Barn pest monitoring and early warning method, system and equipment based on big data and medium

Abstract

The invention discloses a barn pest monitoring and early warning method, system and device based on big data and a medium, particularly relates to the technical field of pest monitoring, and is used for solving the problem that the distribution trend of newly appearing pests in a barn cannot be accurately monitored in the prior art. The method comprises the following steps: dividing a granary into a plurality of monitoring areas, arranging an independent spectrum sensor network, carrying out real-time monitoring based on a spectrum analysis technology, carrying out contrastive analysis with a granary pest library, rapidly identifying new pests, and after the new pests are monitored, generating a distribution trend chart in real time through a spectrum imaging technology, according to the method, a trend chart is obtained, whether early warning needs to be sent out is judged according to the trend chart, and the data quality can be mastered more comprehensively by evaluating the spectral distortion hidden danger and randomness of spectral data in a high-risk monitoring area, so that the accuracy evaluation of the distribution trend chart of newly-emerged pests is improved, and the reliability and accuracy of a granary pest monitoring and early warning system are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of pest monitoring technology, and more specifically, to a method, system, equipment and medium for monitoring and early warning of granary pests based on big data. Background Art

[0002] Granaries are important places for grain storage, so their management and maintenance become particularly important. However, pest problems in granaries have always been a major challenge in the grain storage process. Pests not only directly destroy stored grain and cause economic losses, but also spread pathogens, threatening food security and human health. Traditional granary pest monitoring methods usually rely on manual inspection and physical capture, which is not only time-consuming and labor-intensive, but also has the risk of missed detection and cannot capture the dynamic changes of pests in a timely and accurate manner. Pest monitoring methods based on spectral imaging have gradually become a research hotspot. Pest monitoring based on spectral imaging can identify and monitor pests by capturing the spectral characteristics of pests at specific wavelengths.

[0003] In pest monitoring based on spectral imaging, if the distribution trends of newly emerged pests in granaries cannot be accurately monitored, the monitoring system will not be able to accurately locate the pest activity area and predict the trend, which will affect the reliability and timeliness of the overall monitoring and early warning system. It may also lead to a slow response to outbreaks of newly emerged pests and the wrong deployment of prevention and control measures, making it impossible to effectively suppress the invasion of granaries by newly emerged pests.

[0004] In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a granary pest monitoring and early warning method, system, equipment and medium based on big data to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions: The granary pest monitoring and early warning method based on big data includes the following steps: S1: Divide the granary into multiple monitoring areas, each equipped with an independent spectral sensor network; S2: Real-time monitoring of each monitoring area is carried out based on spectral analysis technology, and by comparing it with the granary pest library, it is determined whether new pests appear in the monitoring area; S3: When new pests appear in the monitoring area, the distribution of the new pests is monitored based on spectral imaging technology, and a distribution trend map of the new pests is generated in real time. The distribution trend map of the new pests is used to determine whether to issue an early warning; S4: Mark the monitoring areas where new pests appear and their adjacent monitoring areas as high-risk monitoring areas; S5: For high-risk monitoring areas: By analyzing the symmetry of pest spectral images at specific wavelengths, evaluate the potential risk of spectral distortion of pest spectral data; use the principle of information entropy to analyze the entropy value of pest spectral data and evaluate the randomness of pest spectral data; S6: Based on a comprehensive analysis of the potential risk of spectral distortion and the randomness of pest spectral data, the accuracy of spectral analysis in high-risk monitoring areas is evaluated, and a determination is made as to whether the accuracy of the distribution trend map of emerging pests meets the standards.

[0007] In a preferred embodiment, each monitoring area is monitored in real time based on spectral analysis technology, and by comparing with the granary pest library, it is determined whether new pests appear in the monitoring area, specifically including: S201: Spectral sensors deployed in each monitoring area capture pest spectral data in real time; S202: performing noise filtering and feature enhancement on the collected pest spectral data, and automatically extracting key spectral features; S203: comparing the preprocessed pest spectral data with the known pest spectral fingerprints in the granary pest library, and using a spectral similarity matching algorithm to preliminarily screen out the pest type that is most similar to the current spectral data; S204: When a new spectral pattern that does not match the granary pest library is identified in the pest spectral data, it is marked as a new pest, and its spectral characteristics and the area information of the first appearance are recorded.

[0008] In a preferred embodiment, when new pests appear in the monitoring area, the distribution of the new pests is monitored based on spectral imaging technology, and a distribution trend map of the new pests is generated in real time. Whether to issue an early warning is determined based on the distribution trend map of the new pests, specifically: S301: collecting distribution spectrum data of newly emerged pests in real time; based on the distribution spectrum data of newly emerged pests in continuous time periods, applying a time series analysis method to generate a dynamic distribution trend diagram of newly emerged pests; S302: Performing pattern recognition analysis on the generated distribution trend graph, combining pest activity hotspots and distribution density changes, to determine whether a warning threshold has been reached; S303: When the analysis result of the distribution trend graph exceeds the preset risk threshold, the system automatically triggers an early warning and sends an alert to the management personnel.

[0009] In a preferred embodiment, S301 specifically includes: Continuously collect spectral data of newly emerged pests at predetermined time intervals; The continuous spectral data of emerging pests are arranged in chronological order to construct a time series data set; the data at each time point include the spectral characteristics of emerging pests at various wavelengths and the spatial distribution of emerging pests in the monitoring area; The time series analysis algorithm is used to analyze the spectral distribution data of new pests at different time points to identify the change pattern of new pests in the time dimension; Based on the analysis results, a dynamic distribution trend map of emerging pests is generated to show the diffusion path, diffusion speed and spatial concentration of emerging pests in the monitoring area.

