A pest population behavior dynamic monitoring method based on an internet of things

Abstract

The application discloses a pest population behavior dynamic monitoring method based on an Internet of Things, and belongs to the technical field of the Internet of Things and biological monitoring. The method comprises the following steps: a multi-physical field sensor network is deployed; according to the environmental characteristics of a monitoring area, multi-dimensional environmental physical quantities are collected; original monitoring data are subjected to space and level mapping, and wind fields, temperature fields and water vapor fields are subjected to normalized conversion to obtain a pretreatment data set constructed by unified quantitative representations of different physical field data; unified physical field data in the pretreatment data set are linked and integrated with pest behavior data collected in the Internet of Things to establish a response model for real-time matching of pest behavior and environmental physical conditions; and the linked monitoring results output by the response model are compared with actual observation results of the monitoring area, and the matching rules of the response model for the pest behavior and the environmental physical conditions are adjusted by using environmental feedback to optimize the response model.

Description

Technical Field

[0001] The present invention belongs to the field of Internet of Things and biological monitoring technology, and more specifically, relates to a method for dynamic monitoring of pest population behavior based on the Internet of Things. Background Art

[0002] Currently, multi-physics field linkage response is limited. In extreme climates, unique geological conditions, or complex hydrological environments, such as those near karst areas or along rivers and lakes, a single IoT monitoring system struggles to capture the nonlinear responses to pest behavior caused by the combined effects of multiple environmental physical fields (such as wind, temperature differences, and water vapor transport). This, to a certain extent, makes it difficult for traditional IoT systems to accurately monitor pest dynamic behavior in such extreme or variable environments. This is especially true when physical signals are transmitted between different levels of the ecosystem, as traditional IoT systems lack a certain degree of cross-system physical quantity conversion capabilities. Summary of the Invention

[0003] In order to address the deficiencies in the prior art, the present invention aims to solve the above-mentioned defects and further propose a method for dynamic monitoring of pest population behavior based on the Internet of Things.

[0004] The present invention adopts the following technical solutions.

[0005] The first aspect of the present invention discloses a method for dynamic monitoring of pest population behavior based on the Internet of Things, the method comprising:

[0006] By deploying a multi-physics field sensor network, multi-dimensional environmental physical quantities are collected according to the environmental characteristics of the monitoring area. The multi-dimensional environmental physical quantities are raw monitoring data from different sensors.

[0007] Performing spatial and hierarchical mapping on the original monitoring data, and normalizing and transforming the wind field, temperature field, and water vapor field to obtain a preprocessed data set constructed by unified quantitative representation of different physical field data;

[0008] Linking and integrating the unified physical field data in the preprocessed dataset with the pest behavior data collected in the Internet of Things to establish a response model for real-time matching of pest behavior with environmental physical conditions;

[0009] Comparing the linkage monitoring results output by the response model with the actual observation results of the monitoring area, and using environmental feedback to adjust the matching rules of the response model for pest behavior and environmental physical conditions to optimize the response model;

[0010] The multi-physical field sensor network is obtained by conducting on-site surveys of the monitoring area and planning and deploying the locations, and the environmental characteristics include climate characteristics, geological characteristics, and hydrological environmental characteristics of the monitoring environment.

[0011] Furthermore, the multi-dimensional environmental physical quantities are collected according to the environmental characteristics of the monitoring area by deploying a multi-physics field sensor network, including:

[0012] Divide the monitoring area into multiple sub-areas based on the climatic, geological and hydrological characteristics of the monitoring area, and calculate the effective monitoring coverage area of ​​each sub-area;

[0013] Determining the sensor type, the number of sensors, and the density of the planned deployment points based on the effective monitoring coverage area of ​​each sub-area, so as to design an installation strategy for the sensor nodes according to the density of the planned deployment points and environmental characteristics;

[0014] The effective monitoring coverage area is the product of the total area of ​​the sub-region, the terrain correction factor, the climate correction factor, and the water temperature correction factor;

[0015] Install and deploy multiple sensors corresponding to various types according to the installation strategy, and perform on-site connectivity testing on each installed sensor node to determine whether each installed sensor node meets the preset connectivity requirements;

[0016] When the installed sensor nodes meet the preset connectivity requirements, the integrated deployment of the multi-physics field sensor network is completed to collect multi-dimensional environmental physical quantities; when the installed sensor nodes do not meet the preset connectivity requirements, the installation strategy is adjusted.

