Intelligent pest and disease damage prediction, prevention and control method based on crop growth state

Abstract

The invention discloses a crop growth state-based intelligent pest and disease damage prediction, prevention and control method, which comprises the following steps of: acquiring multi-source observation data related to a crop growth state and a crop growth environment in a crop growth process, and preprocessing the data to obtain multi-source heterogeneous data of crops; inputting the multi-source heterogeneous data into a pest and disease damage prediction model to predict the occurrence probability of the pest and disease damage in the current growth state and the growth environment, outputting a pest and disease damage prediction result according with the current crop growth state, and performing prevention and control analysis on the pest and disease damage prediction result through a pest and disease damage prevention expert knowledge base. And generating corresponding prevention and control decisions, sending the decisions to corresponding management personnel terminals for disease and pest prevention and control early warning, monitoring disease and pest prevention and control effects of the prevention and control decisions, feeding back the disease and pest prevention and control effects to the disease and pest prediction model for model optimization, and adaptively updating the disease and pest prediction model. The effects of improving the yield and quality of crops and effectively preventing and controlling diseases and pests in the whole growth period of the crops are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pest control, and in particular to an intelligent pest prediction and control method based on crop growth state. BACKGROUND

[0002] At present, with the acceleration of agricultural modernization in China, agricultural pest problems are increasingly prominent, which has brought huge losses to agricultural production. The occurrence of pests not only affects the yield and quality of crops, but also causes serious damage to the ecological environment. Therefore, under the background of the rapid development of artificial intelligence technology, combining artificial intelligence technology to effectively predict and control agricultural pest problems has become one of the important directions of intelligent agricultural transformation and sustainable development.

[0003] The existing agricultural pest control method usually relies on experienced farmers to visually observe the crop growth state and make a subjective judgment of pests based on personal experience. The accuracy of the judgment result depends on personal experience, and the accuracy of pest determination depends on the artificial experience of the determinator, which is likely to lead to a lag in pest discovery and a lag in pest control measures following the growth rate of crops, affecting crop yield and quality. The accuracy of crop pest determination depends on artificial experience, and the lag of control measures lags behind the growth rate of crops. SUMMARY

[0004] In view of the problem that the accuracy of crop pest determination in the prior art depends on artificial experience and the lag of control measures lags behind the growth rate of crops, the present application provides an intelligent pest prediction and control method based on crop growth state, which can introduce machine learning, artificial intelligence and other means to monitor the crop growth state in real time, accurately determine the crop pests at different growth stages, and prevent and control the crop pests at each growth stage, thereby improving the yield and quality of crops and effectively preventing and controlling pests throughout the growth cycle of crops.

[0005] In the first aspect, the above application aims to achieve the following technical solutions: An intelligent pest prediction and control method based on crop growth state, the method comprising: Obtaining crop growth state and crop growth environment related multi-source observation data in the crop growth process, and performing data preprocessing on the multi-source observation data to obtain multi-source heterogeneous data of crops; Inputting the multi-source heterogeneous data into a pre-trained pest prediction model to predict the occurrence probability of pests under the current growth state and growth environment, and outputting a pest prediction result conforming to the current crop growth state; The current pest and disease prediction result is analyzed for prevention and control by the pest and disease control expert knowledge base, and corresponding prevention and control decisions are generated and sent to corresponding management personnel terminals for pest and disease prevention and control early warning. The pest and disease control effect of the prevention and control decision is monitored, and the pest and disease control effect is fed back to the pest and disease prediction model for model optimization, and the pest and disease prediction model is adaptively updated.

[0006] In a preferred example, the present application can be further configured as follows: the training process of the pest and disease prediction model specifically includes: Obtain crop growth state data and growth environment data of the entire growth cycle of the crop, set control groups by SVM algorithm with growth state and growth environment as variables, and obtain pest and disease data of each control group; Through a deep learning algorithm, the pest and disease data and corresponding control group parameters are data trained, the relationship between crop growth state, growth environment and pest and disease occurrence probability is analyzed, and a pest and disease prediction model is constructed.