[0010] In a preferred embodiment, for high-risk monitoring areas, the symmetry of the pest spectral image at a specific wavelength is analyzed to assess the degree of spectral distortion risk of the pest spectral data, specifically: Collect spectral image data of pests in high-risk monitoring areas; The collected spectral image is standardized; the standardized spectral image is subjected to a two-dimensional discrete Fourier transform at a specific wavelength to calculate a complex matrix in the frequency domain; Extracting a phase spectrum from the complex matrix in the frequency domain after Fourier transformation, the phase spectrum represents the geometric structure information of the image; Symmetry analysis of the phase spectrum: compare the phase values ​​of the central symmetric points in the frequency domain and calculate the overall symmetry deviation of the spectrum image; The symmetry of the spectral image at different scales is analyzed by performing scale-space transformation on the phase spectrum. The symmetry calculation steps are repeated for each scale, and the results are weighted combined to obtain a comprehensive index of multi-scale symmetry deviation.

[0011] In a preferred embodiment, for high-risk monitoring areas, the entropy value of pest spectral data is analyzed using the information entropy principle to evaluate the randomness of the pest spectral data, specifically: The edge area is extracted from the spectral image at each wavelength by using the edge detection algorithm; The spectral image is divided into two parts: edge area and non-edge area; For the spectral data in the edge area, the information entropy formula is applied to calculate the edge area entropy value; For the spectral data in the non-edge area, the information entropy formula is applied to calculate the entropy value of the non-edge area; The edge entropy difference is calculated by the difference between the edge area entropy value and the non-edge area entropy value; The edge entropy difference is normalized to obtain the standardized edge entropy difference.

[0012] In a preferred embodiment, based on a comprehensive analysis of the degree of spectral distortion risk and the randomness of the spectral data of pests, the accuracy of the spectral analysis in the high-risk monitoring area is evaluated, and the accuracy of the distribution trend map of new pests is determined to be up to standard, specifically: The multi-scale symmetry deviation comprehensive index and the standardized edge entropy difference are normalized, and the normalized multi-scale symmetry deviation comprehensive index and the standardized edge entropy difference are assigned preset proportional coefficients respectively, and the pest spectral monitoring accuracy evaluation coefficient is calculated; Set the spectral accuracy threshold corresponding to the pest spectral monitoring accuracy assessment coefficient, and compare the pest spectral monitoring accuracy assessment coefficient with the spectral accuracy threshold: When the pest spectral monitoring accuracy assessment coefficient is less than or equal to the spectral accuracy threshold, it is determined that an accurate pest spectral monitoring signal is generated, and it is determined that the accuracy of the distribution trend diagram of the newly emerged pests meets the standard; When the pest spectral monitoring accuracy assessment coefficient is greater than the spectral accuracy threshold, it is determined that a pest spectral monitoring accurate hidden danger warning signal is generated, and it is determined that the accuracy of the distribution trend map of new pests does not meet the standard.

[0013] On the other hand, the present invention provides a granary pest monitoring and early warning system based on big data, including a granary area division module, a new pest monitoring module, a distribution trend early warning module, a risk area marking module, a risk area analysis module and a trend accurate judgment module; Granary area division module: divides the granary into multiple monitoring areas, each of which is equipped with an independent spectral sensor network; New pest monitoring module: Based on spectral analysis technology, each monitoring area is monitored in real time, and by comparing with the granary pest library, it is determined whether new pests appear in the monitoring area; Distribution trend warning module: When new pests appear in the monitoring area, the distribution of the new pests is monitored based on spectral imaging technology, and a distribution trend map of the new pests is generated in real time. The distribution trend map of the new pests is used to determine whether to issue a warning; Risk area marking module: marking the monitoring area where new pests appear and its adjacent monitoring areas as high-risk monitoring areas; Risk area analysis module: For high-risk monitoring areas: By analyzing the symmetry of pest spectral images at specific wavelengths, the degree of spectral distortion risk of pest spectral data is evaluated; the entropy value of pest spectral data is analyzed using the principle of information entropy to evaluate the randomness of pest spectral data; Trend accuracy judgment module: Based on a comprehensive analysis of the degree of spectral distortion risk and the randomness of pest spectral data, the accuracy of spectral analysis in high-risk monitoring areas is evaluated, and it is determined whether the accuracy of the distribution trend map of new pests meets the standards.

[0014] On the other hand, the present invention provides a device, comprising: a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements a granary pest monitoring and early warning method based on big data.

[0015] On the other hand, the present invention provides a computer-readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, a granary pest monitoring and early warning method based on big data is implemented.

[0016] The technical effects and advantages of the granary pest monitoring and early warning method, system, equipment and medium based on big data of the present invention are as follows: By dividing the granary into multiple monitoring areas and equipping an independent spectral sensor network, accurate monitoring and management of each area can be achieved, effectively improving the accuracy and response speed of pest monitoring. Real-time monitoring based on spectral analysis technology and comparative analysis with the granary pest library can quickly identify new pests and ensure that potential threats are discovered at an early stage. When new pests are monitored, a distribution trend map is generated in real time through spectral imaging technology, and it is determined whether an early warning needs to be issued based on the trend map, thereby enhancing the timeliness and accuracy of the early warning system. In addition, by conducting a more in-depth analysis of the spectral data in the high-risk monitoring area, especially evaluating the spectral distortion hazards and randomness of the spectral data, the data quality can be more comprehensively grasped, thereby improving the accuracy assessment of the distribution trend map of new pests and preventing misjudgment or omission. Combining the above measures, the present invention can significantly improve the reliability and accuracy of the granary pest monitoring and early warning system, effectively reduce the risks and losses caused by pest outbreaks, and provide a strong technical guarantee for food storage safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the granary pest monitoring and early warning method based on big data of the present invention; Figure 2 It is a structural schematic diagram of the granary pest monitoring and early warning system based on big data of the present invention. DETAILED DESCRIPTION

[0018] 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.

[0019] Example 1

[0020] Figure 1 The present invention provides a method for monitoring and early warning of granary pests based on big data, which includes the following steps: S1: The granary is divided into multiple monitoring areas, each of which is equipped with an independent spectral sensor network.