[0017] Furthermore, the raw monitoring data is spatially and hierarchically mapped, and the wind field, temperature field, and water vapor field are normalized and transformed to obtain a preprocessed data set constructed by unified quantitative representation of different physical field data, including:

[0018] A linear mapping function is used to convert multiple sets of original monitoring data into a unified dimension, wherein the linear mapping function is selected based on the instrument calibration data at the monitoring site;

[0019] The original monitoring data of each sensor with a unified dimension are integrated to construct a unified conversion matrix; the conversion matrix is ​​composed of the conversion coefficients of the wind field, the temperature field, and the humidity field;

[0020] The product of the conversion matrix and the original data vector and the bias vector are added to obtain a mapped unified physical field data vector; the original data vector is generated by separating and extracting different physical field data detected by various sensors connected to the multi-physical field sensor network in real time;

[0021] Normalizing the mapped unified physical field data vector to map corresponding physical field data in the mapped unified physical field data vector to a first interval to obtain a normalized data vector;

[0022] Selecting normalized data vectors according to preset integration requirements to form the preprocessed data set;

[0023] The mapped unified physical field data vector has the same dimension as the original data vector, and the bias vector is composed of a wind field bias, a temperature field bias, and a humidity field bias.

[0024] Furthermore, the unified physical field data in the preprocessed data set is linked and integrated with the pest behavior data collected in the Internet of Things to establish a response model for real-time matching of pest behavior with environmental physical conditions, including:

[0025] Determining, based on the preprocessed data set, that each sensor node outputs the normalized data vector, each sensor node having spatial coordinates and a timestamp, and constructing a spatiotemporal data matrix based on the spatial coordinates and timestamp combined with the normalized data vector output by each sensor node;

[0026] Based on the physical field data and the corresponding weight coefficients of the physical field data, a pest behavior index is established as a quantitative indicator of the response of the environmental physical field to the dynamic behavior of the pest, so as to construct a pest behavior index model through weighted linear combination;

[0027] Calculating the average pest behavior index, standard deviation, and spatial gradient within the monitoring area based on the pest behavior index of each sensor node and the spatiotemporal data matrix, wherein the average pest behavior index, standard deviation, and spatial gradient are used to characterize the dynamic changes in pest behavior;

[0028] The pest behavior index model is linked and integrated with the pest behavior data collected in the Internet of Things to establish the response model, and the response model is called to generate a linkage monitoring result based on the average pest behavior index, standard deviation and spatial gradient.

[0029] Furthermore, the linkage monitoring results output by the response model are compared with actual observation results of the monitoring area, and environmental feedback is used to adjust the matching rules of the response model for pest behavior and environmental physical conditions to optimize the response model, including:

[0030] Acquiring actual observation results of the monitoring area from the multi-physics field sensor network, and comparing the actual observation results of the monitoring area with the linkage monitoring results to calculate an error vector between the monitoring indicators under the current environment and the historical monitoring indicators in the linkage monitoring results;

[0031] Calibrate and adjust the model parameters of the response model according to the error vector to determine correction coefficients for each monitoring indicator, the correction coefficients including an indicator calibration coefficient of average pest behavior, a standard deviation calibration coefficient, a gradient calibration coefficient, and a time series calibration coefficient, and calculate a calibration correction amount based on the correction coefficients;

[0032] The calibration correction amount is applied to the linkage monitoring result to update the response model, and the final monitoring report and real-time feedback information of the pest behavior are generated by integrating the updated linkage monitoring result and the feedback interface output.

[0033] A second aspect of the present invention discloses a terminal, comprising a processor and a storage medium; the characteristics of the terminal are:

[0034] The storage medium is used to store instructions;

[0035] The processor is configured to operate according to the instructions to execute the steps of the method of the first aspect.

[0036] A third aspect of the present invention discloses a computer-readable storage medium having a computer program stored thereon, wherein the program implements the steps of the method described in the first aspect when executed by a processor.

[0037] The beneficial effects of the present invention are that, compared with the prior art, the present invention has the following advantages:

[0038] (1) Based on the environmental characteristics of the monitoring area, the present invention conducts on-site surveys in extreme climates, special geology and complex hydrological environments (such as karst areas and areas along rivers and lakes), plans and deploys points. In addition to traditional pest behavior monitoring sensors (such as high-resolution cameras and infrared sensors), it focuses on adding wind speed and direction sensors, temperature gradient detectors and water vapor concentration sensors. This can ensure coverage of various physical fields and further collect multi-dimensional environmental physical quantities, providing original input for subsequent physical field conversion, thus solving the limitation that a single IoT node is difficult to fully capture the response of complex environmental physical fields.

[0039] (2) The present invention performs spatial and hierarchical mapping on the raw data from different sensors, normalizing the wind field, temperature field, and humidity field, thereby achieving a unified quantitative expression of data from different physical fields. In addition, during the conversion process, conversion rules are set based on the characteristics of the on-site environment (such as the local temperature difference range, wind speed range, and water vapor concentration changes), and the mapping from local data to the overall physical field is completed. This achieves the conversion and integration of physical signals between different levels of the ecosystem, allowing the unified application of multi-physical field information in complex environments to pest behavior monitoring, solving the problem of insufficient linkage response of multiple physical fields in the environment.