[0007] In a preferred example, the present application can be further configured as follows: the pest and disease prediction model is constructed by a deep learning algorithm, including: Obtain pest and disease characteristics of the pest and disease data through a convolutional neural network, obtain sequence data of the pest and disease data through a recurrent neural network, and obtain a pest and disease correlation by associating the sequence data and the corresponding pest and disease characteristics; According to the pest and disease correlation, the relationship between crop growth state, growth environment and pest and disease occurrence probability is analyzed according to the crop growth cycle, and a pest and disease prediction model is constructed.

[0008] In a preferred example, the present application can be further configured as follows: the training process of the pest and disease prediction model further includes: Test the pest and disease prediction model with a validation data set containing crop growth state data and growth environment data that does not participate in training, and obtain a test result of the validation data set; Artificially verify the validation data set, determine the test result by the artificial verification result, and evaluate the prediction accuracy of the pest and disease prediction model.

[0009] In a preferred example, the present application can be further configured as follows: the artificial verification of the validation data set, the determination of the test result by the artificial verification result, and the evaluation of the prediction accuracy of the pest and disease prediction model specifically include: The error rate, recall rate and F1 score between the manual verification result and the test result are calculated respectively, the model prediction accuracy is comprehensively evaluated according to the error rate, the recall rate and the F1 score, and a model prediction accuracy evaluation result is obtained.

[0010] In a preferred example, the application can be further configured to: the pest control expert knowledge base is used for pest prediction result analysis, and corresponding control decision is generated and sent to the corresponding management personnel end for pest control warning, specifically including: Based on the historical pest data of crops and the corresponding artificial control strategy, a pest control expert knowledge base is constructed; The current pest prediction result is compared and analyzed with the pest control expert knowledge base, and the control strategy conforming to the current pest trend and crop growth state is selected according to the comparison result, and the corresponding control decision is generated; The control decision is sent to the corresponding management personnel end for pest control warning, and the execution state of the control decision is listened to, and pest control effect feedback information is generated.

[0011] In a preferred example, the application can be further configured to: the crop growth state and crop growth environment related multi-source observation data in the crop growth process are obtained, and the multi-source observation data is preprocessed to obtain the data preprocessing process in the multi-source heterogeneous data of crops, specifically including: The multi-source observation data is classified into structured data and unstructured data, the structured data includes farmland environment data, meteorological data and crop growth cycle data, and the unstructured data includes pest image data; The structured data is cleaned and formatted, and the structured data in a unified data format is obtained; The unstructured data is processed by image denoising, size adjustment and color correction, the image data after preliminary processing is processed by grayscale, and the pest image features are identified and classified by a preset classifier; The pest features include shape features, texture features and color features.

[0012] In the second aspect, the above application aims to achieve the following technical solutions: An intelligent pest prediction and control system based on crop growth state, the system is applied to the intelligent pest prediction and control method based on crop growth state, and the system comprises: A data preprocessing module is configured to obtain multi-source observation data related to crop growth state and crop growth environment in the crop growth process, and to preprocess the multi-source observation data to obtain multi-source heterogeneous data of crops. a pest and disease prediction module configured to input the multi-source heterogeneous data into a pre-trained pest and disease prediction model, predict the occurrence probability of pests and diseases under the current growth state and growth environment, and output a pest and disease prediction result conforming to the current crop growth state; a pest and disease control module configured to analyze the current pest and disease prediction result through a pest and disease control expert knowledge base, generate a corresponding control decision, and send the control decision to a corresponding management personnel terminal for pest and disease control warning; a model optimization module configured to monitor the pest and disease control effect of the control decision, feed back the pest and disease control effect to the pest and disease prediction model for model optimization, and adaptively update the pest and disease prediction model.

[0013] In a third aspect, the above object of the present application is achieved by the following technical solution: A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above intelligent pest and disease prediction and control method based on the crop growth state when executing the computer program.

