Crop disease and pest monitoring and early warning system based on satellite remote sensing and AI algorithm

Abstract

The invention relates to the technical field of environmental data monitoring, and discloses a crop disease and insect pest monitoring and early warning system based on satellite remote sensing and an AI algorithm, and the system comprises the following steps: regularly collecting a farmland multispectral image covering a farmland area through a satellite remote sensing technology, and carrying out the preprocessing of the farmland multispectral image, and obtaining a farmland three-dimensional space image; performing pest and disease analysis on the farmland three-dimensional space image to extract a pest and disease image and spatial distribution thereof, analyzing historical pest and disease data based on an AI algorithm analysis model, obtaining pest-inducing influence factors corresponding to different target types of pests, and obtaining a pest and disease development trend corresponding to the target types of pests based on the pest-inducing influence factors; and compiling an intelligent contract for the disease and pest influence degree corresponding to the current disease and pest image, recording the timestamp of the disease and pest influence degree in the intelligent contract, and establishing a disease and pest early warning strategy based on evaluation reports of different time periods.

Description

Technical Field

[0001] The present invention relates to the field of agricultural monitoring technology, and more specifically, to a crop disease and insect pest monitoring and early warning system based on satellite remote sensing and AI algorithms. Background Art

[0002] Traditional crop pest and disease monitoring methods often rely on manual field surveys, which is not only time-consuming and labor-intensive, but also difficult to cover large areas, resulting in low monitoring efficiency and untimely early warning. In addition, although the existing ground sensor-based monitoring system can provide local information, there are still blind spots for monitoring in vast farmland.

[0003] In the field of agricultural environment, with the application of machine learning models, such as the Chinese authorized patent: CN111985370B, a fine-grained recognition method for crop pests and diseases based on an improved hybrid attention module is disclosed. The convolutional network is constructed using the Inception and residual learning ideas, which can reduce the mapping interval and retain the details of the original image. It mainly identifies pests and diseases based on collected images. However, in actual applications, due to the large area of ​​farmland, it is impossible to collect accurate real-time images of pests and diseases. It usually provides basic pest and disease information and there is no integrated intelligent prediction and decision support system. Agricultural managers can only rely on experience to make decisions, which may lead to inaccurate or untimely prevention and control measures.

[0004] In view of this, the present invention proposes a crop disease and pest monitoring and early warning system based on satellite remote sensing and AI algorithm. Summary of the invention In order to overcome the shortcomings of the prior art, the present invention provides a crop disease and pest monitoring and early warning system based on satellite remote sensing and AI algorithm, which has the advantage of more precise operation.

[0005] In a first aspect, the present invention provides a crop pest and disease monitoring and early warning system based on satellite remote sensing and AI algorithm, including a data acquisition module, a data analysis module, a trend prediction module and an early warning decision module, wherein each module is connected by wire or wirelessly; The data acquisition module regularly collects multispectral images of farmland covering the farmland area through satellite remote sensing technology, and pre-processes the multispectral images of farmland to obtain three-dimensional spatial images of farmland; A data analysis module, which performs pest and disease analysis on the three-dimensional spatial image of the farmland to extract pest and disease images and their spatial distribution, wherein the pest and disease images include at least one target type of pest image; The trend prediction module analyzes historical pest data based on the AI ​​algorithm analysis model, obtains the pest inducing factors corresponding to different target pest types, assigns probability values ​​to the pest disaster level based on the pest inducing factors, and performs statistics based on the probability values ​​of the pest disaster level to obtain the pest development trend corresponding to the target pest type; The early warning decision module writes a smart contract for the degree of pest and disease impact corresponding to the current pest and disease image, records the timestamp of the degree of pest and disease impact in the smart contract, and evaluates the actual pest and disease impact considering the transmission time, thereby generating evaluation reports of different time limits and establishing pest and disease early warning strategies based on evaluation reports of different time limits.

[0006] As a preferred technical solution of the first aspect of the present invention, multiple satellites are used to periodically acquire multispectral images of farmland covering multiple bands, wherein the multispectral images of farmland include b1 band spectral images, b2 band spectral images, b3 band spectral images and b4 band spectral images; the multispectral images of farmland are segmented and the farmland planting areas are extracted through spectral feature analysis.

