An intelligent agricultural management system based on big data

Abstract

The application discloses an intelligent agricultural management system based on big data, and a running method of the system, which comprises the following steps: step one, collecting and labeling crop disease images through a network source; step two, establishing a crop disease identification model by using a convolutional neural network; step three, uploading plant disease images and performing feature description by a user in the system; step four, identifying the crop diseases of the images uploaded by the user; and step five, estimating the probability of each type of pest and generating feedback information for the user, wherein the data acquisition and preprocessing module is used for collecting, describing and preparing training data; the crop disease identification model establishment module is used for training and optimizing the identification model; and the user interface and disease identification module is used for user verification, image uploading, feature description, real-time disease identification and feedback; the application has the characteristics of accurate, efficient and objective crop disease diagnosis.

Description

Technical Field

[0001] This invention relates to the field of big data technology, specifically to an intelligent agricultural management system based on big data. Background Technology

[0002] Crop diseases have always been a significant challenge in agricultural production. Traditional disease identification methods typically rely on manual observation and expert experience, which is not only time-consuming and labor-intensive but also susceptible to subjective factors, leading to inconsistent diagnostic results. Therefore, a faster, more accurate, and objective method is needed to address this problem.

[0003] With the continuous development of big data and computer vision technologies, image recognition and deep learning have become effective approaches to solving the problem of crop disease identification. Some existing systems have begun to utilize deep learning techniques, such as convolutional neural networks (CNNs), for automatic identification of crop disease images. However, these systems typically focus only on the image itself, neglecting other important information, such as user-provided feature descriptions and environmental factors. Furthermore, many systems are limited to specific crops and disease types, lacking versatility and scalability. Therefore, it is essential to design an intelligent agricultural management system based on big data that provides accurate, efficient, and objective crop disease diagnosis. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent agricultural management system based on big data to solve the problems mentioned in the background art.

[0005] To address the aforementioned technical problems, this invention provides the following technical solution: a big data-based intelligent agricultural management system, the operation of which includes the following steps:

[0006] Step 1: Collect and label images of crop diseases from online sources;

[0007] Step 2: Establish a crop disease identification model using a convolutional neural network;

[0008] Step 3: Users upload images of plant diseases and provide feature descriptions within the system;

[0009] Step 4: Identify crop diseases in the images uploaded by users;

[0010] Step 5: The system estimates the probability of each pest type and generates feedback information for the user.

[0011] According to the above technical solution, the step of collecting and labeling crop disease images from network sources includes:

[0012] Establish a comprehensive crop disease database, including diverse images, location information, meteorological data, and pest characteristic information;

[0013] Data labeling and detailed description, including image capture time and location, related plant types, pest species and severity;

[0014] Data preprocessing, such as image denoising, data alignment, and format standardization, is used to ensure the quality and consistency of training data.

[0015] According to the above technical solution, the step of establishing a comprehensive crop disease database, including diverse images, location information, meteorological data, and pest characteristic information, includes:

[0016] Collect images of crop diseases from the internet, covering different crop types, different disease development stages, different lighting conditions or shooting angles, and collect meteorological data related to the location and time of each image capture, such as temperature, humidity, wind speed and precipitation. In addition, it is also necessary to obtain characteristic data of different pests, such as the life cycle of pests, degree of damage, and host plant information, in order to establish a rich and comprehensive pest database.

[0017] According to the above technical solution, the step of establishing a crop disease identification model using a convolutional neural network includes:

[0018] The data was divided into 70% training set, 15% validation set, and 15% test set. A convolutional neural network architecture was then chosen for model building, specifically the ResNet-50 model with pre-trained weights to accelerate training and improve model performance. Additionally, a fully connected layer with output neurons matching the number of disease types was added for multi-class classification. Simultaneously, data augmentation techniques were applied to increase the diversity of the training data, including random rotation, horizontal and vertical flipping, random scaling, and random cropping, thereby better generalizing the model to different image variations. Model training then proceeded using a stochastic gradient descent optimizer with learning rate decay to minimize cross-entropy loss. The initial learning rate was set to 0.001, and the batch size was 32. The training process involved forward propagation, loss calculation, backpropagation, and updating of model weights to continuously reduce error. An early stopping strategy was implemented: if the validation loss did not improve within five consecutive epochs, training was halted. The system will then pause and use validation set data to evaluate model performance, measuring accuracy, recall, F1 score, and confusion matrix performance metrics. Performance on the validation set determines whether the system needs further model adjustments or hyperparameter changes. If performance is unsatisfactory, the system will perform hyperparameter tuning, including adjusting the learning rate, selecting the optimizer, and fine-tuning the model architecture. Furthermore, the system uses learning rate decay, with an initial learning rate set to 0.001, reduced to half its original value every 10 epochs. To mitigate overfitting, a Dropout layer is used, with a dropout rate set to 0.5 to randomly shut down half of the neurons during training. Once the model is trained, it is deployed on a cloud server for real-time disease identification. Users can use a mobile application to capture images and upload them to obtain disease identification results. Developers regularly monitor model performance and retrain as needed to adapt to new data and disease types, ensuring the model maintains high performance in constantly changing environments.