[0014] In a fourth aspect, the above object of the present application is achieved by the following technical solution: A computer readable storage medium storing a computer program, wherein the computer program is executable by a processor to implement the steps of the above intelligent pest and disease prediction and control method based on the crop growth state.

[0015] In summary, the present application has at least one of the following beneficial technical effects: 1. The present application takes the crop growth state and growth environment in the crop growth process as variables, cooperatively constructs a pest and disease prediction model through a deep learning algorithm, provides high-quality data support for the model through data classification preprocessing, performs pest and disease prediction through the pest and disease prediction model, provides reliable pest and disease prediction data conforming to the crop growth state for agricultural production, analyzes the current pest and disease prediction result through a pest and disease control expert knowledge base, timely reminds management personnel of pest and disease control warning through a control decision, continuously tracks the pest and disease control effect, continuously optimizes the model, adaptively updates the pest and disease prediction mechanism of the model, improves the prediction accuracy of the model, and further generates a targeted control scheme for crops in each growth stage, improves the yield and quality of crops, and effectively controls pests and diseases in the entire growth cycle of crops. 2、The application jointly analyzes the pest data and the corresponding crop growth parameters through convolutional neural networks and recurrent neural networks, summarizes the rules between the crop growth state, the growth environment and the occurrence probability of pests and diseases according to the crop growth cycle, and then constructs a pest and disease prediction model. The model is tested by using a data set that does not participate in training, and the accuracy of the test results is judged by referring to the artificial verification results, so as to evaluate the prediction accuracy of the model. It is helpful to make the pest and disease prediction model more accurately predict the occurrence trend of pests and diseases in different growth stages and different growth environments, and provide effective data support for pest and disease control. 3、The application also constructs a pest and disease control expert knowledge base through historical pest and disease data of crops and corresponding artificial control strategies, which is used to provide reference for subsequent pest and disease control decision. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the specific embodiments or the prior art description will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, each element or part is not necessarily drawn according to the actual proportion.

[0017] Figure 1 is the implementation flowchart of the intelligent pest and disease prediction and control method based on the crop growth state of the present embodiment.

[0018] Figure 2 is the implementation flowchart of the data preprocessing of the intelligent pest and disease prediction and control method of the present embodiment.

[0019] Figure 3 is the pest and disease prediction model training flowchart of the intelligent pest and disease prediction and control method of the present embodiment.

[0020] Figure 4 is the implementation flowchart of step S30 of the intelligent pest and disease prediction and control method of the present embodiment.

[0021] Figure 5 is the structural block diagram of the intelligent pest and disease prediction and control system based on the crop growth state.

[0022] Figure 6 is the internal structure schematic diagram of the computer equipment for realizing the intelligent pest and disease prediction and control method. DETAILED DESCRIPTION

[0023] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be clearly and completely described below, obviously, the described embodiments are some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work belong to the protection scope of the present application.

[0024] It should be understood that the terms "comprising" and "including" as used in the specification and the appended claims indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0025] It should also be understood that the terms used in the present application specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the present application specification and the appended claims, unless otherwise clearly indicated by the context, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0026] It should be further understood that the term "and / or" used in the present application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0027] In an embodiment, as shown in Figure 1 The present application discloses an intelligent disease and pest prediction and prevention method based on crop growth state, which specifically includes the following steps: S10: Obtain multi-source observation data related to crop growth state and crop growth environment in the crop growth process, and perform data preprocessing on the multi-source observation data to obtain multi-source heterogeneous data of the crop.

[0028] Specifically, the multi-source observation data includes soil humidity, temperature, light intensity, wind speed, and other farmland environment data, which are collected in real time by sensors installed in the farmland, and also includes weather data such as air temperature, rainfall, humidity, which are obtained from the meteorological department or weather station, and also includes disease and pest image data, which are obtained by deploying cameras or unmanned aerial vehicles in the farmland for shooting, for disease and pest identification and judgment. As shown in Figure 2 The data preprocessing process of the multi-source observation data includes: S101: Classify the multi-source observation data into structured data and unstructured data, the structured data includes farmland environment data, weather data and crop growth cycle data, and the unstructured data includes disease and pest image data.