[0007] As a preferred technical solution of the first aspect of the present invention, the logic for acquiring the three-dimensional spatial image of the farmland is: Perform geometric positioning, edge trimming and noise removal on the farmland planting area to obtain the initial processed farmland image; Searching for at least one fixed marker in the initially processed farmland image, marking the fixed marker as a reference point, and performing spatial alignment based on the reference point to obtain a reference farmland image; The reference farmland image is grayed out using a weighted average method, the Rank transformation result of the gray image is used as a matching primitive, and a region matching algorithm based on normalized absolute difference and a measurement function is used to obtain a dense disparity map of the scene; The spatial coordinates of the scene are calculated based on the principle of parallel binocular vision imaging, and a three-dimensional point cloud map is generated; a three-dimensional spatial image of the farmland is generated based on the three-dimensional point cloud map.

[0008] As a preferred technical solution of the first aspect of the present invention, the acquisition logic of the pest image is: Collect the measured species grayscale values ​​of the species corresponding to the three-dimensional spatial image of the farmland, and compare the measured species grayscale values ​​with the species reference grayscale interval corresponding to the same time series information, wherein the species reference grayscale interval is the grayscale pixel interval value of the normal growth of the current species in the experimental planting environment obtained based on the analysis of the historical database; Screening pixel blocks whose measured species grayscale values ​​are not within the species reference grayscale interval, treating the pixel blocks that are not within the species reference grayscale interval as abnormal image blocks, and marking the spatial position corresponding to the abnormal image block as an abnormal spatial position; According to the pre-set continuous time series interval, the abnormal support and abnormal co-occurrence corresponding to the abnormal spatial position in the continuous time series interval are counted, wherein the abnormal support is used to indicate the total number of abnormal marks corresponding to the same abnormal spatial position in the continuous time series interval, and the abnormal co-occurrence is used to indicate the number of times different abnormal spatial positions in the continuous time series interval co-occur; The continuous time series interval is a continuous timestamp of the multispectral images of farmland collected by satellites; when the anomaly support in the continuous time series interval is greater than a preset anomaly support threshold and the anomaly co-occurrence is greater than a preset anomaly expansion threshold, the corresponding abnormal image block is marked as a pest image, and the pest image includes at least one target type pest image and the spatial distribution of the corresponding target type pest.

[0009] As a preferred technical solution of the first aspect of the present invention, the logic for acquiring the target type pest image is: Performing texture analysis on the pest image, extracting pest texture features through a gray-level co-occurrence matrix, and obtaining a pest texture feature correlation amount; The similarity is calculated based on the normalized calculation formula according to the pest texture feature correlation quantity and the pest texture feature defined in the historical database, and the abnormal image blocks with similarity within the pest similarity threshold range are screened out, and the abnormal image blocks are marked as target species pest images.

[0010] As a preferred technical solution of the first aspect of the present invention, the logic for obtaining the development trend of pests and diseases is: The target pest images corresponding to different types of pests and diseases are matrixed according to the abnormal spatial positions to obtain the spatial distribution corresponding to the target pest images; Construct an AI algorithm analysis model, using meteorological data, images of different target species of pests and their spatial distribution as input factors, and output the induced pest impact factors corresponding to the images of different target species of pests; The probability of pest disaster level is assigned based on the pest-inducing impact factor. The larger the probability value, the greater the probability of pest disaster. The probability assignment of different types of pests to the pest disaster level is superimposed to obtain the pest impact degree corresponding to the current pest image, and the pest development trend is characterized based on the pest impact degree.

[0011] As a preferred technical solution of the first aspect of the present invention, the acquisition logic of the pest warning strategy is: Obtain the spatial distribution and collection timestamps of different target pests; store the current pest impact based on the timestamp, and map the historical pest records corresponding to the target pests based on the timestamp; Based on the analysis of historical pest and disease records, and combined with the transmission time, the actual impact of pests and diseases is evaluated, and assessment reports of different time limits are recorded. The timestamps are rewritten according to different time limits, and displayed in the pest and disease early warning strategy corresponding to the smart contract.