[0040] (3) The present invention integrates the converted unified physical field data with the pest behavior data (including images, infrared, and other behavior monitoring information) collected in the Internet of Things. The unified data is also used to establish a response model between pest behavior and environmental physical conditions, which can focus on the nonlinear response of pests to temperature differences, wind direction changes, and water vapor transfer in extreme environments. Through spatiotemporal mapping, real-time matching of pest behavior and environmental changes is achieved, which makes up for the shortcomings of a single monitoring system in capturing pest behavior responses in extreme or changing environments, realizes dynamic monitoring and quantitative description, and improves monitoring accuracy and response speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a flow chart of a method for dynamic monitoring of pest population behavior based on the Internet of Things provided by the present invention. DETAILED DESCRIPTION

[0042] The present application will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present application.

[0043] like Figure 1 As shown, in one embodiment, a method for dynamic monitoring of pest population behavior based on the Internet of Things includes the following steps:

[0044] Step S110 , by deploying a multi-physics field sensor network, multi-dimensional environmental physical quantities are collected according to the environmental characteristics of the monitoring area. The multi-dimensional environmental physical quantities are original monitoring data from different sensors.

[0045] The multi-physics sensor network, developed through on-site surveys and planned deployment of monitoring areas, includes pest behavior monitoring sensors, wind direction and speed sensors, temperature gradient detectors, and water vapor concentration sensors. Environmental characteristics include the climate, geology, and hydrological characteristics of the monitored environment. Pest behavior monitoring sensors include high-resolution cameras and infrared sensors.

[0046] It is understandable that the environmental characteristics mentioned in the present invention, that is, the climate characteristics, geological characteristics and hydrological environment characteristics of the monitored environment, can be data characteristics constructed by other similar historical physical field data; in addition, environmental characteristics are not necessarily concrete data characteristics, but can be a subjective cognitive experience formed based on human experience, etc.

[0047] In some embodiments, the present invention provides a method for dynamic monitoring of pest population behavior based on the Internet of Things, wherein step S110 specifically includes the following steps:

[0048] Step S111 : dividing the monitoring area into multiple sub-areas according to the climate characteristics, geological characteristics and hydrological environment characteristics of the monitoring area, and calculating the effective monitoring coverage area of ​​each sub-area.

[0049] Step S112 , based on the effective monitoring coverage area of ​​each sub-area, determine the sensor type, the number of sensors, and the density of planned deployment points, so as to design an installation strategy for the sensor nodes according to the density of planned deployment points and environmental characteristics.

[0050] Among them, the effective monitoring coverage area is the product of the total area of ​​the sub-area, the terrain correction coefficient, the climate correction coefficient and the water temperature correction coefficient.

[0051] In some embodiments, the present invention provides a method for dynamic monitoring of pest population behavior based on the Internet of Things, wherein step S110 specifically further includes the following steps:

[0052] Step S113 : installing and deploying multiple sensors corresponding to various types according to the installation strategy, and performing on-site connectivity detection on each installed sensor node to determine whether each installed sensor node meets the preset connectivity requirements.

[0053] Step S114, when the installed sensor nodes meet the preset connectivity requirements, the integrated deployment of the multi-physics field sensor network is completed, and multi-dimensional environmental physical quantities are collected; when the installed sensor nodes do not meet the preset connectivity requirements, the installation strategy is adjusted.

[0054] In a specific embodiment, the present invention provides a method for dynamic monitoring of pest population behavior based on the Internet of Things, comprising steps 1 to 4:

[0055] Step 1: Multi-physics sensor network deployment.

[0056] Based on the environmental characteristics of the monitoring area, we conduct on-site surveys of areas with extreme climates, unique geology, and complex hydrological environments (such as karst areas and areas along rivers and lakes). We plan and deploy these sites. In addition to traditional pest behavior monitoring sensors (such as high-resolution cameras and infrared sensors), we also deploy wind speed and direction sensors, temperature gradient detectors, and water vapor concentration sensors to ensure coverage of all physical fields. Step 1 collects multidimensional environmental physical quantities, providing the raw input for subsequent physical field conversion. This overcomes the limitation of a single IoT node that cannot fully capture the complex physical field responses of the environment.

[0057] Specifically, it includes steps 1.1 to 1.4:

[0058] Step 1.1: Region division and calculation of effective coverage area.