[0029] Specifically, according to the data type, it is divided into structured data and unstructured data, which is helpful for classified storage. In the embodiment, the data is processed through data classification. For farmland environment data, the accuracy of the data is ensured through calibration of the sensor and periodic checking of the data. For meteorological data, the reliability of the data is ensured through periodic updating with the official meteorological data source. For image data, the authenticity and effectiveness of the image data are ensured through standardized processing through a pre-set image quality processing procedure.

[0030] S102: The structured data is subjected to data cleaning and data format conversion to obtain structured data in a unified data format.

[0031] Specifically, the structured data is subjected to data cleaning, including removing invalid, erroneous or repeated interference data to ensure the integrity of the data, and unifying data of different data sources and different data formats into a standard format to form structured data in a unified data format, which is convenient for subsequent analysis and processing.

[0032] S103: The unstructured data is subjected to image denoising, size adjustment and color correction processing, the image data after preliminary processing is subjected to grayscale processing, and the features of the pest and disease images are identified and classified through a pre-set classifier.

[0033] Specifically, the unstructured data is subjected to image denoising, and the size of all images is adjusted to a unified size standard. The color of all image data is corrected to adjust the color deviation of the images. Finally, the images are subjected to grayscale processing to convert them into grayscale images, which is convenient for identifying the features of the pest and disease images. The features of the pest and disease include shape features, texture features and color features. The classifier identifies and classifies the features of the pest and disease images of different types.

[0034] Specifically, the geometry of the image position meeting the set grayscale value is calculated, including area, perimeter and circularity, which is used to reflect the morphological features of the pest and disease, such as the shape and size of the pest and disease eating leaves; the repeated patterns and structures in the image are extracted, which are used to analyze the texture features of the pest and disease, such as the spot texture and distribution characteristics of the pest and disease; the grayscale value distribution state of the image is analyzed and compared with the grayscale value distribution database of a plurality of types of pest and disease constructed in advance, and then the features of the pest and disease are identified.

[0035] S20: The multi-source heterogeneous data is input into the pre-trained pest and disease prediction model to predict the occurrence probability of the pest and disease under the current growth state and growth environment, and output the pest and disease prediction result meeting the current crop growth state.

[0036] Specifically, the multi-source heterogeneous data is input into the pre-trained pest and disease prediction model to predict the occurrence probability of pests and diseases under the current growth state and growth environment. The occurrence probability of pests and diseases is the ratio between the current predicted pest and disease result and the historical pest and disease result corresponding to the growth stage. The pest and disease prediction result conforming to the current crop growth state is output by predicting the occurrence trend of pests and diseases under the current growth state and growth environment.

[0037] Specifically, as shown in Figure 3 The training process of the pest and disease prediction model in step S20 specifically includes: S201: Obtain crop growth state data and growth environment data of the complete growth cycle of crops, set control groups by the SVM algorithm with growth state and growth environment as variables respectively, and obtain pest and disease data of each control group.

[0038] Specifically, the historical crop growth state data and growth environment data of the complete growth cycle of crops are obtained according to different quarters and different growth cycles, or the training data is obtained by setting control groups of different growth environments and different growth states. The pest and disease data of each control group is obtained by setting control groups by the SVM algorithm.

[0039] S202: Data training is performed on the pest and disease data and the corresponding control group parameters by a deep learning algorithm to analyze the rules between crop growth state, growth environment and pest and disease occurrence probability, and to construct a pest and disease prediction model.

[0040] Specifically, step S202 includes: S2021: Obtain pest and disease features of pest and disease data by a convolutional neural network, obtain sequence data of pest and disease data by a recurrent neural network, and obtain pest and disease association by associating sequence data and corresponding pest and disease features.

