Forestry pest informatization comprehensive management system and method

Abstract

The invention discloses a forestry pest informatization comprehensive management system and method, and relates to the technical field of forestry informatization, and the method comprises the steps: constructing a fuzzy cognitive map model, predicting the governance priority score of each forest region based on forest region multi-source data after time alignment, and generating a governance priority thermodynamic diagram; based on the governance priority thermodynamic diagram, obtaining the maximum photochemical efficiency in the forest farm and recording distribution coordinates, and generating an ecological baseline data set through space-time matching; according to the priority thermodynamic diagram, a multi-criterion decision algorithm is adopted to generate a pesticide application instruction, and the unmanned aerial vehicle is scheduled to execute differential pesticide application according to the pesticide application instruction; and obtaining a pesticide application amount correction coefficient, calculating an ecological restoration score through an ecological resonance index method in combination with the ecological baseline data set, and performing real-time adjustment on the pesticide application instruction according to the ecological restoration score. According to the method, through the step of constructing the fuzzy cognitive map model, dynamic coupling analysis of the pest risk index and the ecological sensitivity is realized, and the accuracy and adaptability of risk assessment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of forestry informatization, and particularly to an integrated information management system and method for forestry pests and diseases. Background Art

[0002] The informatization management of forestry pests and diseases is an important technical field for forest resource protection and ecological security maintenance. Currently, the technical system in this field is mainly established on the basis of multi-source data collection, spatial analysis, and decision support systems. Conventional methods usually use remote sensing monitoring technology to obtain vegetation index data, combine ground sensor networks to collect environmental parameters, and achieve preliminary identification of pest occurrence areas through GIS spatial analysis. In terms of data processing, existing technologies generally use time series analysis and spatial interpolation methods to preprocess monitoring data and establish a rule-based pest risk assessment model. In recent years, with the development of machine learning technology, some advanced systems have been able to train prediction models by combining historical pest data to achieve a certain degree of prediction of pest occurrence probability. In terms of control decision-making, existing methods mostly use the analytic hierarchy process to determine control priorities and display analysis results through visualization technology to provide decision-making references for forestry management.

[0003] However, there is still room for improvement in the risk assessment model and control decision-making mechanism of existing technologies. On the one hand, conventional pest risk assessment models mostly use static weight allocation methods, which are difficult to adapt to the ecological characteristic differences of different forest areas, resulting in a deviation between the assessment results and the actual damage degree. On the other hand, existing control decision-making systems lack a dynamic monitoring and feedback mechanism for ecological restoration ability and cannot adjust control strategies in real time according to the ecological response after pesticide application, affecting the sustainability and accuracy of control effects. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an integrated information management method for forestry pests and diseases to solve the problems of insufficient evaluation accuracy and lagging control decision-making in the prior art.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides an information-based integrated management method for forestry pests, which includes dividing a forest farm into multiple forest areas, collecting multi-source data of each forest area and aligning them in time; constructing a fuzzy cognitive map model, predicting the governance priority scores of each forest area based on the multi-source data of the forest areas after time alignment, and generating a governance priority heat map; based on the governance priority heat map, obtaining the maximum photochemical efficiency within the forest farm and recording the distribution coordinates, and generating an ecological baseline data set through spatio-temporal matching; according to the priority heat map, using a multi-criteria decision-making algorithm to generate a pesticide application instruction, and dispatching an unmanned aerial vehicle to perform differential pesticide application according to the pesticide application instruction; obtaining a pesticide application amount correction coefficient and combining it with the ecological baseline data set, calculating an ecological restoration score through an ecological resonance index method, and adjusting the pesticide application instruction in real time according to the ecological restoration score.

[0007] As a preferred embodiment of the information-based integrated management method for forestry pests according to the present invention, wherein: the forest farm is divided into multiple forest areas by a spatial clustering analysis method; Using a GIS spatial index to spatially align the multi-source data of each discrete forest area, and performing dynamic time calibration on the space by clock drift compensation to generate the multi-source data of the forest areas after spatio-temporal alignment.

[0008] As a preferred embodiment of the information-based integrated management method for forestry pests according to the present invention, wherein: the steps of constructing the fuzzy cognitive map model are as follows, Define the pest risk index and ecological sensitivity; Through partial least squares path analysis, establish the causal relationship between the pest risk index and ecological sensitivity and the multi-source data of the forest area, and use the Bayesian network inference algorithm to construct the fuzzy cognitive map weight matrix; Define the state transition rule between the pest risk index and ecological sensitivity, and perform iterative update through Sigmoid to generate a dynamic state transition mechanism; Through a fuzzy dynamic cooperation algorithm, combine the fuzzy cognitive map weight matrix with the dynamic state transition mechanism to form a fuzzy cognitive map model.