[0059] Specifically, the monitoring area is divided based on environmental topography, climate, and hydrological survey data, and the effective monitoring coverage area of ​​each sub-area is calculated. The effective monitoring coverage area = the total area of ​​the sub-area × terrain correction factor × climate correction factor × hydrological correction factor. The terrain correction factor ranges from 0.7 to 1.0, reflecting coverage efficiency for both flat and rugged terrain. The climate correction factor ranges from 0.8 to 1.0 and is determined based on climate stability (wind speed, precipitation). The hydrological correction factor ranges from 0.6 to 1.0, reflecting the impact of water bodies and wetland distribution.

[0060] Step 1.2: sensor type selection and point density calculation.

[0061] Specifically, based on the environmental complexity within the sub-area, appropriate sensor types are selected and the sensor density is determined. The number of sensors required is the ratio of the effective coverage area obtained in step 1.1 to the square of the ideal inter-sensor spacing, rounded up to the nearest integer. The ideal inter-sensor spacing is determined based on environmental complexity and is typically between 50 and 200 meters. Next, a sensor type vector is constructed, consisting of a high-resolution camera for capturing images of pest activity, an infrared sensor for nighttime monitoring and heat distribution, a wind speed and direction sensor for monitoring wind speed, a temperature sensor for monitoring ambient temperature differences, and a humidity / water vapor sensor for capturing water vapor transfer.

[0062] Step 1.3, design the physical installation plan of the sensor node.

[0063] Specifically, the sensor node installation plan is designed based on the density of the nodes and the characteristics of the environment to ensure that each sensor has the optimal sampling position in the multi-physics field. The optimal sensor installation height is determined by an empirical formula that combines the terrain relief index (reflecting the flatness of the ground), the vegetation coverage ratio (value range 0-1), and the water proximity factor (value range 0.5-1.0) with various coefficients calibrated based on field tests.

[0064] Step 1.4: Network node physical connectivity detection and final integration.

[0065] Specifically, after completing the physical installation of each sensor node, an on-site connectivity test is performed to ensure that all nodes can achieve physical information collection within the expected range.

[0066] Among them, the connectivity detection expression is:

[0067]

[0068] Where, represents the distance between any two sensor nodes, , represents the plane coordinates of node i, , represents the plane coordinate of node j. This expression is used to detect whether the preset connectivity distance requirement (such as no more than 150m) between nodes is met to ensure the effective interconnection of multi-physics field information.

[0069] Step S120 , spatially and hierarchically mapping the original monitoring data, and normalizing the wind field, temperature field, and water vapor field to obtain a preprocessed data set constructed by unified quantitative representation of different physical field data.

[0070] In some embodiments, the present invention provides a method for dynamic monitoring of pest population behavior based on the Internet of Things, wherein step S120 specifically includes the following steps:

[0071] In step S121 , a linear mapping function is used to convert multiple groups of original monitoring data into a unified dimension. The linear mapping function is selected based on the instrument calibration data at the monitoring site.

[0072] In step S122 , the original monitoring data of the sensors with unified dimensions are integrated to construct a unified conversion matrix. The conversion matrix is ​​composed of the conversion coefficients of the wind field, the temperature field, and the humidity field.

[0073] In step S123, the product of the conversion matrix and the original data vector and the bias vector are added to equal the mapped unified physical field data vector. The original data vector is generated by separating and extracting different physical field data detected by various sensors connected to the multi-physical field sensor network in real time.

[0074] Step S124 , normalizing the mapped unified physical field data vector to map the corresponding physical field data in the mapped unified physical field data vector into the first interval to obtain a normalized data vector.

[0075] Step S125 : selecting normalized data vectors to form a preprocessed data set according to preset integration requirements.

[0076] Among them, the mapped unified physical field data vector has the same dimension as the original data vector, and the bias vector is composed of wind field bias, temperature field bias and humidity field bias.

[0077] In a specific embodiment, the present invention provides a method for dynamic monitoring of pest population behavior based on the Internet of Things, comprising steps 1 to 4:

[0078] Step 2: Construct cross-physical field quantity conversion module.

[0079] Based on the sensor network, a physical quantity conversion module is established. This module performs spatial and hierarchical mapping of raw data from different sensors, normalizing wind field, temperature difference, and water vapor information to achieve a unified quantitative expression of data from different physical fields. During the conversion process, conversion rules are set based on the characteristics of the on-site environment (such as the local temperature difference range, wind speed range, and water vapor concentration changes), and the mapping from local data to the overall physical field is completed. Step 2 is used to realize the conversion and integration of physical signals between different levels of the ecosystem, allowing multi-physics field information in complex environments to be uniformly applied to pest behavior monitoring, solving the problem of insufficient coordinated response of multiple physical fields in the environment.

[0080] Specifically, it includes steps 2.1 to 2.5:

[0081] Step 2.1: Access and extract original physical field data.