[0041] Specifically, pest and disease features in different control groups are extracted by a convolutional neural network, sequence data of pest and disease data in different control groups are obtained by a recurrent neural network, and pest and disease association based on crop growth cycle is obtained by associating sequence data and pest and disease features at the same period according to crop growth cycle.

[0042] S2022: According to the pest and disease association, analyze the rules between crop growth state, growth environment and pest and disease occurrence probability according to crop growth cycle, and construct a pest and disease prediction model.

[0043] Specifically, according to the pest and disease association, the pest and disease occurrence rules between crop growth state, growth environment and pest and disease occurrence probability are summarized according to crop growth cycle, and a pest and disease prediction model is constructed.

[0044] Specifically, the training process of the pest and disease prediction model in step S20 further includes: S203: Test the pest and disease prediction model with a validation data set containing crop growth state data and growth environment data that does not participate in training, to obtain a test result of the validation data set.

[0045] Specifically, the validation data set containing crop growth state data and growth environment data that does not participate in training is input into the pest and disease prediction model for testing, and the test result of the validation data set, i.e., the pest and disease occurrence probability under the corresponding growth state and growth environment, is obtained.

[0046] S204: Manually verify the validation data set, determine the test result through the manual verification result, and evaluate the prediction accuracy of the pest and disease prediction model.

[0047] Specifically, step S204 includes: The error rate, recall rate and F1 score between the manual verification result and the test result are calculated respectively, and the model prediction accuracy is comprehensively evaluated according to the error rate and the recall rate and the F1 score, to obtain the model prediction accuracy evaluation result.

[0048] Specifically, the crop growth data corresponding to the validation data set is manually verified, and the error rate, recall rate and F1 score between the manual verification result and the pest and disease occurrence probability of the model test result are calculated, and the model prediction accuracy is comprehensively evaluated according to the calculation result, to obtain the evaluation result of the model prediction accuracy. If the error rate, recall rate and F1 score reach the corresponding set values, it means that the model prediction accuracy is qualified. If one of the indicators is unqualified, the model is retrained.

[0049] S30: Through the pest and disease control expert knowledge base, the current pest and disease prediction result is analyzed for control, and the corresponding control decision is sent to the corresponding management personnel end for pest and disease control warning.

[0050] Specifically, as shown in Figure 4 step S30 includes: S301: Based on the historical pest and disease data of crops and the corresponding artificial control strategy, a pest and disease control expert knowledge base is constructed.

[0051] Specifically, the historical pest and disease data of crops and the corresponding artificial control strategy are data-associated, and the pest and disease control expert knowledge base is constructed in sequence with the type of pest and disease.

[0052] S302: Compare and analyze the current pest and disease prediction result with the pest and disease control expert knowledge base, select the control strategy that conforms to the current pest and disease trend and crop growth state according to the comparison result, and generate the corresponding control decision.

[0053] Specifically, the current pest prediction result is compared and analyzed with the pest control expert knowledge base, including pest characteristic comparison, crop growth environment parameter comparison, crop growth state comparison, etc., and the control strategy most suitable for the current pest trend and crop growth state is selected to generate the corresponding prevention and control decision through strategy matching of the current pest prediction result.

[0054] S303: Send the prevention and control decision to the corresponding management personnel end for pest prevention and control warning, and listen to the execution state of the prevention and control decision to generate pest prevention and control effect feedback information.

[0055] Specifically, the prevention and control decision is sent to the corresponding management personnel end for pest prevention and control warning, and the prevention and control decision is used as a reference to provide prevention and control suggestions for the management personnel, and the execution state of the prevention and control decision is listened to, including the execution feedback of the management personnel and the crop pest control effect, and the pest position in the subsequent crop image is compared and obtained, such as the reduction or increase of the leaf spot of the pest position, to generate the pest prevention and control effect feedback information.

[0056] S40: Monitor the pest control effect of the prevention and control decision, and feed back the pest control effect to the pest prediction model for model optimization and adaptive updating of the pest prediction model.