[0019] According to the above technical solution, the step of the user uploading plant disease images and performing feature descriptions within the system includes:

[0020] Users need to provide a username and password or create an account through other verification methods. After successful login, the application will actively obtain the device's location information. Subsequently, users can use the application to upload images of plant diseases. Image processing functions will be provided to compress and convert formats before uploading to reduce the data transmission size. After the user uploads the image, the application will provide a feature description interface where users can provide detailed feature descriptions of the plant disease, such as the appearance of the symptoms, the affected parts, and the possible severity of the disease. These descriptions will supplement the image information and help the system more accurately identify the problem. The application will communicate with the crop disease identification system in real time to obtain immediate identification feedback. Through this step, users are provided with a convenient tool to upload plant disease images, provide feature descriptions, and obtain timely disease identification feedback. Such a user interface and application will help farmers better manage their crops and take timely measures to deal with plant disease problems.

[0021] According to the above technical solution, the step of identifying crop diseases in user-uploaded images includes:

[0022] The system extracts keywords added by the user in the feature information. This information includes the user's description of disease characteristics, such as the appearance of leaves, affected parts, and possible severity of the disease. Then, using natural language processing technology, the system matches the extracted keywords. After matching, the system integrates the image recognition results with the feature information provided by the user. The system also obtains the geographic information of the user's current location. Subsequently, the system connects to a meteorological data source to obtain meteorological data of the user's current location, including temperature, humidity, and precipitation. This data will be used to analyze the current environmental conditions to predict the types of pests that may occur. By integrating the image recognition results, the feature information provided by the user, the location information, and the meteorological data, the system will predict the types of pests that may occur. This step provides more comprehensive information to help users understand potential disease risks.

[0023] According to the above technical solution, the steps of estimating the probability of each pest type and generating feedback information for the user include:

[0024] Consider the prior probability of pest types;

[0025] Calculate conditional probabilities, including image features, keywords, temperature, and humidity data;

[0026] Calculate the posterior probability of each pest type using Bayes' theorem;

[0027] Select the pest type with the highest posterior probability as the prediction result;

[0028] Provide user feedback to help them understand potential pest problems and take appropriate measures.

[0029] According to the above technical solution, the step of selecting the pest type with the highest posterior probability as the prediction result includes:

[0030] Consider the prior probability of pest types (P(p(pest))), which represents the probability of each pest type occurring in the absence of other information. This can be estimated based on the pest types prevalent in the current season. Next, consider the conditional probabilities, which include image features, keywords, temperature, and humidity data. Then, calculate the conditional probabilities P(image features|pest), P(keywords|pest), P(temperature|pest), and P(humidity|pest). These conditional probabilities represent the probability that relevant information matches a given pest type. Finally, consider the marginal probabilities of features, namely P(image features), P(keywords), P(temperature), and P(humidity), which are the probabilities of occurrence of their respective features, regardless of the pest type. Using these probability values, the system can use Bayes' theorem to calculate the posterior probability of each pest type (P(pest|image features, keywords, temperature, humidity)). The pest type with the highest posterior probability will be considered the most likely pest type.

[0031] According to the above technical solution, the system includes:

[0032] The data acquisition and preprocessing module is used to collect, describe, and prepare training data;

[0033] A module for establishing crop disease identification models is used to train and optimize the identification models.

[0034] The user interface and disease identification module are used for user verification, image uploading, feature description, and real-time disease identification and feedback.