[0012] In a second aspect, the present invention provides a crop disease and insect pest monitoring and early warning method based on satellite remote sensing and AI algorithm, based on the implementation of a crop disease and insect pest monitoring and early warning system based on satellite remote sensing and AI algorithm described in the first aspect, comprising the following steps: The multispectral images of farmland covering the farmland area are collected regularly through satellite remote sensing technology, and the multispectral images of farmland are preprocessed to obtain the three-dimensional spatial images of farmland; Performing pest and disease analysis on the three-dimensional spatial image of the farmland to extract pest and disease images and their spatial distribution, wherein the pest and disease images include at least one target type of pest image; Analyze historical pest data based on AI algorithm analysis model, obtain the pest inducing factors corresponding to different target pest types, assign probability values ​​to pest disaster levels based on the pest inducing factors, perform statistics based on the probability values ​​of pest disaster levels, and obtain pest development trends corresponding to target pest types; The degree of pest and disease impact corresponding to the current pest and disease image is written into a smart contract, the timestamp of the degree of pest and disease impact is recorded in the smart contract, and the actual pest and disease impact is evaluated considering the transmission time, so as to generate evaluation reports of different time limits, and establish a pest and disease early warning strategy based on the evaluation reports of different time limits.

[0013] In a third aspect, the present invention provides an electronic device, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the crop disease and pest monitoring and early warning system based on satellite remote sensing and AI algorithm described in the first aspect by calling the computer program stored in the memory.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium storing instructions, which, when executed on a computer, enables the computer to execute the crop disease and pest monitoring and early warning system based on satellite remote sensing and AI algorithm as described in the first aspect.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention integrates the monitoring system of satellite remote sensing technology and artificial intelligence algorithm, which can significantly reduce the cost of traditional farmland pest and disease monitoring, improve monitoring efficiency and accuracy, achieve early warning of pests and diseases, and provide scientific support for agricultural decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1It is a schematic diagram of the framework of the crop pest monitoring and early warning system of the present invention; Figure 2 A schematic diagram of a crop disease and insect pest monitoring method in an agricultural environment of the present invention; Figure 3 This is a flow chart of the crop pest monitoring and early warning method of the present invention; Figure 4 The figure is a schematic diagram of the structure of an electronic device of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Example 1: Figure 1-2 As shown, the present invention provides a crop pest monitoring and early warning system based on satellite remote sensing and AI algorithm, including a data acquisition module 100, a data analysis module 200, a trend prediction module 300 and an early warning decision module 400, and each module is connected by wire or wirelessly; The data acquisition module 100 collects farmland multispectral images covering the farmland area regularly through satellite remote sensing technology, and pre-processes the farmland multispectral images to obtain the farmland three-dimensional spatial images; Specifically, multiple high-resolution satellites are used to regularly acquire multispectral images of farmland covering multiple bands such as infrared and visible light. The multispectral images of farmland include b1 band spectral images, b2 band spectral images, b3 band spectral images and b4 band spectral images. The multispectral images of farmland are segmented and the farmland planting areas are extracted through spectral feature analysis, laying the foundation for subsequent vegetation index calculation and pest and disease identification.

[0019] The logic for acquiring the three-dimensional spatial image of farmland is as follows: The farmland planting area is geometrically positioned, edge trimmed and noise removed to obtain the initial processed farmland image. The initial processed farmland image is the farmland image after the farmland planting area is subjected to radiation correction and atmospheric correction, eliminating the influence of atmospheric scattering and absorption, and obtaining the physically quantified apparent radiation brightness to ensure the accuracy of subsequent analysis; It should be noted that the distortion caused by satellite attitude changes and terrain undulations is corrected. In addition, median filtering and Gaussian filtering are used to denoise the image and improve the signal-to-noise ratio of the image.

[0020] Searching for at least one fixed marker in the initially processed farmland image, marking the fixed marker as a reference point, and performing spatial alignment based on the reference point to obtain a reference farmland image for subsequent time series analysis; The reference farmland image is grayed out using a weighted average method, the Rank transformation result of the gray image is used as a matching primitive, and a region matching algorithm based on normalized absolute difference and a measurement function is used to obtain a dense disparity map of the scene; The spatial coordinates of the scene are calculated based on the principle of parallel binocular vision imaging, and a three-dimensional point cloud map is generated; a three-dimensional spatial image of the farmland is generated based on the three-dimensional point cloud map.

[0021] It should be noted that a Rank transform-based three-dimensional modeling method for farmland scenes is adopted. This method grays the image by weighted average method and uses the Rank transform result as a matching primitive to improve the robustness of the matching.

