assessment model and the corresponding pig health status prediction data is output. The pig health status prediction data may include health, injury, disease, etc. In addition, the construction process of the above health assessment model will be further described in detail in the subsequent specific embodiments of this application, and will not be elaborated on here.

[0099] Step S204, performing estimation processing on the target image data based on the pre-trained weight estimation model to generate corresponding pig weight prediction data.

[0100] In this embodiment, the above-mentioned weight estimation model is a pre-built regression model that combines the body shape features of the pig, such as aspect ratio, body shape, etc. to predict the weight, such as ResNet, DenseNet, etc. The specific processing process of the weight estimation model includes: 1. Input layer and feature extraction. ResNet and DenseNet are both deep neural networks composed of multiple convolutional blocks, which can extract rich features from the input image. These features will be used to predict the weight of the pig. 2. Output layer In the regression task, the output layer is usually one or several nodes, which are used to predict one or more continuous variables. In this case, we only have one output node for predicting the weight of the pig. 3. Loss function. For regression problems, common loss functions are mean squared error (MSE) or root mean square error (RMSE). 4. Model architecture: ResNet. ResNet uses residual blocks to alleviate the gradient vanishing problem in deep networks. A typical residual block contains two convolutional layers, each followed by a batch normalization layer and a ReLU activation function. The output of the residual block is the sum of the input and the output of the two convolutional layers. Suppose there is a residual block with an input of x and an output of H(x), then the output of the residual block can be expressed as: y=H(x)=F(x,W)+x. Among them, F(x,W) is a function composed of a convolutional layer, a batch normalization layer, and a ReLU activation function, and W is the weight of the function. 5. Model architecture example: DenseNet. DenseNet uses the concept of dense connections, that is, each layer is directly connected to each subsequent layer. This means that the input of the llth layer is the set of the outputs of all previous layers. In DenseNet, the output of each layer is passed to all subsequent layers as part of the input. 6. Final prediction. The final prediction is usually achieved through a fully connected layer, which converts the extracted features into predicted values. Assuming z is the output of the fully connected layer, w is the weight matrix, and b is the bias vector, then the predicted weight can be expressed as:

[0101] In addition, the training process of the weight estimation model includes: 1. Model initialization: A pre-trained model can be used as a basis to reduce training time and improve the generalization ability of the model. 2. Data partitioning: The data set is divided into a training set, a validation set, and a test set, usually in a ratio of 80%:10%:10%, for model training, hyperparameter adjustment, and final performance evaluation. 3. Training process: Use the training set data to train the model, optimize the model weights through back propagation and gradient descent until the model converges, and use the validation set for performance monitoring and overfitting prevention during the period. 4. Performance evaluation: Evaluate the accuracy, recall rate, F1 score and other indicators of the model on the test set to ensure that the model has good generalization ability on unseen data. 5. Model optimization and testing. Hyperparameter tuning: Find the best combination of model parameters through methods such as grid search, random search or Bayesian optimization. Model fusion: Consider fusing the prediction results of multiple models, using weighted average or voting mechanism to improve the accuracy and robustness of the overall prediction. Error analysis: Conduct in-depth analysis of cases where the model predicts errors, identify the model's weaknesses, and provide direction for subsequent iterations. 6. Continuous iteration and optimization. Model update: Regularly collect new data and retrain the model to adapt to changes in the pig farming environment. Performance monitoring: Continuously monitor the performance of the model to ensure recognition accuracy and stability, and optimize and retrain the model when necessary.

[0102] Step S205, statistically processing the pig health status prediction data and the pig weight prediction data based on a preset risk assessment model to obtain corresponding risk scores.

[0103] In this embodiment, the above-mentioned specific implementation process of statistically processing the pig health status prediction data and the pig weight prediction data based on the preset risk assessment model to obtain the corresponding risk score will be further described in detail in the subsequent specific embodiments of this application and will not be elaborated here.

[0104] Step S206: Generate a corresponding target risk level based on the risk score.