[0035] According to the above technical solution, the data acquisition and preprocessing module includes:

[0036] The data acquisition module is used to collect images of crop diseases from the network, including images of various crop types, different disease stages, and different conditions, as well as relevant meteorological data and pest characteristic data;

[0037] The data labeling and description module is used to label and describe the collected images and data in detail, including shooting information, plant type, and pest species;

[0038] The data preprocessing module is used to preprocess the collected data, including image denoising, data alignment, and format unification, to ensure the quality and consistency of the training data.

[0039] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: First, the system collects crop disease images from the internet and establishes a comprehensive disease database. These images cover various crop types and pest conditions, ensuring data diversity. Then, the data is labeled, cleaned, and preprocessed to ensure high quality and consistency. Subsequently, a crop disease identification model is established using a convolutional neural network, trained using a ResNet-50 model and data augmentation techniques, considering hyperparameter tuning and performance evaluation. Next, a user-friendly mobile application is provided, allowing users to upload plant disease images and provide feature descriptions, while simultaneously acquiring location information. In step four, the trained model is used to identify the user-uploaded images, extract feature information, and estimate the probability of pest types, considering image recognition results, keyword matching, and meteorological data. Finally, feedback information is generated to help users understand potential pest risks, providing them with disease identification and suggestions. This method is characterized by accurate, efficient, and objective crop disease diagnosis. Attached Figure Description

[0040] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0041] Figure 1 A flowchart illustrating the intelligent agricultural management method based on big data provided in Embodiment 1 of the present invention;

[0042] Figure 2 This is a schematic diagram of the module composition of the intelligent agricultural management system based on big data provided in Embodiment 2 of the present invention. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] Example 1: Figure 1 This is a flowchart of a big data-based intelligent agricultural management method provided in Embodiment 1 of the present invention. This embodiment can be applied to the scenario of crop disease identification. This method can be executed by the big data-based intelligent agricultural management system provided in this embodiment, such as... Figure 1 As shown, the method specifically includes the following steps:

[0045] Step 1: Collect and label images of crop diseases from online sources;

[0046] In this embodiment of the invention, a comprehensive crop disease database is established to support subsequent big data training and accurate crop disease identification.

[0047] For example, images of crop diseases collected from the internet can cover different crop types, different disease development stages, different lighting conditions or shooting angles. Meteorological data of pests related to the location and time of each image shooting can also be collected, such as temperature, humidity, wind speed and precipitation. In addition, it is also necessary to obtain characteristic data of different pests, such as the life cycle of pests, the degree of damage, host plants and other detailed information, in order to establish a rich and comprehensive pest database.

[0048] For example, each piece of collected data needs to be labeled and described in detail. For crop disease images collected from the internet, a detailed description is needed, including the time and location of the image capture, the type of plant involved, the type of pest, and the severity of the pest. The collected data is then uploaded to a cloud storage platform, and an index and metadata are created for the data, with access control set. After the data is collected, it is cleaned and preprocessed, such as image denoising, data alignment, and data format standardization, to ensure the quality and consistency of the training data. Through this step, a crop disease dataset can be established, providing a solid foundation for subsequent big data training and accurate crop disease identification.

[0049] Step 2: Establish a crop disease identification model using a convolutional neural network;

[0050] In this embodiment of the invention, a crop disease identification model is established by using a convolutional neural network to train the collected data.

[0051] For example, the data is divided into a 70% training set, a 15% validation set, and a 15% test set. Then, a convolutional neural network architecture is selected for model building. Specifically, a ResNet-50 model is used with pre-trained weight parameters to accelerate training and improve model performance. In addition, a fully connected layer with output neurons matching the number of disease types is added for multi-class classification. At the same time, data augmentation techniques are applied to increase the diversity of training data, including random rotation, horizontal and vertical flipping, random scaling, and random cropping, so that the model can better generalize to different image variations. Then, the model is trained using a stochastic gradient descent optimizer with learning rate decay to minimize cross-entropy loss. The initial learning rate is set to 0.001, and the batch size is 32. During training, the data is forward propagated, the loss is calculated, backpropagation is performed, and the model weights are updated to continuously reduce the error. An early stopping strategy is set, and training stops if the validation loss does not improve in 5 consecutive rounds.