[0021] S2: Real-time monitoring of each monitoring area is carried out based on spectral analysis technology, and by comparing it with the granary pest library, it is determined whether new pests appear in the monitoring area.

[0022] S3: When new pests appear in the monitoring area, the distribution of the new pests is monitored based on spectral imaging technology, and a distribution trend map of the new pests is generated in real time. The distribution trend map of the new pests is used to determine whether to issue an early warning.

[0023] S4: The monitoring areas where new pests appear and their adjacent monitoring areas will be marked as high-risk monitoring areas.

[0024] S5: For high-risk monitoring areas: By analyzing the symmetry of the pest spectral image at a specific wavelength, the degree of spectral distortion risk of the pest spectral data is evaluated; using the principle of information entropy to analyze the entropy value of the pest spectral data, the randomness of the pest spectral data is evaluated.

[0025] S6: Based on a comprehensive analysis of the potential risk of spectral distortion and the randomness of pest spectral data, the accuracy of spectral analysis in high-risk monitoring areas is evaluated, and a determination is made as to whether the accuracy of the distribution trend map of emerging pests meets the standards.

[0026] The granary pest monitoring and early warning method fully reflects the advantages and characteristics of big data through the implementation of multiple steps. First, in step S1, the granary is divided into multiple monitoring areas and equipped with an independent spectral sensor network, which can collect a large amount of spectral data in real time. Then, steps S2 to S4 use big data analysis technology to compare, monitor and trend the data of each area, identify new pests and dynamically adjust the key monitoring areas. Steps S5 and S6 achieve accurate monitoring and early warning of high-risk areas by conducting in-depth analysis of the complexity, randomness and distortion risks of a large amount of spectral data. The entire process relies on the real-time collection, analysis and comprehensive judgment of big data, thereby improving the accuracy and timeliness of monitoring and ensuring that the granary pest early warning system can efficiently respond to potential risks.

[0027] The granary is divided into multiple monitoring areas, each of which is equipped with an independent spectral sensor network, specifically: It is divided into several monitoring areas. The size and shape of each monitoring area should be optimized according to the actual layout and specific needs of the granary.

[0028] When dividing a granary into multiple monitoring areas, the following factors can be considered to determine the boundaries and characteristics of each area: Internal structure of the silo: The silo is divided according to its physical layout and structural characteristics. For example, different storage units, shelves, aisles, and entrance and exit locations can serve as natural partition boundaries.

[0029] Grain type and storage method: Different types of grain may attract different types of pests. By designating an area where the same or similar types of grain are stored as a monitoring area, you can monitor specific pests in a targeted manner.

[0030] Differences in environmental conditions: Consider differences in environmental factors such as temperature, humidity, light and ventilation. Environmental conditions have a significant impact on the activity and reproduction of pests, so areas with similar environmental conditions can be divided into one monitoring area.

[0031] Historical pest activity data: Analyze hot spots of pest activity in history. Identify areas with high pest activity through historical data and classify them as high-risk monitoring areas.

[0032] Logistics and human flow: The material handling routes and personnel flow in the granary are also important factors in the spread of pests. Areas can be divided according to high-frequency areas of logistics and human flow to more effectively monitor possible pest transmission routes.

[0033] Technical parameters of monitoring equipment: Consider the coverage and monitoring capabilities of the spectral sensor. Determine the size of the area that each sensor can effectively monitor based on the technical specifications of the equipment (such as sensing distance, angle, and sensitivity).

[0034] Select a suitable spectral sensor based on monitoring requirements. The sensor should cover the wavelength range required for pest spectral fingerprinting and have sufficient resolution and sensitivity.

[0035] Design the spectral sensor network to ensure that the sensors in each monitoring area can fully cover their monitoring range. The network design should consider the optimal layout and signal coverage between sensors.

[0036] Install a spectral sensor in each divided monitoring area. During the installation process, the angle and position of the sensor need to be adjusted to ensure that the spectral data in the target area can be effectively captured.

[0037] Connect all spectral sensors to the central monitoring system. Set up network synchronization to ensure that the data collected from each sensor can be transmitted to the data processing center in real time and accurately.

[0038] Based on spectral analysis technology, each monitoring area is monitored in real time, and by comparing with the granary pest library, it is determined whether new pests appear in the monitoring area, including: S201: Spectral sensors deployed in each monitoring area capture pest spectral data in real time, and the system automatically adjusts acquisition parameters to adapt to different ambient light conditions to ensure data quality.

[0039] In each monitoring area, a wavelength range that covers the spectral characteristics of pests is selected.

[0040] Among them, the spectral sensor continuously collects pest spectral data at a high frequency, and the pest spectral data is the spectral data of pest activities. The system monitors the ambient light conditions (such as light intensity and angle changes) in real time and automatically adjusts the sensor's acquisition parameters, such as exposure time and gain control, to ensure that the collected data is of high quality and stability.

[0041] S202: Noise filtering and feature enhancement are performed on the collected pest spectral data, and key spectral features are automatically extracted to prepare for subsequent comparative analysis: Apply noise filtering algorithms, such as adaptive filtering or wavelet transform, to the collected pest spectral data to remove random noise caused by sensor noise or environmental interference and ensure signal purity.

[0042] At the same time, in view of possible spectral baseline drift, a baseline correction method is used to restore the accuracy of the spectral signal.

[0043] Use spectral feature enhancement techniques, such as derivative spectral analysis and smooth difference technology, to improve the resolution and contrast of spectral data and highlight the significance of pest characteristic bands.

[0044] Extract key spectral feature points, such as absorption peaks and reflection valleys at specific wavelengths, for subsequent pest identification and comparison.

[0045] S203: Compare the pre-processed pest spectral data with the known pest spectral fingerprints in the granary pest library: The pre-processed pest spectral data are compared with the standard spectral fingerprints stored in the granary pest library. The pest library stores the standard spectral features of known pests, covering the spectral data of various pest types and their different life cycles.