[0105] In this embodiment, the specific implementation process of generating the corresponding target risk level based on the risk score will be further described in detail in subsequent specific embodiments of the present application, and will not be elaborated on here.

[0106] Step S207, returning the target risk level to the farmer.

[0107] In this embodiment, a special risk level display page can be designed to display the target risk level in a clear and intuitive manner. For example, the target risk level can be displayed in the form of a chart, a list or a comparison table to return the target risk level to the farmer.

[0108] The present application first obtains the pig image data uploaded by the farmer; and pre-processes the pig image data to obtain the corresponding target image data; then evaluates and processes the target image data based on a pre-trained health assessment model to generate corresponding pig health status prediction data; then estimates and processes the target image data based on a pre-trained weight estimation model to generate corresponding pig weight prediction data; subsequently, statistically processes the pig health status prediction data and the pig weight prediction data based on a preset risk assessment model to obtain a corresponding risk score; further generates a corresponding target risk level based on the risk score; and finally returns the target risk level to the farmer. After obtaining the pig image data uploaded by the farmer and pre-processing the pig image data to obtain the target image data, the present application performs risk assessment processing on the target image data based on the use of a pre-trained health assessment model, a weight estimation model, and a risk assessment model, so as to quickly and accurately generate the corresponding risk score, and then automatically generate the corresponding target risk level based on the risk score, effectively improving the processing efficiency and assessment accuracy of the risk assessment of pig farming, and ensuring the accuracy of the obtained target risk level.

[0109] In some optional implementations, step S205 includes the following steps:

[0110] The pig health status prediction data is converted based on a preset conversion strategy to obtain a corresponding health score.

[0111] In this embodiment, the health score corresponding to the predicted data of the health status of the pigs can be queried based on a preset conversion data table. The conversion data table is a pre-constructed data table storing a one-to-one correspondence between the health status data of the pigs and the health score. In addition, the predicted data of the health status of the pigs may include health, injury, and disease. Exemplarily, the health score corresponding to the health status data of the pigs being healthy is 0, the health score corresponding to the health status data of the pigs being injured is 1, and the health score corresponding to the health status data of the pigs being injured is 2.

[0112] Obtain a first weight corresponding to the health score.

[0113] In this embodiment, each input feature has a corresponding weight within the model of the risk assessment model. These weights are automatically learned during the model training process through an optimization algorithm (such as gradient descent) to maximize the model's prediction accuracy for the risk level.

[0114] A second weight corresponding to the pig weight prediction data is obtained.

[0115] In this embodiment, a suitable risk assessment model can be selected according to business requirements and data characteristics. Preferably, the risk assessment model is a regression model based on statistics, and specifically, linear regression, logistic regression, polynomial regression, etc. can be used. Preferably, for the risk assessment of pig farms, since the goal is to obtain a continuous risk level numerical score, linear regression or polynomial regression may be a suitable choice.

[0116] Based on the risk assessment model, the health score, the pig weight prediction data, the first weight and the second weight are calculated and processed using a preset risk score calculation formula to obtain a corresponding calculation result.

[0117] In this embodiment, after the health score and the pig weight prediction data are input into the risk assessment model, different weights (first weight and second weight) are assigned in the risk assessment model according to the importance of various risk factors, and the health score, the pig weight prediction data, the first weight and the second weight are calculated based on the internal statistical algorithm to obtain the corresponding calculation results. The above statistical algorithm can be specifically a weighted summation algorithm.

[0118] The calculation result is used as the risk score.

[0119] The present application converts the predicted data of the health status of the pigs based on a preset conversion strategy to obtain a health score. Then, a first weight corresponding to the health score is obtained, and a second weight corresponding to the predicted weight data of the pigs is obtained. Then, based on the risk assessment model, the health score, the predicted weight data of the pigs, the first weight and the second weight are calculated using a preset risk score calculation formula, so that the corresponding risk score can be automatically and accurately generated, the calculation efficiency of the risk score is improved, and the accuracy of the obtained risk score is ensured.