[0022] The data analysis module 200 performs pest and disease analysis on the three-dimensional spatial image of the farmland to extract pest and disease images and their spatial distribution, wherein the pest and disease images include at least one target type of pest image; It should be noted that the types and spatial distribution of pests and diseases are automatically identified, and the results are intuitively superimposed on the remote sensing image to show the location and severity of the pests and diseases.

[0023] Specifically, the acquisition logic of the pest and disease image is: Collect the measured species grayscale values ​​of the species corresponding to the three-dimensional spatial image of the farmland, and compare the measured species grayscale values ​​with the species reference grayscale interval corresponding to the same time series information, wherein the species reference grayscale interval is the grayscale pixel interval value of the normal growth of the current species in the experimental planting environment obtained based on the analysis of the historical database; Screening pixel blocks whose measured species grayscale values ​​are not within the species reference grayscale interval, treating the pixel blocks that are not within the species reference grayscale interval as abnormal image blocks, and marking the spatial position corresponding to the abnormal image block as an abnormal spatial position; According to the pre-set continuous time series interval, the abnormal support and abnormal co-occurrence corresponding to the abnormal spatial position in the continuous time series interval are counted, wherein the abnormal support is used to indicate the total number of abnormal marks corresponding to the same abnormal spatial position in the continuous time series interval, and the abnormal co-occurrence is used to indicate the number of times different abnormal spatial positions in the continuous time series interval co-occur; The continuous time series interval is a continuous timestamp of the multispectral images of farmland collected by satellites; when the anomaly support in the continuous time series interval is greater than a preset anomaly support threshold and the anomaly co-occurrence is greater than a preset anomaly expansion threshold, the corresponding abnormal image block is marked as a pest image, and the pest image includes at least one target type pest image and the spatial distribution of the corresponding target type pest.

[0024] The logic for acquiring the target type pest image: Performing texture analysis on the pest image, extracting pest texture features through a gray-level co-occurrence matrix, and obtaining a pest texture feature correlation amount; The similarity is calculated based on the normalized calculation formula according to the pest texture feature correlation quantity and the pest texture feature defined in the historical database, and the abnormal image blocks with similarity within the pest similarity threshold range are screened out, and the abnormal image blocks are marked as target species pest images.

[0025] The trend prediction module 300 analyzes historical pest data based on an AI algorithm analysis model, obtains pest inducing impact factors corresponding to different target pest types, assigns probability values ​​to pest disaster levels based on the pest inducing impact factors, performs statistics based on the probability values ​​of pest disaster levels, and obtains pest development trends corresponding to target pest types; Specifically, the logic for obtaining the development trend of pests and diseases is: The target pest images corresponding to different types of pests and diseases are matrixed according to the abnormal spatial positions to obtain the spatial distribution corresponding to the target pest images; Construct an AI algorithm analysis model, using meteorological data, images of different target species of pests and their spatial distribution as input factors, and output the induced pest impact factors corresponding to the images of different target species of pests; The probability of pest disaster level is assigned based on the pest-inducing impact factor. The larger the probability value, the greater the probability of pest disaster. The probability assignment of different types of pests to the pest disaster level is superimposed to obtain the pest impact degree corresponding to the current pest image, and the pest development trend is characterized based on the pest impact degree.

[0026] The early warning decision module 400 compiles a smart contract for the degree of impact of pests and diseases corresponding to the current pest and disease image, records the timestamp of the degree of impact of pests and diseases in the smart contract, and evaluates the actual impact of pests and diseases considering the transmission time, thereby generating evaluation reports of different time limits, and establishing a pest and disease early warning strategy based on the evaluation reports of different time limits.

[0027] Specifically, the acquisition logic of the pest and disease early warning strategy is: Obtain the spatial distribution and collection timestamps of different target pests; store the current pest impact based on the timestamp, and map the historical pest records corresponding to the target pests based on the timestamp; Based on the analysis of historical pest and disease records, and combined with the transmission time, the actual impact of pests and diseases is evaluated, and assessment reports of different time limits are recorded. The timestamps are rewritten according to different time limits, and displayed in the pest and disease early warning strategy corresponding to the smart contract.