[0046] A spectral similarity matching algorithm, such as cosine similarity or Euclidean distance, is used to preliminarily screen out the pest type that is most similar to the current spectral data.

[0047] S204: When a new spectral pattern that does not match the granary pest library is identified in the pest spectral data, it is marked as a new pest, and its spectral characteristics and the area where it first appeared are recorded: The results of each spectral matching are analyzed, and when the spectral data are identified to not match any known pest spectral fingerprint in the pest library, the data are marked as potential new pests.

[0048] Further validation of the spectral data of emerging pests, including repeated sampling and analysis, ensures the reliability of the detection results of emerging pests.

[0049] Once the presence of new pests is confirmed, their detailed spectral characteristics are automatically recorded, including absorption peaks at key wavelengths, reflection valleys, and the overall shape of the spectral graph.

[0050] At the same time, record the regional information where the new pest first appeared, including time, environmental conditions, etc., as a reference for subsequent analysis and treatment.

[0051] Among them, the spectral fingerprint of pests refers to the unique spectral characteristics of a certain pest within a specific spectral range. Specifically, when light shines on the body surface of the pest, light of different wavelengths will be reflected, absorbed or transmitted by the body surface of the pest in different ways. These reflection or absorption characteristics collected by the spectral sensor appear as a set of specific absorption peaks and reflection valleys on the spectrum graph. The combination of these characteristics is the fingerprint of the pest. Each pest has a unique spectral fingerprint due to its different physiological structure and chemical composition. This uniqueness makes it possible to identify specific pests through spectral analysis technology.

[0052] The Granary Pest Database is a database system that stores and manages information on various known pests. The database includes basic biological information of pests that have appeared in the granary in history (such as species, life history, habitat, etc.), as well as spectral data in different spectral bands, i.e. spectral fingerprint information. The main function of the Granary Pest Database is to provide reference data for the spectral analysis system, and to help identify and confirm the types of pests and monitor new pests by comparing with the real-time monitored spectral data.

[0053] The standard spectral fingerprint in the granary pest library refers to the typical spectral response characteristics recorded for each known pest in the granary pest library. These standard fingerprints are obtained through multiple sampling and averaging under experimental conditions, and represent the most common spectral characteristics of the pest under specific environmental conditions. The standard spectral fingerprint is used to compare with the real-time monitoring data to determine whether the monitored pest is a known species. If the monitored spectral data matches the standard spectral fingerprint in the granary pest library, the pest is identified as a known species; if it does not match, it may be a new pest.

[0054] When new pests appear in the monitoring area, the distribution of new pests is monitored based on spectral imaging technology, and a distribution trend map of the new pests is generated in real time. The distribution trend map of the new pests is used to determine whether to issue an early warning. Specifically: S301: Real-time collection of distribution spectrum data of newly emerged pests; based on the distribution spectrum data of newly emerged pests in continuous time periods, a time series analysis method is applied to generate a dynamic distribution trend diagram of newly emerged pests to reflect the spread or concentration trend of pests: After confirming the emergence of new pests in the monitoring area, the spectral data of the new pests will be continuously collected at the predetermined time interval to ensure that the distribution information of the new pests at different time points can be captured. The data collection frequency is dynamically adjusted according to the intensity of the new pest activity and environmental changes to capture high-frequency activity data; the predetermined time interval is set according to actual needs.

[0055] The continuous spectral data of emerging pests obtained from the spectral imaging equipment are arranged in chronological order to construct a time series data set. The data at each time point includes the spectral characteristics of emerging pests at various wavelengths and their spatial distribution locations in the monitoring area.

[0056] Time series analysis algorithms (such as autoregressive integrated moving average model, long short-term memory network, etc.) are used to analyze the spectral distribution data of new pests at different time points and identify their changing patterns in the time dimension.

[0057] Based on the analysis results, a dynamic distribution trend map of emerging pests is generated to show the diffusion path, diffusion speed and spatial concentration of emerging pests in the monitoring area.

[0058] Update the distribution trend chart of emerging pests in real time, integrate the newly collected data into the existing trend analysis, and dynamically reflect the latest changes in the distribution of emerging pests.

[0059] The distribution trend of emerging pests is displayed in a visual form, making it easier for managers to understand and respond quickly.

[0060] S302: Perform pattern recognition analysis on the generated distribution trend graph, and determine whether the warning threshold is reached by combining the pest activity hotspots and distribution density changes: Use machine learning and pattern recognition algorithms (such as K-means clustering, support vector machine, etc.) to analyze the generated distribution trend map of emerging pests and identify hot spots and high-risk points of emerging pest activities.

[0061] Analyze changes in the distribution density of emerging pests and identify trends of concentrated outbreaks or spread to determine the severity of emerging pest activities.

[0062] The system presets different levels of new pest risk thresholds (such as low risk, medium risk and high risk) based on historical data and professional technicians in this field.

[0063] Compare the analysis results of the current distribution trend chart of emerging pests with the preset thresholds to evaluate the speed, area and density of the spread of emerging pests and determine whether they exceed the safety range.

[0064] S303: When the distribution trend chart analysis results exceed the preset risk threshold, the system automatically triggers an early warning and sends an alert to the management personnel, prompting them to take countermeasures: When the analysis results of the distribution trend chart of emerging pests show that the distribution of emerging pests exceeds the set risk threshold, the system automatically triggers the early warning mechanism. The early warning level is automatically classified according to the degree of exceeding the threshold, such as red alert (high risk), orange alert (medium risk), and yellow alert (low risk).

[0065] The system promptly transmits early warning information of emerging pests to relevant managers through various channels (such as text messages, emails, alarms, etc.). The alert content includes the current distribution status of emerging pests, estimated spread trends, potentially affected areas, and recommended countermeasures.

[0066] Based on the warning level, the system provides specific response measures, such as strengthening monitoring, activating pest control equipment, or closing specific areas, etc. Managers can take prompt action based on the recommendations.