[0082] Specifically, various sensor data are accessed in real time from the established multi-physics sensor network, and different physical fields (wind field, temperature field, humidity / water vapor field) are separated and extracted. For each sensor node, its output data constitutes a raw data vector, which is composed of wind speed or direction data, temperature data, and humidity / water vapor concentration data. Each sensor data is verified on-site before being passed as input.

[0083] Step 2.2, multi-physics field data mapping function design.

[0084] Specifically, in order to express different physical field data on the same scale, a mapping function is designed to convert the original data into a unified dimension. Corresponding to each physical quantity, the mapped data is recorded as , this example uses a linear mapping function, whose general form is:

[0085]

[0086] Where, is the mapped physical field, is the physical field before mapping, is the physical field conversion coefficient, is the physics field offset value.

[0087] Combining the above formula, we can know that for the wind field: After mapping, we get , is the wind speed conversion coefficient, the constant range can be set to 0.8~1.2 (determined according to the instrument calibration results), is the wind speed offset value, and the constant range is generally -0.5~0.5m / s. The same applies to the temperature field and humidity / water vapor field.

[0088] It should be noted that the selection of the mapping function is determined based on the field instrument calibration data.

[0089] Step 2.3, transformation matrix construction and bias correction.

[0090] Specifically, the mapping results of each sensor are integrated to construct a unified transformation matrix to ensure that the physical field data can be compared within a unified scale. The transformation model constructed based on the transformation matrix is ​​as follows:

[0091]

[0092] Where, is the unified physical field data vector after mapping, and its dimension is the same as the original data vector X. is the transformation matrix, is the bias vector, which is composed of the wind field bias, temperature field bias and humidity field bias. The conversion matrix is ​​composed of the conversion coefficients of the wind field, temperature field and humidity field.

[0093] The above conversion model can ensure that all data have a unified dimension and standard after conversion. The conversion coefficient and bias vector can be adjusted according to the instrument calibration data, and the optimal value can be obtained by field testing.

[0094] Step 2.4, normalization and standardization preprocessing.

[0095] Specifically, in order to facilitate the direct comparison of data in different regions, the unified physical field data vector obtained by the mapping of the transformation matrix needs to be A mapping process is performed to map each physical field data to between 0 and 1.

[0096] Among them, the normalization formula is:

[0097]

[0098] Where, is the normalized data vector, 、 They are respectively the maximum and minimum values ​​of each physical field data in the historical preset data.

[0099] Step 2.5, integration testing and final output.

[0100] Specifically, the entire conversion module is integrated and tested to verify whether the normalized data Z meets the expected requirements. During the test, the output data of some sensor nodes is compared with the actual field measurements to adjust the conversion matrix M, bias vector B, and normalization parameters until the data output from all sensor nodes is stable. Finally, the final data set, i.e., the preprocessed data set, is integrated and output.

[0101] Step S130 , integrating the unified physical field data in the preprocessed data set with the pest behavior data collected in the Internet of Things to establish a response model for real-time matching of pest behavior with environmental physical conditions.

[0102] Among them, the Internet of Things is mainly used to collect pest behavior data. This data often comes from on-site monitoring equipment, video sensors, RFID or other devices used to record pest movement, aggregation and activity trajectories.

[0103] In some embodiments, the present invention provides a method for dynamic monitoring of pest population behavior based on the Internet of Things, wherein step S130 specifically includes the following steps:

[0104] Step S131 : determining the normalized data vector output by each sensor node based on the preprocessed data set, where each sensor node has spatial coordinates and a timestamp, and constructing a spatiotemporal data matrix based on the spatial coordinates and timestamp combined with the normalized data vector output by each sensor node.

[0105] Step S132 , establishing a pest behavior index as a quantitative index of the environmental physical field response to the dynamic behavior of pests based on the physical field data and the weight coefficient corresponding to the physical field data, so as to construct a pest behavior index model through weighted linear combination.

[0106] Step S133 , calculating the average pest behavior index, standard deviation, and spatial gradient in the monitoring area based on the pest behavior index of each sensor node and the spatiotemporal data matrix. The average pest behavior index, standard deviation, and spatial gradient are used to characterize the dynamic changes in pest behavior.

[0107] Step S134 , integrating the pest behavior index model with the pest behavior data collected in the Internet of Things to establish a response model, and calling the response model to generate a linkage monitoring result based on the average pest behavior index, standard deviation, and spatial gradient.

[0108] In a specific embodiment, the present invention provides a method for dynamic monitoring of pest population behavior based on the Internet of Things, comprising steps 1 to 4:

[0109] Step 3: Linked monitoring of pest behavior and environmental physical fields.