[0028] It should be noted that in order to provide more effective prevention and control measures, the system will compare and analyze the identification results with historical pest and disease data, use time series models to analyze the development trend of pests and diseases, and predict the peak period in the future. Based on these prediction results, pest and disease monitoring reports are automatically generated to help farmers take prevention and control measures in advance.

[0029] This embodiment supports automated and regular remote sensing image acquisition. By regularly acquiring remote sensing images of farmland, the system can achieve continuous monitoring of farmland health conditions, avoid the limitations of manual inspections, and have a wide coverage and high efficiency. Using multispectral image processing technology and deep learning models, the system can automatically identify the types and distribution of pests and diseases, with high analysis accuracy and fast recognition speed, greatly improving the efficiency and accuracy of data analysis. The system combines historical data with current recognition results, infers the development trend of pests and diseases through prediction models, and provides real-time early warning information. Agricultural managers can take prevention and control measures in advance based on this information to avoid further spread of pests and diseases. Combined with meteorological data and environmental factors, the system provides accurate prevention and control suggestions to help agricultural managers make scientific decisions. This function makes farmland management more intelligent and precise. Through automated and regular remote sensing image acquisition and intelligent analysis, the system reduces dependence on manual inspections, improves detection efficiency, and ensures that farmland pests and diseases can be discovered in a timely manner. Through advanced image processing and deep learning technologies, the system can accurately identify the types and distribution of pests and diseases, and provide early warning information in combination with prediction models, ensuring that agricultural managers can take measures before pests and diseases spread on a large scale. The intelligent decision-making support function provided by the system enables agricultural managers to formulate more scientific and accurate prevention and control strategies based on real-time monitoring data and prediction results, thereby improving the effectiveness of pest and disease prevention and control and reducing crop losses. Combined with multi-source data (such as meteorological data and historical pest and disease data), the system can conduct a comprehensive analysis of farmland, provide all-round management support, and improve the overall level of agricultural management.

[0030] Example 2: Please refer to Figure 3 As shown, the part not described in detail in this embodiment is described in Example 1. This embodiment provides a crop disease and insect pest monitoring and early warning method based on satellite remote sensing and AI algorithm, including the following steps: The multispectral images of farmland covering the farmland area are collected regularly through satellite remote sensing technology, and the multispectral images of farmland are preprocessed to obtain the three-dimensional spatial images of farmland; Performing pest and disease analysis on the three-dimensional spatial image of the farmland to extract pest and disease images and their spatial distribution, wherein the pest and disease images include at least one target type of pest image; Analyze historical pest data based on AI algorithm analysis model, obtain the pest inducing factors corresponding to different target pest types, assign probability values ​​to pest disaster levels based on the pest inducing factors, perform statistics based on the probability values ​​of pest disaster levels, and obtain pest development trends corresponding to target pest types; The degree of pest and disease impact corresponding to the current pest and disease image is written into a smart contract, the timestamp of the degree of pest and disease impact is recorded in the smart contract, and the actual pest and disease impact is evaluated considering the transmission time, so as to generate evaluation reports of different time limits, and establish a pest and disease early warning strategy based on the evaluation reports of different time limits.

[0031] Multiple satellites are used to regularly acquire farmland multispectral images covering multiple bands, wherein the farmland multispectral images include b1 band spectral images, b2 band spectral images, b3 band spectral images and b4 band spectral images; the farmland multispectral images are segmented and the farmland planting areas are extracted through spectral feature analysis.

[0032] The logic for acquiring the three-dimensional spatial image of farmland is as follows: Perform geometric positioning, edge trimming and noise removal on the farmland planting area to obtain the initial processed farmland image; Searching for at least one fixed marker in the initially processed farmland image, marking the fixed marker as a reference point, and performing spatial alignment based on the reference point to obtain a reference farmland image; The reference farmland image is grayed out using a weighted average method, the Rank transformation result of the gray image is used as a matching primitive, and a region matching algorithm based on normalized absolute difference and a measurement function is used to obtain a dense disparity map of the scene; The spatial coordinates of the scene are calculated based on the principle of parallel binocular vision imaging, and a three-dimensional point cloud map is generated; a three-dimensional spatial image of the farmland is generated based on the three-dimensional point cloud map.