[0120] In some optional implementations of this embodiment, step S206 includes the following steps:

[0121] A first risk score range corresponding to a first preset risk level is obtained, a second risk score range corresponding to a second preset risk level is obtained, and a third risk score range corresponding to a third preset risk level is obtained.

[0122] In this embodiment, according to the actual business needs related to pig farming, a series of risk levels corresponding to pig farming (including high risk, medium risk, and low risk) are predefined, and a corresponding risk score range is set for each risk level. Among them, the first preset risk level may correspond to high risk, the second preset risk level may correspond to medium risk, and the first preset risk level may correspond to low risk.

[0123] If the risk score is within the first risk score range, the target risk level corresponding to the risk score is set to the first preset risk level.

[0124] In this embodiment, the risk score obtained by the risk assessment model is mapped to the predefined risk level, that is, the risk score is compared with the risk score range of each risk level. If it is detected that the risk score is within the first risk score range corresponding to the first preset risk level, the target risk level corresponding to the risk score is set to the above-mentioned first preset risk level.

[0125] If the risk score is within the second risk score range, the target risk level corresponding to the risk score is set to the second preset risk level.

[0126] In this embodiment, if it is detected that the risk score is within a second risk score range corresponding to a second preset risk level, the target risk level corresponding to the risk score is set to the second preset risk level.

[0127] If the risk score is within the third risk score range, the target risk level corresponding to the risk score is set to the third preset risk level.

[0128] In this embodiment, if it is detected that the risk score is within a third risk score range corresponding to a third preset risk level, the target risk level corresponding to the risk score is set to the third preset risk level.

[0129] The present application obtains a first risk score range corresponding to a first preset risk level, obtains a second risk score range corresponding to a second preset risk level, and obtains a third risk score range corresponding to a third preset risk level; if the risk score is in the first risk score range, the target risk level corresponding to the risk score is set to the first preset risk level; if the risk score is in the second risk score range, the target risk level corresponding to the risk score is set to the second preset risk level; if the risk score is in the third risk score range, the target risk level corresponding to the risk score is set to the third preset risk level. The present application maps the risk score obtained by the risk assessment model with the predefined risk level, that is, the risk score is compared with the risk score range of each risk level, so as to quickly and accurately determine the target risk level corresponding to the risk score according to the obtained mapping result, and effectively ensure the data accuracy of the obtained target risk level.

[0130] In some optional implementations, step S202 includes the following steps:

[0131] The pig image data is subjected to noise removal processing to obtain corresponding first image data.

[0132] In this embodiment, the noise removal process may include image filtering and threshold processing. Among them, image filtering includes: using filters (such as Gaussian filtering, median filtering) to smooth the image and reduce random noise. These filters can help remove fine particles or spots in the image while maintaining edge information. Threshold processing includes: for binary images or situations where they need to be converted into binary images, threshold processing can be used to remove noise. By setting a pixel value threshold, pixels below or above the threshold are regarded as background or noise and removed.

[0133] The first image data is subjected to image resizing processing to obtain corresponding second image data.

[0134] In this embodiment, the image resizing process may include scaling and cropping. The scaling process includes scaling the image to a uniform size to meet the input requirements of the model. This can be achieved by using an interpolation algorithm (such as bilinear interpolation, bicubic interpolation) to maintain image quality. The cropping process includes cropping the image if the image only contains a portion of the target area or to remove irrelevant background information.

[0135] Perform contrast enhancement processing on the second image data to obtain corresponding third image data.

[0136] In this embodiment, the above-mentioned contrast enhancement processing may include: histogram equalization processing and contrast stretching processing. Among them, the histogram equalization processing includes: enhancing the contrast by adjusting the histogram of the image to make the pixel value distribution of the image more uniform. This helps to improve the visual effect of the image and may improve the accuracy of AI recognition. The contrast stretching processing includes: stretching the contrast by adjusting the minimum and maximum pixel values ​​of the image to make the dark part of the image darker and the bright part brighter.

[0137] The third image data is standardized to obtain corresponding fourth image data.

[0138] In this embodiment, the third image data is standardized to ensure that all images are within the same numerical range, which is helpful for model training and convergence.