[0110] The converted unified physical field data is integrated with pest behavior data collected from the Internet of Things (IoT), including image, infrared, and other behavioral monitoring information. The monitoring system leverages this unified data to establish a response model between pest behavior and environmental physical conditions, focusing on the nonlinear responses of pests to temperature differences, wind direction changes, and water vapor transport in extreme environments. Through spatiotemporal mapping, pest behavior is matched to environmental changes in real time, addressing the limitations of traditional single-monitoring systems in capturing pest behavioral responses in extreme or changing environments. This enables dynamic monitoring and quantitative description, improving monitoring accuracy and response speed.

[0111] Specifically, it includes steps 3.1 to 3.4:

[0112] Step 3.1, unified data integration and spatiotemporal mapping.

[0113] Specifically, based on the unified physical field data from the converted and normalized preprocessed dataset, a data vector is constructed for each sensor node's output. This data vector is composed of the normalized wind field data, temperature field data, and humidity / water vapor field data. Combined with the spatial coordinates and timestamps of each sensor node, a corresponding spatiotemporal data matrix is ​​constructed. This spatiotemporal data matrix consists of each type of temperature field data, its corresponding plane coordinates, and its corresponding timestamp.

[0114] Step 3.2: Construction of pest behavior indicator model.

[0115] Specifically, a pest behavior index is established as a quantitative indicator of the environmental physical field's response to the pest's dynamic behavior, and then a pest behavior index model is constructed through weighted linear combination, which is expressed as follows:

[0116]

[0117] Where, represents the pest behavior response index of a single sensor node, 、 、 They are the normalized wind field data, temperature field data, and humidity / water vapor field data, 、 、 They are the weight coefficients corresponding to wind field data, temperature field data, and humidity / water vapor field data, that is, the impact of each physical field on the impact of the impact on the activity and living environment, which are obtained by field survey and historical data correction.

[0118] Step 3.3: Spatiotemporal dynamic behavior analysis and local statistical calculation.

[0119] Specifically, based on the pest behavior index B and spatiotemporal coordinates of each sensor node, local statistics and spatiotemporal dynamic analysis are performed to calculate the average pest behavior index, standard deviation, and spatial gradient in the monitoring area to reflect the dynamic changes in pest behavior.

[0120] The standard deviation reflects the fluctuation range of pest behavior indicators. A larger value indicates a more dramatic change in pest behavior response in space or time. The spatial gradient is approximated using a two-dimensional difference method, which calculates the gradient by selecting adjacent sensor nodes.

[0121] Step 3.4: Generate linkage monitoring results and preliminary dynamic behavior analysis results.

[0122] Specifically, the aforementioned statistical indicators are used to generate linkage monitoring results and preliminary dynamic behavior analysis results. The report content includes the average value, fluctuation range and time / space changes of the pest behavior indicators in the area, and then the final monitoring report is output. The final monitoring report is composed of the average pest behavior indicator, standard deviation, spatial gradient and time series analysis results. The time series analysis results are expressed as trend curve parameters of the changes in pest behavior indicators in different time periods. Its specific form can be an array of pest behavior indicators corresponding to a group of time nodes.

[0123] Step S140 , comparing the linkage monitoring results output by the response model with the actual observation results of the monitoring area, and using environmental feedback to adjust the matching rules of the response model for pest behavior and environmental physical conditions to optimize the response model.

[0124] It is understood that actual observation results can be derived from a highly integrated summary of actual observed pest behavior data. In the present invention, pest behavior data actually incorporates data of multiple dimensions, such as pest movement data, pest aggregation data, and activity trajectory data. The actual observation results integrate pest behavior data of different dimensions into data of the same dimension as the linked monitoring results, facilitating classification learning within the algorithm. In some embodiments, pest behavior data can be converted into actual observation results using techniques commonly employed, such as principal component analysis (PCA) and multimodal feature fusion.

[0125] In some embodiments, the present invention provides a method for dynamic monitoring of pest population behavior based on the Internet of Things, wherein step S140 specifically includes the following steps:

[0126] Step S141, obtaining actual observation results of the monitoring area from the multi-physics field sensor network, and comparing the actual observation results of the monitoring area with the linkage monitoring results to calculate the error vector between the monitoring indicators under the current environment and the historical monitoring indicators in the linkage monitoring results.

[0127] Step S142: calibrate and adjust the model parameters of the response model according to the error vector to determine the correction coefficients of each monitoring indicator. The correction coefficients include the indicator calibration coefficient of the average pest behavior, the standard deviation calibration coefficient, the gradient calibration coefficient, and the time series calibration coefficient, and calculate the calibration correction amount based on the correction coefficients.