[0067] The system also enters monitoring mode, tracking the effectiveness of response measures in real time and adjusting warnings and recommendations for emerging pests based on the latest data.

[0068] The monitoring areas where new pests appear and their adjacent monitoring areas are marked as high-risk monitoring areas: When new pests are detected in a monitoring area, the monitoring area is immediately marked as a high-risk monitoring area.

[0069] At the same time, the monitoring areas adjacent to this high-risk monitoring area will also be automatically marked as high-risk monitoring areas. This is because new pests may affect adjacent areas through migration or spread, thereby increasing the risk of these areas.

[0070] Pests have a tendency to spread naturally, especially in closed or semi-closed environments such as granaries, where they may quickly spread to adjacent areas through air movement, human activities, and grain handling.

[0071] By marking adjacent monitoring areas as high-risk areas, it is possible to prevent new pests from spreading outward from the initial area, reduce monitoring blind spots, and improve the coverage and response speed of the entire monitoring system.

[0072] For high-risk monitoring areas, the symmetry of pest spectral images at specific wavelengths is analyzed to assess the potential spectral distortion risk of pest spectral data. Specifically: In high-risk monitoring areas, high-resolution spectral imaging equipment is used to collect spectral image data of pests at multiple predetermined key wavelengths (such as 450nm, 650nm, and 850nm). This ensures that the data is minimally disturbed by ambient light during collection and that the captured images cover the main areas where pests are distributed.

[0073] The collected spectral images are standardized, including removing background noise, correcting possible light source inhomogeneity, and smoothing the images. The purpose is to ensure the purity and consistency of Fourier transform input data to improve the reliability of the analysis results.

[0074] The standardized spectral image is subjected to a two-dimensional discrete Fourier transform (2D-DFT) at a specific wavelength, and a complex matrix in the frequency domain is calculated.

[0075] The Fourier transform formula is as follows: .

[0076] in, Indicates the specific wavelength to which the spectral image corresponds, usually in nanometers (nm).

[0077] and They represent the horizontal and vertical frequency components of the spectral image in the frequency domain after Fourier transform, and the unit is usually cycles / unit length.

[0078] and They represent the horizontal and vertical position coordinates of the spectral image in the spatial domain, respectively, in pixels.

[0079] and Respectively represent the width and height of the spectral image, in pixels.

[0080] Indicated in wavelength Next, the spectral image is at position The spectral intensity value at .

[0081] Represents the spectrum image at wavelength In the following figure, the complex value (including amplitude and phase information) in the frequency domain after the two-dimensional Fourier transform.

[0082] is the basis function in Fourier transform, which represents the transformation relationship between frequency domain and spatial domain. Is an imaginary unit.

[0083] Extract the phase spectrum from the complex matrix in the frequency domain after Fourier transform. The phase spectrum represents the geometric structure information of the image: .

[0084] Indicated in wavelength The phase spectrum of the spectral image below reflects the phase information of the image in the frequency domain.

[0085] Indicates the phase angle (phase value) of the complex matrix after Fourier transformation.

[0086] Symmetry analysis of the phase spectrum is performed. The specific method is to compare the phase values ​​of the central symmetric points in the frequency domain, such as comparing and , and calculate the sum of the squares of these differences: ; Here Values ​​represent wavelength The overall symmetry deviation of the lower spectral image. A value of 0 indicates significant asymmetry in the spectral image, which may indicate the presence of spectral distortion.

[0087] Indicated in wavelength In the lower frequency domain Symmetrical position The phase value at .

[0088] It means to sum the symmetrical points from the center to the boundary in the frequency domain and calculate the phase difference of each symmetrical point respectively.

[0089] In order to further enhance the accuracy of the analysis, multi-scale analysis is introduced. By performing scale space transformation (such as wavelet transform) on the phase spectrum, the symmetry of the spectral image at different scales (frequency levels) is analyzed. For each scale, the symmetry calculation steps are repeated and the results are weighted combined: .

[0090] It is a comprehensive index of multi-scale symmetry deviation.

[0091] Indicates the scale number, from 1 to .

[0092] Indicates the number of scale levels used for analysis, which is usually set according to the image resolution and analysis accuracy requirements.

[0093] Indicates The weight of each scale is usually set according to the degree of influence of different scales on the analysis results.

[0094] Indicated in The symmetry deviation value is obtained by the phase symmetry analysis formula on each scale.

[0095] The larger the multi-scale symmetry deviation comprehensive index is, the higher the potential spectral distortion of pest spectral data is. This is because the larger This reflects that at different scales and frequency levels, the spectral data has obvious structural asymmetry or distortion, which may be caused by environmental interference, optical equipment problems or other external factors during the spectral data acquisition process. Such distortion may affect the accuracy of spectral analysis and lead to misjudgment or misclassification.

[0096] By analyzing the spectral data of pests in high-risk monitoring areas, the technology can effectively identify and quantify potential distortion hazards in spectral images. By introducing Fourier phase spectrum and multi-scale analysis technology, the structural asymmetry of spectral data can be accurately detected at multiple frequency levels, thereby revealing spectral distortions that may be caused by environmental interference, optical equipment problems or other external factors. A large multi-scale symmetry deviation composite index provides clear instructions to help quickly identify spectral data anomalies in high-risk areas, avoid misjudgment or misclassification, and improve the accuracy and reliability of overall spectral analysis.

[0097] For high-risk monitoring areas, the entropy value of pest spectral data is analyzed using the principle of information entropy to evaluate the randomness of pest spectral data. Specifically: The edge region is extracted from the spectral image at each wavelength by edge detection algorithms (such as Sobel or Canny edge detection). The purpose of edge detection is to determine the regions with significant changes in the spectral image, which usually contain important feature information.

[0098] The spectral image is divided into two parts: edge region and non-edge region. The edge region contains the pixels identified during edge detection, while the non-edge region contains the rest of the image.