[0033] The acquisition logic of the pest and disease image is: Collect the measured species grayscale values ​​of the species corresponding to the three-dimensional spatial image of the farmland, and compare the measured species grayscale values ​​with the species reference grayscale interval corresponding to the same time series information, wherein the species reference grayscale interval is the grayscale pixel interval value of the normal growth of the current species in the experimental planting environment obtained based on the analysis of the historical database; Screening pixel blocks whose measured species grayscale values ​​are not within the species reference grayscale interval, treating the pixel blocks that are not within the species reference grayscale interval as abnormal image blocks, and marking the spatial position corresponding to the abnormal image block as an abnormal spatial position; According to the pre-set continuous time series interval, the abnormal support and abnormal co-occurrence corresponding to the abnormal spatial position in the continuous time series interval are counted, wherein the abnormal support is used to indicate the total number of abnormal marks corresponding to the same abnormal spatial position in the continuous time series interval, and the abnormal co-occurrence is used to indicate the number of times different abnormal spatial positions in the continuous time series interval co-occur; The continuous time series interval is a continuous timestamp of the multispectral images of farmland collected by satellites; when the anomaly support in the continuous time series interval is greater than a preset anomaly support threshold and the anomaly co-occurrence is greater than a preset anomaly expansion threshold, the corresponding abnormal image block is marked as a pest image, and the pest image includes at least one target type pest image and the spatial distribution of the corresponding target type pest.

[0034] The logic for acquiring the target type pest image: Performing texture analysis on the pest image, extracting pest texture features through a gray-level co-occurrence matrix, and obtaining a pest texture feature correlation amount; The similarity is calculated based on the normalized calculation formula according to the pest texture feature correlation quantity and the pest texture feature defined in the historical database, and the abnormal image blocks with similarity within the pest similarity threshold range are screened out, and the abnormal image blocks are marked as target species pest images.

[0035] The logic for obtaining the development trend of pests and diseases is as follows: The target pest images corresponding to different types of pests and diseases are matrixed according to the abnormal spatial positions to obtain the spatial distribution corresponding to the target pest images; Construct an AI algorithm analysis model, using meteorological data, images of different target species of pests and their spatial distribution as input factors, and output the induced pest impact factors corresponding to the images of different target species of pests; The probability of pest disaster level is assigned based on the pest-inducing impact factor. The larger the probability value, the greater the probability of pest disaster. The probability assignment of different types of pests to the pest disaster level is superimposed to obtain the pest impact degree corresponding to the current pest image, and the pest development trend is characterized based on the pest impact degree.

[0036] The acquisition logic of the pest and disease early warning strategy is: Obtain the spatial distribution and collection timestamps of different target pests; store the current pest impact based on the timestamp, and map the historical pest records corresponding to the target pests based on the timestamp; Based on the analysis of historical pest and disease records, and combined with the transmission time, the actual impact of pests and diseases is evaluated, and assessment reports of different time limits are recorded. The timestamps are rewritten according to different time limits, and displayed in the pest and disease early warning strategy corresponding to the smart contract.

[0037] Example 3 An electronic device according to an exemplary embodiment includes: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the above-mentioned crop disease and pest monitoring and early warning system based on satellite remote sensing and AI algorithm by calling the computer program stored in the memory.

[0038] Figure 4 It is a structural diagram of an electronic device provided in an embodiment of the present application. The electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) and one or more memories, wherein at least one computer program is stored in the memory, and the at least one computer program is loaded and executed by the processor to implement a crop pest monitoring and early warning system based on satellite remote sensing and AI algorithms provided in the above-mentioned various method embodiments. The electronic device may also include other components for realizing the functions of the device. For example, the electronic device may also have components such as a wired or wireless network interface and an input and output interface for input and output. The embodiments of the present application will not be described in detail here.

[0039] Example 4 According to an exemplary embodiment, a computer-readable storage medium is shown, on which a rewritable computer program is stored; When the computer program runs on a computer device, the computer device executes the above-mentioned crop disease and insect pest monitoring and early warning system based on satellite remote sensing and AI algorithm.

[0040] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including at least one computer program, and the at least one computer program is executable by a processor to complete a crop pest monitoring and early warning system based on satellite remote sensing and AI algorithm in the above embodiment. For example, the computer-readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, etc.