[0128] Step S143 , applying the calibration correction amount to the linkage monitoring results to update the response model, and generating a final monitoring report and real-time feedback information of the pest behavior by integrating the updated linkage monitoring results and the feedback interface output.

[0129] The calibration correction amount is obtained by multiplying the correction coefficient by the elements in the error vector item by item.

[0130] In a specific embodiment, the present invention provides a method for dynamic monitoring of pest population behavior based on the Internet of Things, comprising steps 1 to 4:

[0131] Step 4: Dynamic calibration and real-time feedback optimization.

[0132] A dynamic calibration mechanism is established to compare the linked monitoring results obtained in step 3 with actual field observations, leveraging environmental feedback to adjust the mapping rules of the physical quantity conversion module. The system interacts with the agricultural pest control platform through a pre-set feedback interface, updating the monitoring model of the relationship between the field environment and pest behavior in real time. The final results are then displayed to monitoring personnel via a visualization platform. Step 4 is designed to improve monitoring accuracy and real-time responsiveness in extreme or complex physical environments, ensuring that the system can automatically calibrate according to field environmental changes and maintain the consistency of data transmission and monitoring results.

[0133] Specifically, it includes steps 4.1 to 4.4:

[0134] Step 4.1: Compare and verify the real-time data with the preliminary monitoring results.

[0135] Specifically, first, the new physical field data collected from the multi-physics field sensor network for real-time monitoring on site are compared with the previously generated linkage monitoring results to calculate the error vector between the monitoring indicators in the current environment and the indicators in the preliminary linkage monitoring results, which is composed of the difference between the corresponding data of the two.

[0136] Step 4.2, error calculation and calibration parameter adjustment.

[0137] Specifically, based on the error vector calculated in step 4.1, a calibration parameter adjustment model is constructed to determine the correction coefficients for each monitoring indicator. These correction coefficients are composed of the average pest behavior indicator calibration coefficient, the standard deviation calibration coefficient, the gradient calibration coefficient, and the time series calibration coefficient. The calibration corrections are then obtained by multiplying each element of the correction coefficient by each element of the error vector.

[0138] Step 4.3: Dynamic calibration model update and feedback interface construction.

[0139] Specifically, the calibration corrections obtained in step 4.2 are applied to the preliminary linked monitoring results to update the monitoring model (response model) and present it to the field in real time via the feedback interface. The updated monitoring model is the updated monitoring report data set, formatted identically to the preliminary linked monitoring results and representing the sum of the preliminary monitoring report and the calibration corrections. The feedback interface transmits the updated monitoring report data set to a display terminal to implement dynamic feedback optimization, including formatting the data into charts, reports, and warning signals, and generating real-time response information based on pre-set rules.

[0140] Step 4.4: Generate the final monitoring report and real-time feedback information.

[0141] Specifically, the calibration update results and feedback interface output are integrated to generate a final monitoring report and real-time feedback information. This report comprises the calibrated average pest behavior indicators, standard deviations, spatial gradients, and time series behavior indicators within the monitoring area, along with real-time response information generated by the feedback interface. This integrated display of the final monitoring report and real-time feedback ensures that field technicians receive complete and accurate decision-making support data.

[0142] The present disclosure may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0143] A computer-readable storage medium can be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punched card or raised structure in a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse passing through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0144] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0145] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, the state information of the computer-readable program instructions is used to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), so that the electronic circuit can execute the computer-readable program instructions, thereby implementing various aspects of the present disclosure.

[0146] Various aspects of the present disclosure are described herein with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.