[0099] For the spectral data in the edge area, the information entropy formula is applied to calculate the edge area entropy value, which is expressed as: ;in, is the edge area entropy value, Indicates the edge area Pixel value The probability of appearing in the edge area, is the total number of possible pixel values ​​in the edge area, The number of pixel values ​​in the edge area.

[0100] Similarly, for the spectral data in the non-edge area, the information entropy formula is applied to calculate the entropy value of the non-edge area, and its expression is: ;in, is the entropy value of the non-edge area, In the non-edge area, Pixel value The probability of appearing in the non-edge area, is the total number of possible pixel values ​​in the non-edge area, It is the number of pixel values ​​in the non-edge area.

[0101] The edge entropy difference is calculated by the difference between the edge area entropy value and the non-edge area entropy value: ;in, is the marginal entropy difference.

[0102] The edge entropy difference indicates the difference in information complexity between edge regions and non-edge regions. A larger edge entropy difference means that the information complexity of edge regions is significantly higher than that of non-edge regions, indicating that the image contains important and complex spectral features.

[0103] In order to compare between different spectral images, the edge entropy difference is normalized to obtain the standardized edge entropy difference: ;in, is the normalized marginal entropy difference.

[0104] Therefore, the smaller the normalized edge entropy difference is, the higher the randomness of the pest spectral data is. This is because when the edge entropy difference is small, the entropy value difference between the edge area and the non-edge area is not significant, indicating that the overall information density of the spectral image is low, and there may be a lot of noise or random signals. In this case, the feature information in the spectral image is relatively scattered, and it is difficult to extract effective identification features, which may lead to a decrease in the accuracy of pest identification. Therefore, as an important indicator for evaluating the quality and complexity of spectral data, the normalized edge entropy difference can effectively warn of potential random problems in the data and help take timely measures to ensure the reliability of monitoring and early warning.

[0105] By using the standardized edge entropy difference to evaluate the quality of pest spectral data in high-risk monitoring areas, this method can sensitively detect the differences in information complexity and randomness in spectral data. The size of the standardized edge entropy difference directly reflects the fidelity and noise interference of key feature areas in the spectral image. A larger edge entropy difference indicates that there is significant and complex feature information in the spectral image, which helps to improve the accuracy of pest identification; a smaller edge entropy difference indicates that the data may be affected by random noise, which requires attention and further processing, which can improve the reliability of spectral data analysis and enhance the effectiveness of the granary pest monitoring and early warning system.

[0106] Based on a comprehensive analysis of the potential spectral distortion risk and randomness of pest spectral data, the accuracy of spectral analysis in high-risk monitoring areas is evaluated, and the accuracy of the distribution trend map of emerging pests is determined to be up to standard. Specifically: The multi-scale symmetry deviation comprehensive index and the standardized edge entropy difference are normalized, and the normalized multi-scale symmetry deviation comprehensive index and the standardized edge entropy difference are respectively assigned preset proportional coefficients to calculate the accurate assessment coefficient of pest spectral monitoring.

[0107] The specific implementation method of calculating the accurate assessment coefficient of pest spectral monitoring is not specifically limited here. Any method that can calculate the accurate assessment coefficient of pest spectral monitoring by assigning preset proportional coefficients to the normalized multi-scale symmetry deviation comprehensive index and the standardized edge entropy difference respectively is sufficient. In order to realize the technical solution of the present invention, the present invention provides a specific method for calculating the accurate assessment coefficient of pest spectral monitoring.

[0108] Set two preset scale factors and ,in is the preset proportionality coefficient for normalizing the composite index of multi-scale symmetry deviation, is the preset proportional coefficient for normalizing the normalized edge entropy difference. The preset proportional coefficient can be adjusted according to the actual application scenario or empirical data so that the weights of the two indicators in the total evaluation coefficient reflect their relative importance.

[0109] The expression of the accuracy evaluation coefficient of pest spectral monitoring is: ;in, Accurate evaluation coefficient for pest spectral monitoring, and are the preset proportional coefficients of the multi-scale symmetry deviation comprehensive index and the standardized edge entropy difference, respectively, where greater than 0, Less than 0.

[0110] Therefore, the greater the accuracy assessment coefficient of pest spectral monitoring, the worse the accuracy of spectral analysis in high-risk monitoring areas.

[0111] Set the spectral accuracy threshold corresponding to the pest spectral monitoring accuracy assessment coefficient, and compare the pest spectral monitoring accuracy assessment coefficient with the spectral accuracy threshold: When the pest spectral monitoring accuracy assessment coefficient is less than or equal to the spectral accuracy threshold, it is determined that the pest spectral monitoring accurate signal is generated, and the accuracy of the distribution trend map of new pests is determined to meet the standard. This indicates that the spectral data quality is high, the distortion risk and randomness are low, and the monitoring system has a high accuracy in identifying and judging the distribution trend of new pests. At this time, it can be confirmed that the current monitoring data can be trusted, and the distribution trend map of new pests generated based on these data is accurate, which can be used as the basis for subsequent prevention and control decisions, and no additional adjustment measures are required.

[0112] When the pest spectral monitoring accuracy assessment coefficient is greater than the spectral accuracy threshold, a pest spectral monitoring accuracy hidden danger warning signal is generated, and the accuracy of the distribution trend map of new pests is determined to be substandard. This indicates that the spectral data has a large distortion risk or randomness, which may lead to unreliable monitoring results, large errors, and insufficient accuracy of the distribution trend map of new pests.

[0113] Generate accurate warning signals for pest spectral monitoring, indicating potential problems in the system. Further inspection and optimization of the spectral data collection process may be required, which may require recalibration of equipment, adjustment of environmental conditions, or re-collection of data, and temporarily refrain from making important decisions based on the current distribution trend chart.