[0057] Specifically, the pest control effect of the prevention and control decision is listened to, such as the change of the image gray value of the same pest position after the execution of the pest prevention and control decision, including the change of the shape feature, texture feature, color feature, etc., and according to the change trend, such as the pest shape expansion, texture deepening or color deepening, etc., as invalid pest control effect, such as the pest shape becoming smaller, the texture or color becoming lighter, etc., as effective pest control effect, the pest control effect is fed back to the model, the pest prediction result of the model is fed back, the pest prediction mechanism of the model is optimized, and the model is adaptively updated.

[0058] It should be understood that the size of the serial number of each step in the above embodiments does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0059] In an embodiment, an intelligent pest prediction and control system based on crop growth state is provided. The intelligent pest prediction and control system based on crop growth state corresponds to the intelligent pest prediction and control method based on crop growth state in the above embodiments. As shown in the figure, the intelligent pest prediction and control system based on crop growth state includes a data preprocessing module, a pest prediction module, a pest control module, and a model optimization module. The functions of each module are described in detail as follows: Figure 5 ​The data preprocessing module is configured to acquire multi-source observation data related to the crop growth state and the crop growth environment in the crop growth process, and perform data preprocessing on the multi-source observation data to obtain multi-source heterogeneous data of the crop.

[0060] The pest and disease prediction module is configured to input the multi-source heterogeneous data into a pre-trained pest and disease prediction model, predict the occurrence probability of pests and diseases under the current growth state and growth environment, and output a pest and disease prediction result corresponding to the current crop growth state.

[0061] The pest and disease control module is configured to perform control analysis on the current pest and disease prediction result by using a pest and disease control expert knowledge base, and generate a corresponding control decision and send the control decision to a corresponding management personnel terminal for pest and disease control warning.

[0062] The model optimization module is configured to monitor the pest and disease control effect of the control decision, feed back the pest and disease control effect to the pest and disease prediction model for model optimization, and perform self-adaptive update on the pest and disease prediction model.

[0063] Preferably, the training process of the pest and disease prediction model specifically includes: The training data setting submodule is configured to acquire crop growth state data and growth environment data in a complete crop growth cycle, set a control group by taking the growth state and the growth environment as variables respectively by using an SVM algorithm, and acquire pest and disease data of each control group.

[0064] The model construction submodule is configured to perform data training on the pest and disease data and corresponding control group parameters by using a deep learning algorithm, analyze the relationship between the crop growth state, the growth environment, and the occurrence probability of pests and diseases, and construct a pest and disease prediction model.

[0065] Preferably, the model construction submodule includes: The correlation relationship analysis unit is configured to acquire pest and disease characteristics of the pest and disease data by using a convolutional neural network, acquire sequence data of the pest and disease data by using a recurrent neural network, and obtain a pest and disease correlation relationship by associating the sequence data and the corresponding pest and disease characteristics.

[0066] The rule analysis unit is configured to analyze the relationship between the crop growth state, the growth environment, and the occurrence probability of pests and diseases according to the pest and disease correlation relationship, and construct a pest and disease prediction model according to the crop growth cycle.

[0067] Preferably, the training process of the pest and disease prediction model further includes: The model testing submodule is configured to test the pest and disease prediction model by using a verification data set containing crop growth state data and growth environment data which does not participate in the training, and obtain a test result of the verification data set.

[0068] The model evaluation submodule is configured to perform manual verification on the validation dataset, determine the test result based on the manual verification result, and evaluate the prediction accuracy of the pest and disease prediction model.

[0069] Preferably, the model evaluation submodule specifically comprises: The error rate, recall rate and F1 score between the manual verification result and the test result are calculated respectively, and the model prediction accuracy is comprehensively evaluated based on the error rate, recall rate and F1 score to obtain the model prediction accuracy evaluation result.

[0070] Preferably, the pest and disease control module specifically comprises: The knowledge base construction submodule is configured to construct a pest and disease control expert knowledge base based on historical pest and disease data of crops and corresponding artificial control strategies.