[0139] The fourth image data is used as the target image data.

[0140] The present application performs noise removal processing on the pig image data to obtain the corresponding first image data; then performs image resizing processing on the first image data to obtain the corresponding second image data; then performs contrast enhancement processing on the second image data to obtain the corresponding third image data; then performs standardization processing on the third image data to obtain the corresponding fourth image data; and finally uses the fourth image data as the target image data. The present application performs noise removal, image resizing, contrast enhancement, and standardization processing on the pig image data to quickly and accurately complete the preprocessing of the pig image data and obtain the corresponding target image data, which effectively ensures the accuracy of the obtained target image data, and is conducive to improving the accuracy of the subsequent evaluation processing of the target image data by the health assessment model and the weight estimation model.

[0141] In some optional implementations, after step S206, the electronic device may further perform the following steps:

[0142] The total value of the pigs is calculated based on the pig weight prediction data.

[0143] In this embodiment, the corresponding total value of the pigs can be calculated using the pre-recorded number of pigs, the predicted pig weight data estimated by the weight estimation model, and the current market price of the pigs.

[0144] Acquire adjustment demand information corresponding to the farmer.

[0145] In this embodiment, the above adjustment demand information is a specific insurance amount demand proposed by the farmer according to his actual situation. For example, the insurance amount demand may be 1.2 times higher than the total value of the live pigs.

[0146] The total value of the live pigs is adjusted based on the adjustment demand information to obtain a corresponding insurance amount.

[0147] In this embodiment, the product between the adjustment requirement information and the total value of the live pigs can be calculated, and the obtained product can be used as the adjusted insurance amount.

[0148] Based on the target risk level, the insurance amount is calculated using a preset premium calculation strategy to obtain a corresponding target premium.

[0149] In this embodiment, the above-mentioned specific implementation process of calculating and processing the insurance amount based on the target risk level using a preset premium calculation strategy to obtain the corresponding target premium will be further described in detail in subsequent specific embodiments of this application and will not be elaborated on here.

[0150] A corresponding insurance proposal is generated based on the insurance amount and the target premium.

[0151] In this embodiment, the insurance amount and the target premium can be filled into a preset insurance suggestion template to generate a corresponding insurance suggestion. The insurance suggestion template is a pre-constructed template including an insurance amount area and a premium area.

[0152] The insurance proposal is sent to the farmer.

[0153] In this embodiment, a special insurance suggestion page can be designed to present the insurance suggestion in a clear and intuitive manner. For example, the insurance suggestion can be presented in the form of a chart, a list or a comparison table to complete the process of sending the insurance suggestion to the farmer.

[0154] After generating the corresponding target risk level based on the risk score, the present application will automatically and intelligently calculate the total value of the pigs based on the predicted pig weight data, and adjust the total value of the pigs based on the adjustment demand information corresponding to the farmer to obtain the corresponding insurance amount, and then based on the target risk level, use a preset premium calculation strategy to calculate the insurance amount to obtain the corresponding target premium, and subsequently generate corresponding insurance recommendations based on the insurance amount and the target premium and send them to the farmers, so as to provide farmers with personalized insurance recommendations intelligently and accurately, which is conducive to improving the applicability and satisfaction of insurance products, and improving the user experience of farmers.

[0155] In some optional implementations of this embodiment, the method of calculating the insurance amount based on the target risk level and using a preset premium calculation strategy to obtain a corresponding target premium includes the following steps:

[0156] Get the preset rate table.

[0157] In this embodiment, a rate table is pre-built in the insurance company, which is a data table with matching rates based on the insurance amount and risk level. The insurance rate is determined based on multiple factors such as the risk level and the insurance amount. Usually, the insurance company has a rate table that lists the rates corresponding to different risk levels and insurance amount ranges.

[0158] A target rate matching the insurance amount and the target risk level is obtained from the rate table.

[0159] In this embodiment, the above-mentioned rate table can be queried to find out the target rate that matches the insurance amount and the target risk level from the rate table.

[0160] Obtain a target credit rating for the farmer.