[0114] The setting of the spectral accuracy threshold is based on historical data analysis and experience accumulation, and is determined through the evaluation of a large amount of monitoring data. Specifically, the quality distribution of spectral data in different monitoring areas is first analyzed, and a dividing point that can distinguish high-quality data from low-quality data is set according to actual conditions. This threshold is usually determined by the system in the initial debugging stage, and can be adjusted based on feedback from actual applications to ensure that the threshold can effectively reflect the accuracy of spectral data in different environments and conditions.

[0115] Only the degree of hidden dangers and randomness of spectral distortion of pest spectral data in high-risk monitoring areas are analyzed, and these two factors are comprehensively evaluated to evaluate the accuracy of spectral analysis, mainly because these areas have a higher risk of emerging pest activities, and the impact of data quality on monitoring results is more critical. In high-risk monitoring areas, emerging pest activities are frequent and complex, and spectral data are easily disturbed by factors such as environmental changes and equipment errors, resulting in increased spectral distortion or information randomness. These problems may directly affect the monitoring system's accurate judgment of pest types, quantities and distribution. By focusing on the analysis of spectral distortion hazards and randomness in these areas, potential problems in the data can be identified in advance, data processing and monitoring strategies can be optimized, and the reliability of monitoring results can be improved. In addition, the monitoring accuracy of high-risk monitoring areas has an important impact on the overall safety of granaries. Therefore, a more detailed and comprehensive analysis of these areas can ensure the efficient operation of the early warning system and reduce the potential risk of emerging pest outbreaks.

[0116] Example 2

[0117] The difference between Example 2 of the present invention and Example 1 is that this example introduces a granary pest monitoring and early warning system based on big data.

[0118] Figure 2 A structural schematic diagram of the granary pest monitoring and early warning system based on big data of the present invention is given. The granary pest monitoring and early warning system based on big data includes a granary area division module, an emerging pest monitoring module, a distribution trend early warning module, a risk area marking module, a risk area analysis module and a trend accurate judgment module.

[0119] Granary area division module: The granary is divided into multiple monitoring areas, each of which is equipped with an independent spectral sensor network.

[0120] Emerging pest monitoring module: Based on spectral analysis technology, each monitoring area is monitored in real time, and by comparing with the granary pest library, it is determined whether new pests appear in the monitoring area.

[0121] Distribution trend warning module: When new pests appear in the monitoring area, the distribution of the new pests is monitored based on spectral imaging technology, and a distribution trend map of the new pests is generated in real time. The distribution trend map of the new pests is used to determine whether to issue a warning.

[0122] Risk area marking module: The monitoring area where new pests appear and its adjacent monitoring areas are marked as high-risk monitoring areas.

[0123] Risk area analysis module: For high-risk monitoring areas: By analyzing the symmetry of pest spectral images at specific wavelengths, the degree of spectral distortion risk of pest spectral data is evaluated; the entropy value of pest spectral data is analyzed using the principle of information entropy to evaluate the randomness of pest spectral data.

[0124] Trend accuracy judgment module: Based on a comprehensive analysis of the degree of spectral distortion risk and the randomness of pest spectral data, the accuracy of spectral analysis in high-risk monitoring areas is evaluated, and it is determined whether the accuracy of the distribution trend map of new pests meets the standards.

[0125] Example 3

[0126] A device comprises: a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, a granary pest monitoring and early warning method based on big data is implemented.

[0127] Example 4

[0128] A computer-readable storage medium stores a program or instruction, and when the program or instruction is executed by a processor, a granary pest monitoring and early warning method based on big data is implemented.

[0129] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0130] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or may be transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0131] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0132] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0133] In the several embodiments provided in the present application, 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 modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules 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 modules, which can be electrical, mechanical or other forms.

[0134] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0135] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0136] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0137] 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.

[0138] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A granary pest monitoring and early warning method based on big data, characterized in that: The steps include: S1: Divide the granary into multiple monitoring areas, each equipped with an independent spectral sensor network; S2: Real-time monitoring of each monitoring area is carried out based on spectral analysis technology, and by comparing it with the granary pest library, it is determined whether new pests appear in the monitoring area; S3: When new pests appear in the monitoring area, the distribution of the new pests is monitored based on spectral imaging technology, and a distribution trend map of the new pests is generated in real time. The distribution trend map of the new pests is used to determine whether to issue an early warning; S4: Mark the monitoring areas where new pests appear and their adjacent monitoring areas as high-risk monitoring areas; S5: For high-risk monitoring areas: By analyzing the symmetry of pest spectral images at specific wavelengths, evaluate the potential risk of spectral distortion of pest spectral data; use the principle of information entropy to analyze the entropy value of pest spectral data and evaluate the randomness of pest spectral data; S6: Based on a comprehensive analysis of the potential risk of spectral distortion and the randomness of pest spectral data, the accuracy of spectral analysis in high-risk monitoring areas is evaluated, and a determination is made as to whether the accuracy of the distribution trend map of emerging pests meets the standards.

2. The method for monitoring and early warning of granary pests based on big data according to claim 1 is characterized in that: Based on spectral analysis technology, each monitoring area is monitored in real time, and by comparing with the granary pest library, it is determined whether new pests appear in the monitoring area, including: S201: Spectral sensors deployed in each monitoring area capture pest spectral data in real time; S202: performing noise filtering and feature enhancement on the collected pest spectral data, and automatically extracting key spectral features; S203: comparing the preprocessed pest spectral data with the known pest spectral fingerprints in the granary pest library, and using a spectral similarity matching algorithm to preliminarily screen out the pest type that is most similar to the current spectral data; S204: When a new spectral pattern that does not match the granary pest library is identified in the pest spectral data, it is marked as a new pest, and its spectral characteristics and the area information of the first appearance are recorded.