[0041] In an exemplary embodiment, a computer program product or a computer program is also provided, the computer program product or the computer program includes one or more program codes, and the one or more program codes are stored in a computer-readable storage medium. One or more processors of an electronic device can read the one or more program codes from the computer-readable storage medium, and the one or more processors execute the one or more program codes, so that the electronic device can execute the above-mentioned crop pest and disease monitoring and early warning system based on satellite remote sensing and AI algorithm.

[0042] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not 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 embodiments of the present application.

[0043] It should be understood that determining B based on A does not mean determining B only based on A. B can also be determined based on A and / or other information.

[0044] A person of ordinary skill in the art will understand that all or part of the steps to implement the above embodiments can be completed by hardware, or can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. The above-mentioned storage medium can be a read-only memory, a disk or an optical disk, etc.

[0045] The above description is only an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0046] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0047] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0048] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0049] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A crop pest monitoring and early warning system based on satellite remote sensing and AI algorithm, characterized in that: It comprises a data acquisition module (100), a data analysis module (200), a trend prediction module (300) and an early warning decision module (400), wherein the modules are connected to each other via wired or wireless connections; A data acquisition module (100) is used to regularly collect multi-spectral images of farmland covering the farmland area through satellite remote sensing technology, and to pre-process the multi-spectral images of the farmland to obtain a three-dimensional spatial image of the farmland; A data analysis module (200) performs pest and disease analysis on the three-dimensional spatial image of the farmland to extract pest and disease images and their spatial distribution, wherein the pest and disease images include at least one target type of pest image; The trend prediction module (300) analyzes historical pest data based on an AI algorithm analysis model, obtains pest inducing impact factors corresponding to different target pest types, assigns probability values ​​to pest disaster levels based on the pest inducing impact factors, performs statistics based on the probability values ​​of pest disaster levels, and obtains pest development trends corresponding to target pest types; The early warning decision module (400) compiles a smart contract for the degree of pest impact corresponding to the current pest image, records the timestamp of the degree of pest impact in the smart contract, and evaluates the actual pest impact by considering the transmission time, thereby generating evaluation reports of different time limits, and establishing pest early warning strategies based on the evaluation reports of different time limits.

2. The crop pest monitoring and early warning system based on satellite remote sensing and AI algorithm according to claim 1, characterized in that: Multiple satellites are used to regularly acquire farmland multispectral images covering multiple bands, wherein the farmland multispectral images include b1 band spectral images, b2 band spectral images, b3 band spectral images and b4 band spectral images; the farmland multispectral images are segmented and the farmland planting areas are extracted through spectral feature analysis.

3. The crop pest monitoring and early warning system based on satellite remote sensing and AI algorithm according to claim 2 is characterized by: The logic for acquiring the three-dimensional spatial image of farmland is as follows: Perform geometric positioning, edge trimming and noise removal on the farmland planting area to obtain the initial processed farmland image; Searching for at least one fixed marker in the initially processed farmland image, marking the fixed marker as a reference point, and performing spatial alignment based on the reference point to obtain a reference farmland image; The reference farmland image is grayed out using a weighted average method, the Rank transformation result of the gray image is used as a matching primitive, and a region matching algorithm based on normalized absolute difference and a measurement function is used to obtain a dense disparity map of the scene; The spatial coordinates of the scene are calculated based on the principle of parallel binocular vision imaging, and a three-dimensional point cloud map is generated; a three-dimensional spatial image of the farmland is generated based on the three-dimensional point cloud map.

4. The crop pest and disease monitoring and early warning system based on satellite remote sensing and AI algorithm according to claim 3 is characterized by: The acquisition logic of the pest and disease image is: Collect the measured species grayscale values ​​of the species corresponding to the three-dimensional spatial image of the farmland, and compare the measured species grayscale values ​​with the species reference grayscale interval corresponding to the same time series information, wherein the species reference grayscale interval is the grayscale pixel interval value of the normal growth of the current species in the experimental planting environment obtained based on the analysis of the historical database; Screening pixel blocks whose measured species grayscale values ​​are not within the species reference grayscale interval, treating the pixel blocks that are not within the species reference grayscale interval as abnormal image blocks, and marking the spatial position corresponding to the abnormal image block as an abnormal spatial position; According to the pre-set continuous time series interval, the abnormal support and abnormal co-occurrence corresponding to the abnormal spatial position in the continuous time series interval are counted, wherein the abnormal support is used to indicate the total number of abnormal marks corresponding to the same abnormal spatial position in the continuous time series interval, and the abnormal co-occurrence is used to indicate the number of times different abnormal spatial positions in the continuous time series interval co-occur; The continuous time series interval is a continuous timestamp of the multispectral images of farmland collected by satellites; when the anomaly support in the continuous time series interval is greater than a preset anomaly support threshold and the anomaly co-occurrence is greater than a preset anomaly expansion threshold, the corresponding abnormal image block is marked as a pest image, and the pest image includes at least one target type pest image and the spatial distribution of the corresponding target type pest.