[0161] In this embodiment, the target credit rating of the farmer can be found out by obtaining the farmer's user information, wherein the user information may include the name of the farmer, and the credit rating may include excellent, good, qualified, poor, etc.

[0162] A premium discount corresponding to the target credit rating is determined.

[0163] In this embodiment, in addition to the basic rate, possible discounts or surcharges need to be considered. According to the actual insurance business processing requirements, corresponding premium discounts are set in advance for users with different credit ratings. For example, if the farmer has a high credit rating or a good historical claim record, he may receive a certain premium discount. On the contrary, if the farmer has a low credit rating or a poor historical claim record, the premium may increase accordingly.

[0164] Get the preset premium calculation formula.

[0165] In this embodiment, the above premium calculation formula may specifically include: premium = insurance amount × rate × (1-premium discount).

[0166] The insurance amount, the target rate and the premium discount are calculated based on the premium calculation formula to obtain the corresponding target premium.

[0167] In this embodiment, the corresponding target premium can be obtained by substituting the insurance amount, the target rate and the premium discount into the corresponding positions in the above premium calculation formula.

[0168] The present application obtains a preset rate table, and obtains a target rate that matches the insurance amount and the target risk level from the rate table; then obtains the target credit rating of the farmer, and determines the premium discount corresponding to the target credit rating; then obtains a preset premium calculation formula; and subsequently calculates and processes the insurance amount, the target rate, and the premium discount based on the premium calculation formula to obtain the corresponding target premium. The present application obtains a target rate that matches the insurance amount and the target risk level from a preset rate table, and then determines the premium discount corresponding to the target credit rating of the farmer, and then calls the preset premium calculation formula to calculate and process the insurance amount, the target rate, and the premium discount, so as to achieve rapid and accurate generation of the corresponding target premium, improve the calculation efficiency of the target premium, and ensure the data accuracy of the obtained target premium.

[0169] In other embodiments, the risk assessment system can also dynamically adjust the insurance terms based on the obtained risk assessment results to ensure that the insurance products match the aquaculture risks, while improving the risk management capabilities of the insurance company.

[0170] In some optional implementations of this embodiment, before step S203, the electronic device may further perform the following steps:

[0171] Access pre-collected historical pig image data.

[0172] In this embodiment, the above historical pig image data includes a certain number of image data of healthy and sick pigs collected within a historical time period. There is no specific limitation on the time selection of the above historical time period, which can be determined according to actual processing requirements, for example, it can refer to the previous month or the previous six months before the current time, etc.

[0173] The historical pig image data is preprocessed to obtain corresponding sample data.

[0174] In this embodiment, the data preprocessing may include data cleaning, data enhancement, and data annotation. Data cleaning includes removing noise and irrelevant information. Data enhancement includes increasing data diversity through operations such as rotation, flipping, and scaling. Data standardization includes scaling pixel values ​​to the same range, such as [0, 1].

[0175] Call the preset initial neural network model.

[0176] In this embodiment, the model architecture design of the above-mentioned initial neural network model includes at least a convolution layer, an activation function, a pooling layer, a fully connected layer, and an output layer. Among them, the convolution layer: uses the convolution kernel to slide on the input image to extract features. The weight of the convolution kernel is obtained by back propagation learning. Activation function: ReLU, used to introduce nonlinearity. Pooling layer: downsampling, reducing computational complexity while maintaining important features. Fully connected layer: used for classification, mapping the features extracted by the convolution layer to the classification space. Output layer: usually a softmax layer, used to output the probability of each category.

[0177] Determine the target loss function and target optimization algorithm.

[0178] In this embodiment, the above-mentioned target loss function can specifically adopt cross-entropy loss (Cross-Ent ropyLoss). The above-mentioned target optimization algorithm can specifically adopt algorithms such as stochastic gradient descent (SGD), Adam, and Adagrad. Among them, Dropout can also be used to randomly discard some neurons to prevent overfitting. And L1 or L2 regularization can also be further added to the loss function to penalize large weights to avoid overfitting.