[0071] The control decision submodule is configured to compare and analyze the current pest and disease prediction result with the pest and disease control expert knowledge base, select a control strategy that conforms to the current pest and disease trend and the growth state of crops based on the comparison result, and generate a corresponding control decision.

[0072] The control feedback submodule is configured to send the control decision to the corresponding management personnel terminal for pest and disease control warning, listen to the execution state of the control decision, and generate pest and disease control effect feedback information.

[0073] Preferably, the data preprocessing process in the data preprocessing module specifically comprises: The data classification submodule is configured to classify the multi-source observation data into structured data and unstructured data, wherein the structured data includes farmland environment data, meteorological data and crop growth cycle data, and the unstructured data includes pest and disease image data.

[0074] The data format conversion submodule is configured to clean and format the structured data to obtain structured data in a unified data format.

[0075] The data processing submodule is configured to perform image denoising, size adjustment and color correction processing on the unstructured data, perform grayscale processing on the preliminarily processed image data, and identify and classify the pest and disease image features through a pre-set classifier.

[0076] The pest and disease features include shape features, texture features and color features.

[0077] Specific limitations regarding the intelligent pest and disease prediction and control system based on crop growth status can be found in the limitations of the intelligent pest and disease prediction and control method based on crop growth status mentioned above, and will not be repeated here. Each module in the aforementioned intelligent pest and disease prediction and control system based on crop growth status can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0078] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 6 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides the environment for the operation of the operating system and computer programs in the non-volatile storage media. The database stores data related to pest and disease prediction and control. The network interface communicates with external terminals via a network. When the computer program is executed by the processor, it implements an intelligent pest and disease prediction and control method based on crop growth status.

[0079] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program being executed by a processor to implement the steps of an intelligent method for predicting and controlling pests and diseases based on crop growth status.

[0080] Those skilled in the art will recognize that the units of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application of the technical solution and the constraints involved. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of the invention.

[0081] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical functional division. In actual implementation, there may be other division methods, such as multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored.

[0082] In addition, each function unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software function unit.

[0083] When the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the part of the prior art that essentially contributes or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0084] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of each embodiment of the present application, and they should be covered in the scope of the claims and the description of the present application.

Claims

1. An intelligent method for predicting and controlling pests and diseases based on crop growth status, characterized in that, The method includes: Multi-source observation data related to crop growth status and crop growth environment during the crop growth process are acquired, and the multi-source observation data are preprocessed to obtain multi-source heterogeneous data of crops. The multi-source heterogeneous data is input into a pre-trained pest and disease prediction model to predict the probability of pest and disease occurrence under the current growth status and growth environment, and output pest and disease prediction results that conform to the current crop growth status. By using the knowledge base of pest and disease control experts, the current pest and disease forecast results are analyzed for prevention and control, and corresponding prevention and control decisions are generated and sent to the corresponding management personnel for pest and disease control early warning. The effectiveness of the pest and disease control decisions is monitored, and the effectiveness of pest and disease control is fed back to the pest and disease prediction model for model optimization. The pest and disease prediction model is then adaptively updated.

2. The intelligent pest and disease prediction and control method based on crop growth status according to claim 1, characterized in that, The training process of the pest and disease prediction model specifically includes: Obtain crop growth status data and growth environment data for the entire crop growth cycle. Use the SVM algorithm to set up control groups with growth status and growth environment as variables, and obtain pest and disease data for each control group. By using deep learning algorithms to train the pest and disease data and corresponding control group parameters, the relationship between crop growth status, growth environment and the probability of pest and disease occurrence is analyzed, and a pest and disease prediction model is constructed.

3. The intelligent pest and disease prediction and control method based on crop growth status according to claim 2, characterized in that, The process involves training the pest and disease data and corresponding control group parameters using a deep learning algorithm to analyze the relationship between crop growth status, growth environment, and the probability of pest and disease occurrence, thereby constructing a pest and disease prediction model, including: The pest and disease features of the pest and disease data are obtained by using a convolutional neural network, and the sequence data of the pest and disease data is obtained by using a recurrent neural network. The pest and disease association relationship is obtained by associating the sequence data with the corresponding pest and disease features. Based on the aforementioned disease and pest correlation, and by analyzing the relationship between crop growth status, growth environment, and the probability of disease and pest occurrence according to the crop growth cycle, a disease and pest prediction model is constructed.