[0052] For example, validation set data is used to evaluate the model's performance, measuring performance metrics such as accuracy, recall, F1 score, and confusion matrix. Performance on the validation set determines whether the system needs further model adjustments or hyperparameter changes. If performance is unsatisfactory, the system will perform hyperparameter tuning, including adjusting the learning rate, selecting the optimizer, and fine-tuning the model architecture. In addition, the system uses learning rate decay, with the initial learning rate set to 0.001 and reduced to half its original value every 10 epochs. To mitigate overfitting, a Dropout layer is also used, with a dropout rate set to 0.5 to randomly shut down half of the neurons during training. Once the model is trained, it is deployed on a cloud server for real-time disease identification. Users can use a mobile application to take images and upload them to obtain disease identification results. Developers regularly monitor model performance and retrain the model as needed to adapt to new data and disease types, ensuring that the model maintains high performance in a constantly changing environment.

[0053] Step 3: Users upload images of plant diseases and provide feature descriptions within the system;

[0054] In this embodiment of the invention, through the user interface, users can easily upload images of plant diseases, provide feature descriptions, and obtain immediate recognition feedback;

[0055] For example, users need to provide a username and password or create an account through other verification methods. After successful login, the application will actively obtain the device's location information. Subsequently, users can use the application to upload images of plant diseases. Image processing functions will be provided to compress and convert formats before uploading to reduce the data transmission size. After the user uploads the image, the application will provide a feature description interface where users can provide detailed feature descriptions of the plant disease, such as the appearance of the symptoms, the affected parts, and the possible severity of the disease. These descriptions will supplement the image information and help the system more accurately identify the problem. The application will communicate with the crop disease identification system in real time to obtain immediate identification feedback. Through this step, users are provided with a convenient tool to upload plant disease images, provide feature descriptions, and obtain timely disease identification feedback. Such a user interface and application will help farmers better manage their crops and take timely measures to deal with plant disease problems.

[0056] Step 4: Identify crop diseases in the images uploaded by users;

[0057] In this embodiment of the invention, the system uses a pre-trained crop disease identification model to analyze plant disease images uploaded by users, and also receives the user's location information. The model will identify the disease type in the image and provide initial disease identification results, laying the foundation for subsequent steps.

[0058] For example, the system extracts keywords added by the user in the feature information. This information includes the user's description of disease characteristics, such as the appearance of the leaves, the affected parts, and the possible severity of the disease. Then, using natural language processing technology, the system matches the extracted keywords. After matching, the system integrates the image recognition results and the feature information provided by the user. The system also obtains the geographic information of the user's current location. Subsequently, the system connects to a meteorological data source to obtain meteorological data of the user's current location, including temperature, humidity, precipitation, etc. This data will be used to analyze the current environmental conditions to predict the types of pests that may occur. By integrating the image recognition results, the feature information provided by the user, the location information, and the meteorological data, the system will predict the types of pests that may occur. This step will provide more comprehensive information to help the user understand the potential disease risks.

[0059] Step 5: The system estimates the probability of each pest type and generates feedback information for the user.

[0060] In this embodiment of the invention, by combining image recognition results, keyword matching, and meteorological data, the system can estimate possible pest types.

[0061] For example, first, consider the prior probability of pest type (P(p(pest))), which represents the probability of each pest type occurring in the absence of other information. This can be estimated based on the pest types prevalent in the current season. Next, consider the conditional probabilities, which include image features, keywords, temperature, and humidity data. Then, calculate the conditional probabilities P(image features|pest), P(keywords|pest), P(temperature|pest), and P(humidity|pest). These conditional probabilities represent the probability that relevant information matches a given pest type. Finally, consider the marginal probabilities of features, namely P(image features), P(keywords), P(temperature), and P(humidity), which are the probabilities of occurrence of their respective features, regardless of the pest type. Using these probability values, the system can use Bayes' theorem to calculate the posterior probability of each pest type (P(pest|image features, keywords, temperature, humidity)). The pest type with the highest posterior probability will be considered the most likely pest type.