[0179] Based on the target loss function and the target optimization algorithm, the initial neural network model is trained using the sample data to obtain a trained specified neural network model.

[0180] In this embodiment, the initial neural network model is trained by using the sample data, and the model weights of the initial neural network model are optimized by using the target loss function and the target optimization algorithm until the model converges, and a trained specified neural network model is obtained.

[0181] The designated neural network model is used as the health assessment model.

[0182] The present application obtains the historical pig image data collected in advance; and performs data preprocessing on the historical pig image data to obtain the corresponding sample data; then calls the preset initial neural network model; then determines the target loss function and the target optimization algorithm; subsequently, based on the target loss function and the target optimization algorithm, uses the sample data to train the initial neural network model to obtain the trained designated neural network model; and finally uses the designated neural network model as the health assessment model. The present application obtains sample data by performing data preprocessing on the historical pig image data collected in advance, and determines the target loss function and the target optimization algorithm; then, based on the target loss function and the target optimization algorithm, uses the sample data to train the preset initial neural network model to obtain the trained designated neural network model and use it as the final health assessment model, thereby automatically and intelligently completing the construction process of the health assessment model, improving the construction efficiency of the health assessment model, and ensuring the model effect of the health assessment model.

[0183] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0184] It should be emphasized that in order to further ensure the privacy and security of the above target risk level, the above target risk level can also be stored in a node of a blockchain.

[0185] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0186] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0187] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0188] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0189] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0190] Further references Figure 3 , as a response to the above Figure 2 The present application provides an embodiment of a risk assessment device based on artificial intelligence. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0191] like Figure 3 As shown, the risk assessment device 300 based on artificial intelligence described in this embodiment includes: a first acquisition module 301, a preprocessing module 302, an assessment module 303, an estimation module 304, a statistics module 305, a first generation module 306 and a return module 307. Among them:

[0192] The first acquisition module 301 is used to acquire pig image data uploaded by farmers;

[0193] A preprocessing module 302 is used to preprocess the pig image data to obtain corresponding target image data;

[0194] An evaluation module 303 is used to evaluate the target image data based on a pre-trained health evaluation model to generate corresponding pig health status prediction data;

[0195] An estimation module 304 is used to estimate the target image data based on a pre-trained weight estimation model to generate corresponding pig weight prediction data;

[0196] A statistical module 305 is used to perform statistical processing on the pig health status prediction data and the pig weight prediction data based on a preset risk assessment model to obtain a corresponding risk score;

[0197] A first generating module 306, configured to generate a corresponding target risk level based on the risk score;

[0198] The return module 307 is used to return the target risk level to the farmer.

[0199] In some optional implementations of this embodiment, the statistics module 305 includes:

[0200] A conversion submodule, used to convert the pig health status prediction data based on a preset conversion strategy to obtain a corresponding health score;

[0201] A first acquisition submodule, used to acquire a first weight corresponding to the health score;

[0202] A second acquisition submodule, used to acquire a second weight corresponding to the pig weight prediction data;

[0203] A first calculation submodule is used to calculate the health score, the pig weight prediction data, the first weight and the second weight based on the risk assessment model using a preset risk score calculation formula to obtain a corresponding calculation result;

[0204] The first determination submodule is configured to use the calculation result as the risk score.

[0205] In some optional implementations of this embodiment, the first generating module 306 includes:

[0206] A third acquisition submodule, used to acquire a first risk score range corresponding to a first preset risk level, acquire a second risk score range corresponding to a second preset risk level, and acquire a third risk score range corresponding to a third preset risk level;

[0207] a first setting submodule, configured to set a target risk level corresponding to the risk score to the first preset risk level if the risk score is within the first risk score range;

[0208] a second setting submodule, configured to set a target risk level corresponding to the risk score to the second preset risk level if the risk score is within the second risk score range;

[0209] The third setting submodule is configured to set the target risk level corresponding to the risk score to the third preset risk level if the risk score is within the third risk score range.