[0147] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device is generated that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, where these instructions cause the computer, programmable data processing device, and / or other device to operate in a specific manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0148] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0149] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and the part of the module, program segment or instruction contains one or more executable instructions for realizing the prescribed logical function. In some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the prescribed function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0150] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for dynamic monitoring of pest population behavior based on the Internet of Things, characterized in that: The method comprises: By deploying a multi-physics field sensor network, multi-dimensional environmental physical quantities are collected according to the environmental characteristics of the monitoring area. The multi-dimensional environmental physical quantities are raw monitoring data from different sensors. Performing spatial and hierarchical mapping on the original monitoring data, and normalizing and transforming the wind field, temperature field, and water vapor field to obtain a preprocessed data set constructed by unified quantitative representation of different physical field data; Linking and integrating the preprocessed dataset with pest behavior data collected from the Internet of Things to establish a response model that matches pest behavior with environmental physical conditions in real time; Comparing the linkage monitoring results output by the response model with the actual observation results of the monitoring area, and using environmental feedback to adjust the matching rules of the response model for pest behavior and environmental physical conditions to optimize the response model; The multi-physical field sensor network is obtained by conducting on-site surveys of the monitoring area and planning and deploying the locations, and the environmental characteristics include climate characteristics, geological characteristics, and hydrological environmental characteristics of the monitoring environment; The raw monitoring data is spatially and hierarchically mapped, and the wind field, temperature field, and water vapor field are normalized and converted to obtain a preprocessed data set constructed by unified quantitative representation of different physical field data, including: A linear mapping function is used to convert multiple sets of original monitoring data into a unified dimension, wherein the linear mapping function is selected based on the instrument calibration data at the monitoring site; The original monitoring data of each sensor with a unified dimension are integrated to construct a unified conversion matrix; the conversion matrix is ​​composed of the conversion coefficients of the wind field, the temperature field, and the humidity field; The product of the conversion matrix and the original data vector and the bias vector are added to obtain a mapped unified physical field data vector; the original data vector is generated by separating and extracting different physical field data detected by various sensors connected to the multi-physical field sensor network in real time; Normalizing the mapped unified physical field data vector to map corresponding physical field data in the mapped unified physical field data vector to a first interval to obtain a normalized data vector; Selecting normalized data vectors according to preset integration requirements to form the preprocessed data set; The mapped unified physical field data vector has the same dimension as the original data vector, and the bias vector is composed of a wind field bias, a temperature field bias, and a humidity field bias; The method of integrating the pre-processed dataset with the pest behavior data collected in the Internet of Things to establish a response model for real-time matching of pest behavior with environmental physical conditions includes: Determining the normalized data vector output by each sensor node based on the preprocessed data set, each sensor node having spatial coordinates and a timestamp, and constructing a spatiotemporal data matrix based on the spatial coordinates and timestamp combined with the normalized data vector output by each sensor node; Based on the physical field data and the corresponding weight coefficients of the physical field data, a pest behavior index is established as a quantitative indicator of the response of the environmental physical field to the dynamic behavior of the pest, so as to construct a pest behavior index model through weighted linear combination; Calculating the average pest behavior index, standard deviation, and spatial gradient within the monitoring area based on the pest behavior index of each sensor node and the spatiotemporal data matrix, wherein the average pest behavior index, standard deviation, and spatial gradient are used to characterize the dynamic changes in pest behavior; The pest behavior index model is linked and integrated with the pest behavior data collected in the Internet of Things to establish the response model, and the response model is called to generate a linkage monitoring result based on the average pest behavior index, standard deviation and spatial gradient.

2. The method for dynamic monitoring of pest population behavior based on the Internet of Things according to claim 1, characterized in that: The multi-dimensional environmental physical quantities are collected based on the environmental characteristics of the monitoring area by deploying a multi-physics field sensor network, including: Divide the monitoring area into multiple sub-areas based on the climatic, geological and hydrological characteristics of the monitoring area, and calculate the effective monitoring coverage area of ​​each sub-area; Based on the effective monitoring coverage area of ​​each sub-region, determine the sensor type, number of sensors, and density of the planned deployment points, so as to design a sensor node installation strategy based on the density of the planned deployment points and environmental characteristics; wherein the effective monitoring coverage area is the product of the total area of ​​the sub-region, the terrain correction factor, the climate correction factor, and the water temperature correction factor; Install and deploy multiple sensors corresponding to various types according to the installation strategy, and perform on-site connectivity testing on each installed sensor node to determine whether each installed sensor node meets the preset connectivity requirements; When the installed sensor nodes meet the preset connectivity requirements, the integrated deployment of the multi-physics field sensor network is completed to collect multi-dimensional environmental physical quantities; when the installed sensor nodes do not meet the preset connectivity requirements, the installation strategy is adjusted.

3. The method for dynamic monitoring of pest population behavior based on the Internet of Things according to claim 2, characterized in that: The method of comparing the linkage monitoring results output by the response model with the actual observation results of the monitoring area and adjusting the matching rules of the response model between pest behavior and environmental physical conditions by using environmental feedback to optimize the response model includes: Acquiring actual observation results of the monitoring area from the multi-physics field sensor network, and comparing the actual observation results of the monitoring area with the linkage monitoring results to calculate an error vector between the monitoring indicators under the current environment and the historical monitoring indicators in the linkage monitoring results; Calibrate and adjust the model parameters of the response model according to the error vector to determine correction coefficients for each monitoring indicator, the correction coefficients including an indicator calibration coefficient of average pest behavior, a standard deviation calibration coefficient, a gradient calibration coefficient, and a time series calibration coefficient, and calculate a calibration correction amount based on the correction coefficients; Applying the calibration correction to the linkage monitoring results to update the response model, and generating a final monitoring report and real-time feedback information on pest behavior by integrating the updated linkage monitoring results and feedback interface output; The calibration correction amount is obtained by multiplying the correction coefficient by the elements in the error vector item by item.

4. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1 to 3.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.

Images