5. The crop pest monitoring and early warning system based on satellite remote sensing and AI algorithm according to claim 4 is characterized by: The logic for acquiring the target type pest image: Performing texture analysis on the pest image, extracting pest texture features through a gray-level co-occurrence matrix, and obtaining a pest texture feature correlation amount; The similarity is calculated based on the normalized calculation formula according to the pest texture feature correlation quantity and the pest texture feature defined in the historical database, and the abnormal image blocks with similarity within the pest similarity threshold range are screened out, and the abnormal image blocks are marked as target species pest images.

6. The crop pest monitoring and early warning system based on satellite remote sensing and AI algorithm according to claim 5, characterized in that: The logic for obtaining the development trend of pests and diseases is as follows: The target pest images corresponding to different types of pests and diseases are matrixed according to the abnormal spatial positions to obtain the spatial distribution corresponding to the target pest images; Construct an AI algorithm analysis model, using meteorological data, images of different target species of pests and their spatial distribution as input factors, and output the induced pest impact factors corresponding to the images of different target species of pests; The probability of pest disaster level is assigned based on the pest-inducing impact factor. The larger the probability value, the greater the probability of pest disaster. The probability assignment of different types of pests to the pest disaster level is superimposed to obtain the pest impact degree corresponding to the current pest image, and the pest development trend is characterized based on the pest impact degree.

7. The crop pest monitoring and early warning system based on satellite remote sensing and AI algorithm according to claim 6, characterized in that: The acquisition logic of the pest and disease early warning strategy is: Obtain the spatial distribution and collection timestamps corresponding to different types of target pests; Store the current degree of impact of pests and diseases based on timestamps, and map the historical pest and disease records corresponding to the target pest types based on timestamps; Based on the analysis of historical pest and disease records, and combined with the transmission time, the actual impact of pests and diseases is evaluated, and assessment reports of different time limits are recorded. The timestamps are rewritten according to different time limits, and displayed in the pest and disease early warning strategy corresponding to the smart contract.

8. A crop pest and disease monitoring and early warning method based on satellite remote sensing and AI algorithm, based on the implementation of a crop pest and disease monitoring and early warning system based on satellite remote sensing and AI algorithm according to any one of claims 1 to 7, characterized in that: The following steps are involved: The multispectral images of farmland covering the farmland area are collected regularly through satellite remote sensing technology, and the multispectral images of farmland are preprocessed to obtain the three-dimensional spatial images of farmland; Performing pest and disease analysis on the three-dimensional spatial image of the farmland to extract pest and disease images and their spatial distribution, wherein the pest and disease images include at least one target type of pest image; Analyze historical pest data based on AI algorithm analysis model, obtain the pest inducing factors corresponding to different target pest types, assign probability values ​​to pest disaster levels based on the pest inducing factors, perform statistics based on the probability values ​​of pest disaster levels, and obtain pest development trends corresponding to target pest types; The degree of pest and disease impact corresponding to the current pest and disease image is written into a smart contract, the timestamp of the degree of pest and disease impact is recorded in the smart contract, and the actual pest and disease impact is evaluated considering the transmission time, so as to generate evaluation reports of different time limits, and establish a pest and disease early warning strategy based on the evaluation reports of different time limits.

9. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the crop disease and pest monitoring and early warning system based on satellite remote sensing and AI algorithm as described in any one of claims 1 to 7 by calling the computer program stored in the memory.

10. A computer-readable storage medium, characterized in that: Instructions are stored, and when the instructions are executed on a computer, the computer executes a crop disease and insect pest monitoring and early warning system based on satellite remote sensing and AI algorithm as described in any one of claims 1 to 7.

Images