3. The method for monitoring and early warning of granary pests based on big data according to claim 2 is characterized in that: When new pests appear in the monitoring area, the distribution of new pests is monitored based on spectral imaging technology, and a distribution trend map of the new pests is generated in real time. The distribution trend map of the new pests is used to determine whether to issue an early warning. Specifically: S301: collecting distribution spectrum data of newly emerged pests in real time; based on the distribution spectrum data of newly emerged pests in continuous time periods, applying a time series analysis method to generate a dynamic distribution trend diagram of newly emerged pests; S302: Performing pattern recognition analysis on the generated distribution trend graph, combining pest activity hotspots and distribution density changes, to determine whether a warning threshold has been reached; S303: When the analysis result of the distribution trend graph exceeds the preset risk threshold, the system automatically triggers an early warning and sends an alert to the management personnel.

4. The method for monitoring and early warning of granary pests based on big data according to claim 3 is characterized in that: S301 specifically includes: Continuously collect spectral data of newly emerged pests at predetermined time intervals; The continuous spectral data of emerging pests are arranged in chronological order to construct a time series data set; the data at each time point include the spectral characteristics of emerging pests at various wavelengths and the spatial distribution of emerging pests in the monitoring area; The time series analysis algorithm is used to analyze the spectral distribution data of new pests at different time points to identify the change pattern of new pests in the time dimension; Based on the analysis results, a dynamic distribution trend map of emerging pests is generated to show the diffusion path, diffusion speed and spatial concentration of emerging pests in the monitoring area.

5. The method for monitoring and early warning of granary pests based on big data according to claim 4 is characterized in that: For high-risk monitoring areas, the symmetry of pest spectral images at specific wavelengths is analyzed to assess the potential spectral distortion risk of pest spectral data. Specifically: Collect spectral image data of pests in high-risk monitoring areas; The collected spectral image is standardized; the standardized spectral image is subjected to a two-dimensional discrete Fourier transform at a specific wavelength to calculate a complex matrix in the frequency domain; Extracting a phase spectrum from the complex matrix in the frequency domain after Fourier transformation, the phase spectrum represents the geometric structure information of the image; Symmetry analysis of the phase spectrum: compare the phase values ​​of the central symmetric points in the frequency domain and calculate the overall symmetry deviation of the spectrum image; The symmetry of the spectral image at different scales is analyzed by performing scale-space transformation on the phase spectrum. The symmetry calculation steps are repeated for each scale, and the results are weighted combined to obtain a comprehensive index of multi-scale symmetry deviation.

6. The method for monitoring and early warning of granary pests based on big data according to claim 5 is characterized in that: For high-risk monitoring areas, the entropy value of pest spectral data is analyzed using the principle of information entropy to evaluate the randomness of pest spectral data. Specifically: The edge area is extracted from the spectral image at each wavelength by using the edge detection algorithm; The spectral image is divided into two parts: edge area and non-edge area; For the spectral data in the edge area, the information entropy formula is applied to calculate the edge area entropy value; For the spectral data in the non-edge area, the information entropy formula is applied to calculate the entropy value of the non-edge area; The edge entropy difference is calculated by the difference between the edge area entropy value and the non-edge area entropy value; The edge entropy difference is normalized to obtain the standardized edge entropy difference.

7. The method for monitoring and early warning of granary pests based on big data according to claim 6 is characterized in that: Based on a comprehensive analysis of the potential spectral distortion risk and randomness of pest spectral data, the accuracy of spectral analysis in high-risk monitoring areas is evaluated, and the accuracy of the distribution trend map of emerging pests is determined to be up to standard. Specifically: The multi-scale symmetry deviation comprehensive index and the standardized edge entropy difference are normalized, and the normalized multi-scale symmetry deviation comprehensive index and the standardized edge entropy difference are assigned preset proportional coefficients respectively, and the pest spectral monitoring accuracy evaluation coefficient is calculated; Set the spectral accuracy threshold corresponding to the pest spectral monitoring accuracy assessment coefficient, and compare the pest spectral monitoring accuracy assessment coefficient with the spectral accuracy threshold: When the pest spectral monitoring accuracy assessment coefficient is less than or equal to the spectral accuracy threshold, it is determined that an accurate pest spectral monitoring signal is generated, and it is determined that the accuracy of the distribution trend diagram of the newly emerged pests meets the standard; When the pest spectral monitoring accuracy assessment coefficient is greater than the spectral accuracy threshold, it is determined that a pest spectral monitoring accurate hidden danger warning signal is generated, and it is determined that the accuracy of the distribution trend map of new pests does not meet the standard.

8. A granary pest monitoring and early warning system based on big data, used to implement the granary pest monitoring and early warning method based on big data as described in any one of claims 1 to 7, characterized in that: It includes granary area division module, new pest monitoring module, distribution trend warning module, risk area marking module, risk area analysis module and trend accurate judgment module; Granary area division module: divides the granary into multiple monitoring areas, each of which is equipped with an independent spectral sensor network; New pest monitoring module: Based on spectral analysis technology, each monitoring area is monitored in real time, and by comparing with the granary pest library, it is determined whether new pests appear in the monitoring area; Distribution trend warning module: When new pests appear in the monitoring area, the distribution of the new pests is monitored based on spectral imaging technology, and a distribution trend map of the new pests is generated in real time. The distribution trend map of the new pests is used to determine whether to issue a warning; Risk area marking module: marking the monitoring area where new pests appear and its adjacent monitoring areas as high-risk monitoring areas; Risk area analysis module: For high-risk monitoring areas: by analyzing the symmetry of pest spectral images at specific wavelengths, the degree of spectral distortion risk of pest spectral data is evaluated; The entropy value of pest spectral data is analyzed using the principle of information entropy to evaluate the randomness of pest spectral data; Trend accuracy judgment module: Based on a comprehensive analysis of the degree of spectral distortion risk and the randomness of pest spectral data, the accuracy of spectral analysis in high-risk monitoring areas is evaluated, and it is determined whether the accuracy of the distribution trend map of new pests meets the standards.

9. A device, characterized in that: The device includes: a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, the granary pest monitoring and early warning method based on big data as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores programs or instructions, and when the programs or instructions are executed by the processor, the granary pest monitoring and early warning method based on big data as described in any one of claims 1 to 7 is implemented.

Images