[0062] For example, suppose there are three types of pests: A, B, and C, and their corresponding conditional and prior probabilities:

[0063] 1. Prior probability:

[0064] P(A) = 0.4, P(B) = 0.3, P(C) = 0.3;

[0065] 2. Conditional probability (P(image features|pest)):

[0066] P(image feature|A) = 0.8, P(image feature|B) = 0.2, P(image feature|C) = 0.1;

[0067] 3. Conditional probability (P(keyword|pest)):

[0068] P(keyword|A) = 0.6, P(keyword|B) = 0.2, P(keyword|C) = 0.1;

[0069] 4. Conditional probabilities (P(temperature|pest)) and P(humidity|pest)):

[0070] P(temperature|A) = 0.7, P(humidity|A) = 0.4;

[0071] P(temperature|B) = 0.9, P(humidity|B) = 0.3;

[0072] P(temperature|C) = 0.6, P(humidity|C) = 0.2;

[0073] 5. Calculate the posterior probability (P(p(insect pest | image features, keywords, temperature, humidity)) using Bayes' theorem:

[0074] P(A|data) = P(A) * P(image features|A) * P(keywords|A) * P(temperature|A) * P(humidity|A) / (P(data))

[0075] P(B|data) = P(B) * P(image features|B) * P(keywords|B) * P(temperature|B) * P(humidity|B) / (P(data))

[0076] P(C|data) = P(C) * P(image features|C) * P(keywords|C) * P(temperature|C) * P(humidity|C) / (P(data))

[0077] Where P(data) is the normalization factor, ensuring that the sum of all posterior probabilities equals 1.

[0078] Finally, by comparing the posterior probabilities, the pest type with the highest posterior probability is selected as the system's prediction result. This prediction result will be fed back to the user to help them better understand potential pest problems and take corresponding agricultural measures.

[0079] Example 2: Example 2 of the present invention provides an intelligent agricultural management system based on big data. Figure 2This is a schematic diagram of the module composition of the big data-based intelligent agricultural management system provided in Embodiment 2 of the present invention, as shown below. Figure 2 As shown, the system includes:

[0080] The data acquisition and preprocessing module is used to collect, describe, and prepare training data;

[0081] A module for establishing crop disease identification models is used to train and optimize the identification models.

[0082] The user interface and disease identification module are used for user verification, image uploading, feature description, and real-time disease identification and feedback.

[0083] In some embodiments of the present invention, the data acquisition and preprocessing module includes:

[0084] The data acquisition module is used to collect images of crop diseases from the network, including images of various crop types, different disease stages, and different conditions, as well as relevant meteorological data and pest characteristic data;

[0085] The data labeling and description module is used to label and describe the collected images and data in detail, including shooting information, plant type, pest type, etc.

[0086] The data preprocessing module is used to preprocess the collected data, including image denoising, data alignment, and format unification, to ensure the quality and consistency of the training data;

[0087] In some embodiments of the present invention, the crop disease identification model establishment module includes:

[0088] The data segmentation and model selection module is used to divide the data into training, validation and test sets, and select a convolutional neural network architecture (ResNet-50) for model building;

[0089] The model training and optimization module is used to train the model using the stochastic gradient descent optimizer, learning rate decay, batch size, and other settings.

[0090] The model deployment module is used to deploy the trained model on a cloud server to provide users with real-time disease identification services.

[0091] In some embodiments of the present invention, the user interface and the disease identification module include:

[0092] The user verification and data upload module is used by users to create accounts or verify their identities through the application and upload images of plant diseases they have taken.

[0093] The feature description module provides a user interface that allows users to provide detailed feature descriptions of plant diseases, such as symptoms, affected parts, and disease severity.

[0094] The image recognition and feedback module is used to identify crop diseases in user-uploaded images using a trained model.