[0210] In some optional implementations of this embodiment, the preprocessing module 302 includes:

[0211] A first processing submodule, configured to remove noise from the pig image data to obtain corresponding first image data;

[0212] A second processing submodule, configured to perform image resizing processing on the first image data to obtain corresponding second image data;

[0213] A third processing submodule, configured to perform contrast enhancement processing on the second image data to obtain corresponding third image data;

[0214] a fourth processing submodule, configured to perform standardization processing on the third image data to obtain corresponding fourth image data;

[0215] The second determining submodule is configured to use the fourth image data as the target image data.

[0216] In some optional implementations of this embodiment, the risk assessment device based on artificial intelligence further includes:

[0217] A first calculation module, used for calculating the total value of the pigs based on the pig weight prediction data;

[0218] A second acquisition module is used to acquire adjustment demand information corresponding to the breeder;

[0219] An adjustment module, used for adjusting the total value of the live pigs based on the adjustment demand information to obtain a corresponding insurance amount;

[0220] A second calculation module is used to calculate the insurance amount based on the target risk level using a preset premium calculation strategy to obtain a corresponding target premium;

[0221] A second generating module, configured to generate corresponding insurance recommendations based on the insurance amount and the target premium;

[0222] A sending module is used to send the insurance proposal to the breeder.

[0223] In some optional implementations of this embodiment, the second calculation module includes:

[0224] The third acquisition submodule is used to obtain a preset rate table;

[0225] A fourth acquisition submodule is used to acquire a target rate matching the insurance amount and the target risk level from the rate table;

[0226] A fifth acquisition submodule is used to obtain the target credit rating of the farmer;

[0227] A third determination submodule is used to determine a premium discount corresponding to the target credit rating;

[0228] The sixth acquisition submodule is used to obtain a preset premium calculation formula;

[0229] The second calculation submodule is used to calculate the insurance amount, the target rate and the premium discount based on the premium calculation formula to obtain the corresponding target premium.

[0230] In some optional implementations of this embodiment, the risk assessment device based on artificial intelligence further includes:

[0231] The third acquisition module is used to acquire the historical pig image data collected in advance;

[0232] A processing module, used for performing data preprocessing on the historical pig image data to obtain corresponding sample data;

[0233] A calling module, used to call a preset initial neural network model; wherein the initial neural network model includes at least a convolutional layer, a pooling layer, a fully connected layer and an output layer;

[0234] A first determination module is used to determine a target loss function and a target optimization algorithm;

[0235] A training module, used to train the initial neural network model using the sample data based on the target loss function and the target optimization algorithm to obtain a trained specified neural network model;

[0236] The second confirmation module is used to use the specified neural network model as the health assessment model.

[0237] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4This is a basic structural block diagram of the computer device in this embodiment.

[0238] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 4 with components 41-43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0239] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0240] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the computer device 4. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions of a risk assessment method based on artificial intelligence, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0241] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run computer-readable instructions stored in the memory 41 or process data, such as computer-readable instructions for running the artificial intelligence-based risk assessment method.

[0242] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0243] The present application also provides another embodiment, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the above-mentioned artificial intelligence-based risk assessment method.

[0244] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0245] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A risk assessment method based on artificial intelligence, characterized in that: The steps include: Obtain pig image data uploaded by farmers; Preprocessing the pig image data to obtain corresponding target image data; The target image data is evaluated and processed based on a pre-trained health assessment model to generate corresponding pig health status prediction data; The target image data is estimated based on a pre-trained weight estimation model to generate corresponding pig weight prediction data; Performing statistical processing on the predicted data of the health status of the pigs and the predicted data of the weight of the pigs based on a preset risk assessment model to obtain corresponding risk scores; generating a corresponding target risk level based on the risk score; The target risk level is returned to the farmer.

2. The artificial intelligence-based risk assessment method according to claim 1, characterized in that: The step of performing statistical processing on the predicted data of the health status of the pigs and the predicted data of the weight of the pigs based on the preset risk assessment model to obtain corresponding risk scores specifically includes: The pig health status prediction data is converted based on a preset conversion strategy to obtain a corresponding health score; Obtaining a first weight corresponding to the health score; Obtaining a second weight corresponding to the pig weight prediction data; Based on the risk assessment model, the health score, the pig weight prediction data, the first weight and the second weight are calculated and processed using a preset risk score calculation formula to obtain a corresponding calculation result; The calculation result is used as the risk score.