4. The intelligent pest and disease prediction and control method based on crop growth status according to claim 2, characterized in that, The training process of the pest and disease prediction model also includes: The pest and disease prediction model was tested using a validation dataset containing crop growth status data and growth environment data that was not used in the training, and the test results of the validation dataset were obtained. The validation dataset is manually validated, and the test results are judged based on the manual validation results to evaluate the prediction accuracy of the pest and disease prediction model.

5. The intelligent pest and disease prediction and control method based on crop growth status according to claim 4, characterized in that, The step of manually validating the validation dataset, judging the test results based on the manual validation results, and evaluating the prediction accuracy of the pest and disease prediction model specifically includes: The error rate, recall rate, and F1 score between the manual validation results and the test results are calculated separately. The model prediction accuracy is comprehensively evaluated based on the error rate, recall rate, and F1 score to obtain the model prediction accuracy evaluation result.

6. The intelligent pest and disease prediction and control method based on crop growth status according to claim 1, characterized in that, The process involves using a pest and disease control expert knowledge base to analyze current pest and disease forecasts, generating corresponding control decisions, and sending these decisions to relevant management personnel for pest and disease control early warning. Specifically, this includes: Based on historical crop pest and disease data and corresponding manual control strategies, a knowledge base of experts in pest and disease control will be constructed. The current pest and disease prediction results are compared and analyzed with the pest and disease control expert knowledge base. Based on the comparison results, control strategies that conform to the current pest and disease trends and crop growth status are selected, and corresponding control decisions are generated. The control and prevention decisions are sent to the corresponding management personnel for pest and disease control early warning, and the execution status of the control and prevention decisions is monitored to generate feedback information on the effectiveness of pest and disease control.

7. The intelligent pest and disease prediction and control method based on crop growth status according to claim 1, characterized in that, The process of acquiring multi-source observation data related to crop growth status and crop growth environment during crop growth, and preprocessing the multi-source observation data to obtain multi-source heterogeneous data of crops, specifically includes: The multi-source observation data is classified into structured and unstructured data. The structured data includes farmland environment data, meteorological data, and crop growth cycle data, while the unstructured data includes pest and disease image data. The structured data is cleaned and formatted to obtain structured data with a unified data format. The unstructured data is subjected to image denoising, resizing, and color correction. The pre-processed image data is then converted to grayscale. Finally, a preset classifier is used to identify and classify the features of the pest and disease images. The characteristics of the pests and diseases include shape characteristics, texture characteristics, and color characteristics.

8. An intelligent pest and disease prediction and control system based on crop growth status, characterized in that, The system is applied to the intelligent pest and disease prediction and control method based on crop growth status as described in any one of claims 1-7, and the system comprises: The data preprocessing module is used to acquire multi-source observation data related to crop growth status and crop growth environment during the crop growth process, and to perform data preprocessing on the multi-source observation data to obtain multi-source heterogeneous data of crops. The pest and disease prediction module is used to input the multi-source heterogeneous data into the pre-trained pest and disease prediction model, predict the probability of pest and disease occurrence under the current growth status and growth environment, and output pest and disease prediction results that conform to the current crop growth status. The pest and disease control module is used to analyze the current pest and disease prediction results through the pest and disease control expert knowledge base, generate corresponding control decisions, and send them to the corresponding management personnel for pest and disease control early warning. The model optimization module is used to monitor the pest and disease control effect of the control decision, feed the pest and disease control effect back to the pest and disease prediction model for model optimization, and adaptively update the pest and disease prediction model.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the intelligent pest and disease prediction and control method based on crop growth status as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the intelligent pest and disease prediction and control method based on crop growth status as described in any one of claims 1 to 7.

Images