[0095] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0096] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A smart agricultural management method based on big data, comprising the following steps: Step 1: Collect and label images of crop diseases from online sources; Step 2: Establish a crop disease identification model using a convolutional neural network; Step 3: Users upload images of plant diseases and provide feature descriptions within the system; Step 4: Identify crop diseases in the images uploaded by users; Step 5: The system estimates the probability of each pest type and generates feedback information for the user; The steps of collecting and labeling crop disease images from network sources include: Establish a comprehensive crop disease database, including diverse images, location information, meteorological data, and pest characteristic information; Data labeling and detailed description, including image capture time and location, related plant types, pest species and severity; Data preprocessing: image denoising, data alignment, and format standardization to ensure the quality and consistency of training data; The steps for establishing a crop disease identification model using a convolutional neural network include: The data was divided into 70% training set, 15% validation set, and 15% test set. A convolutional neural network architecture was then chosen for model building, specifically the ResNet-50 model with pre-trained weights to accelerate training and improve model performance. Additionally, a fully connected layer with output neurons matching the number of disease types was added for multi-class classification. Simultaneously, data augmentation techniques were applied to increase the diversity of the training data, including random rotation, horizontal and vertical flipping, random scaling, and random cropping, thereby better generalizing the model to different image variations. Model training then proceeded using a stochastic gradient descent optimizer with learning rate decay to minimize cross-entropy loss. The initial learning rate was set to 0.001, and the batch size was 32. The training process involved forward propagation, loss calculation, backpropagation, and updating of model weights to continuously reduce error. An early stopping strategy was implemented: if the validation loss did not improve within five consecutive epochs, training was halted. The system will stop and use validation set data to evaluate the model's performance, measuring accuracy, recall, F1 score, and confusion matrix performance metrics. Performance on the validation set determines whether the system needs further model adjustments or hyperparameter changes. If performance is unsatisfactory, the system will perform hyperparameter tuning, including adjusting the learning rate, selecting the optimizer, and fine-tuning the model architecture. In addition, the system uses learning rate decay, with the initial learning rate set to 0.001 and reduced to half its original value every 10 rounds. To mitigate overfitting, a Dropout layer is also used, with a dropout rate set to 0.5 to randomly shut down half of the neurons during training. Once the model is trained, it will be deployed on a cloud server for real-time disease identification. Users can use a mobile application to take pictures and upload them to obtain disease identification results. Developers regularly monitor model performance and retrain the model as needed to adapt to new data and disease types, ensuring that the model maintains high performance in a constantly changing environment. The steps for identifying crop diseases in user-uploaded images include: The system extracts keywords added by the user in the feature information. Then, using natural language processing technology, the system matches the extracted keywords. After matching, the system integrates the image recognition results and the feature information provided by the user. The system also obtains the geographic information of the user's current location. Subsequently, the system connects to the meteorological data source to obtain the meteorological data of the user's current location, including temperature, humidity, and precipitation. This data will be used to analyze the current environmental conditions to predict the types of pests that may occur. By integrating the image recognition results, the feature information provided by the user, the location information, and the meteorological data, the system will predict the types of pests that may occur. The steps of estimating the probability of each pest type and generating feedback information for the user include: Consider the prior probability of pest types; Calculate conditional probabilities, including image features, keywords, temperature, and humidity data; Calculate the posterior probability of each pest type using Bayes' theorem; Select the pest type with the highest posterior probability as the prediction result; Provide user feedback to help them understand potential pest problems and take appropriate measures; The step of selecting the pest type with the highest posterior probability as the prediction result includes: Consider the prior probability P(pest), representing the probability of each pest type occurring in the absence of other information. This can be estimated based on the prevalent pest types of the current season. Next, consider the conditional probabilities, which include image features, keywords, temperature, and humidity data. Then, calculate the conditional probabilities P(image features|pest), P(keywords|pest), P(temperature|pest), and P(humidity|pest). These conditional probabilities represent the probability that relevant information matches a given pest type. Finally, consider the marginal probabilities of features, namely P(image features), P(keywords), P(temperature), and P(humidity), which are the probabilities of occurrence of their respective features, regardless of the pest type. Using these probability values, the system can use Bayes' theorem to calculate the posterior probability P(pest|image features, keywords, temperature, humidity) for each pest type. The pest type with the highest posterior probability will be considered the most likely pest type.

2. A smart agricultural management system based on big data, characterized by: The intelligent agricultural management system executes the big data-based intelligent agricultural management method as described in claim 1, and the system includes: The data acquisition and preprocessing module is used to collect, describe, and prepare training data; A module for establishing crop disease identification models is used to train and optimize the identification models. The user interface and disease identification module are used for user verification, image uploading, feature description, and real-time disease identification and feedback.

3. The intelligent agricultural management system based on big data according to claim 2, characterized in that: The data acquisition and preprocessing module includes: The data acquisition module is used to collect images of crop diseases from the network, including images of various crop types, different disease stages, and different conditions, as well as relevant meteorological data and pest characteristic data; The data labeling and description module is used to label and describe the collected images and data in detail, including shooting information, plant type, and pest species; The data preprocessing module is used to preprocess the collected data, including image denoising, data alignment, and format unification, to ensure the quality and consistency of the training data.

Images