3. The artificial intelligence-based risk assessment method according to claim 1, characterized in that: The step of generating a corresponding target risk level based on the risk score specifically includes: Obtaining a first risk score range corresponding to a first preset risk level, obtaining a second risk score range corresponding to a second preset risk level, and obtaining a third risk score range corresponding to a third preset risk level; If the risk score is within the first risk score range, setting the target risk level corresponding to the risk score to the first preset risk level; If the risk score is within the second risk score range, setting the target risk level corresponding to the risk score to the second preset risk level; If the risk score is within the third risk score range, the target risk level corresponding to the risk score is set to the third preset risk level.

4. The artificial intelligence-based risk assessment method according to claim 1, characterized in that: The step of preprocessing the pig image data to obtain corresponding target image data specifically includes: Performing noise removal processing on the pig image data to obtain corresponding first image data; Performing image resizing processing on the first image data to obtain corresponding second image data; performing contrast enhancement processing on the second image data to obtain corresponding third image data; performing standardization processing on the third image data to obtain corresponding fourth image data; The fourth image data is used as the target image data.

5. The artificial intelligence-based risk assessment method according to claim 1, characterized in that: After the step of generating a corresponding target risk level based on the risk score, the method further includes: Calculate the total value of pigs based on the predicted pig weight data; Obtaining adjustment demand information corresponding to the breeder; Adjust the total value of the live pigs based on the adjustment demand information to obtain a corresponding insurance amount; Based on the target risk level, the insurance amount is calculated using a preset premium calculation strategy to obtain a corresponding target premium; generating a corresponding insurance proposal based on the insurance amount and the target premium; The insurance proposal is sent to the farmer.

6. The artificial intelligence-based risk assessment method according to claim 5, characterized in that: The step of calculating the insurance amount based on the target risk level using a preset premium calculation strategy to obtain a corresponding target premium specifically includes: Get the preset rate table; Obtaining a target rate that matches the insurance amount and the target risk level from the rate table; Obtaining a target credit rating for the farmer; determining a premium discount corresponding to the target credit rating; Get the preset premium calculation formula; The insurance amount, the target rate and the premium discount are calculated based on the premium calculation formula to obtain the corresponding target premium.

7. The artificial intelligence-based risk assessment method according to claim 1, characterized in that: Before the step of evaluating the target image data based on the pre-trained health assessment model to generate corresponding pig health status prediction data, the method further includes: Obtain pre-collected historical pig image data; Performing data preprocessing on the historical pig image data to obtain corresponding sample data; Calling a preset initial neural network model; wherein the initial neural network model includes at least a convolutional layer, a pooling layer, a fully connected layer, and an output layer; Determine the target loss function and target optimization algorithm; Based on the target loss function and the target optimization algorithm, the initial neural network model is trained using the sample data to obtain a trained specified neural network model; The designated neural network model is used as the health assessment model.

8. A risk assessment device based on artificial intelligence, characterized in that: include: The first acquisition module is used to acquire pig image data uploaded by farmers; A preprocessing module, used for preprocessing the pig image data to obtain corresponding target image data; An evaluation module, used to evaluate and process the target image data based on a pre-trained health evaluation model to generate corresponding pig health status prediction data; An estimation module, used to estimate the target image data based on a pre-trained weight estimation model to generate corresponding pig weight prediction data; A statistical module, used for performing statistical processing on the predicted data of the health status of the pigs and the predicted data of the weight of the pigs based on a preset risk assessment model to obtain corresponding risk scores; A first generating module, configured to generate a corresponding target risk level based on the risk score; A return module is used to return the target risk level to the farmer.

9. A computer device, characterized in that: It comprises a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the risk assessment method based on artificial intelligence as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the artificial intelligence-based risk assessment method according to any one of claims 1 to 7.

Images