[0009] As a preferred embodiment of the information-based integrated management method for forestry pests according to the present invention, wherein: the steps of predicting the governance priority scores of each forest area and generating a governance priority heat map are as follows, The fuzzy cognitive map model predicts the governance priority scores of each forest area based on the multi-source data of the aligned forest areas, the pest risk index and ecological sensitivity; Perform clustering analysis on the historical pest diffusion rate of the forest area by the K-means algorithm, and define the low-priority threshold and high-priority threshold; According to the comparison results of the governance priority scores of each forest area with the low-priority threshold and high-priority threshold, mark the hazards of each forest area; Perform probability density interpolation on each forest area after marking through spatial kernel density estimation to generate a heat map of treatment priorities.

[0010] As a preferred embodiment of the comprehensive information management method for forest pests and diseases described in the present invention, wherein: the steps of obtaining the maximum photochemical efficiency in the forest farm and recording the distribution coordinates, and generating an ecological baseline data set through spatio-temporal matching are as follows. Obtain the spatial mask of the high-hazard forest areas in the heat map of treatment priorities, and through the stratified random sampling method, measure the initial chlorophyll fluorescence and maximum chlorophyll fluorescence of dark-adapted leaves in the spatial mask of the high-hazard forest areas by an AM fluorometer, and use the Genty formula to obtain the maximum photochemical efficiency, and at the same time record the distribution coordinates through GPS. Use the Kriging spatial interpolation method to perform spatial matching on the NDVI vegetation index, maximum photochemical efficiency, and distribution coordinates in the forest farm, and combine the improved dynamic time warping algorithm for time series alignment to generate an ecological baseline data set.

[0011] As a preferred embodiment of the comprehensive information management method for forest pests and diseases described in the present invention, wherein: the steps of generating a pesticide application instruction by using a multi-criteria decision-making algorithm are as follows. Perform resolution registration on the spatial mask of the high-hazard forest areas and the ecological baseline data set through the GDAL library to generate a multi-variable raster stack. Adopt the band splitting method to separate the three criterion layers from the multi-variable raster stack, and perform weight allocation through the analytic hierarchy process. According to the weights of each criterion layer, perform spatial weighted overlay analysis by using MCDA to obtain the coordinates and application rates of the pesticide application forest areas, and generate a treatment path according to the treatment priority scores of each forest area. Map the coordinates, application rates, and treatment paths of the pesticide application forest areas into pesticide application instructions through multi-criteria decision mapping.

[0012] As a preferred embodiment of the comprehensive information management method for forest pests and diseases described in the present invention, wherein: the steps of calculating the ecological restoration score by using the ecological resonance index method and adjusting the pesticide application instruction in real time according to the ecological restoration score are as follows. Based on the application rate in the pesticide application instruction and the maximum photochemical efficiency in the ecological baseline data set, calculate the application rate correction coefficient by using the dose-response surface method, and combine the ecological baseline data set to calculate the ecological restoration score of each forest area by using the ecological resonance index method. Based on the ecological restoration scores of each forest area, identify and mark the ecological restoration deviation status through dynamic cluster analysis, and update the application rate correction coefficient through the incremental learning algorithm, and optimize the pesticide application instruction according to the updated application rate correction coefficient.

[0013] In a second aspect, the present invention provides an information-based integrated management system for forestry pests, including a data collection module, a priority division module, a baseline generation module, an instruction generation module, and an ecological restoration evaluation module; The data collection module is used to divide the forest farm into multiple forest areas, collect multi-source data of each forest area, and perform time alignment; The priority division module is used to construct a fuzzy cognitive map model, predict the governance priority scores of each forest area based on the multi-source data of the forest areas after time alignment, and generate a governance priority heat map; The baseline generation module is used to obtain the maximum photochemical efficiency in the forest farm based on the governance priority heat map, record the distribution coordinates, and generate an ecological baseline data set through spatio-temporal matching; The instruction generation module is used to generate a pesticide application instruction according to the priority heat map by using a multi-criteria decision-making algorithm, and dispatch an unmanned aerial vehicle to perform differential pesticide application according to the pesticide application instruction; The ecological restoration evaluation module is used to obtain a pesticide application amount correction coefficient, combine it with the ecological baseline data set, calculate an ecological restoration score by using the ecological resonance index method, and adjust the pesticide application instruction in real time according to the ecological restoration score.

[0014] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the information-based integrated management method for forestry pests as described in the first aspect of the present invention is implemented.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the information-based integrated management method for forestry pests as described in the first aspect of the present invention is implemented.

[0016] The beneficial effects of the present invention are as follows: through the step of constructing a fuzzy cognitive map model, the dynamic coupling analysis of the pest risk index and ecological sensitivity is realized, and the accuracy and adaptability of risk assessment are improved; by calculating the ecological restoration scores of each forest area by using the ecological resonance index method, the dynamic correlation assessment of the pesticide application effect and the ecological restoration state is realized, and the pertinence and timeliness of the prevention and control measures are enhanced. Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1It is a flowchart of the information-based comprehensive management method for forestry pests.

[0019] Figure 2 It is a schematic diagram of the information-based comprehensive management system for forestry pests.

[0020] Figure 3 It is a flowchart of the construction of the fuzzy cognitive map model.

[0021] Figure 4 It is a flowchart of the generation of the heat map of control priorities. Specific implementation manners

[0022] To make the above objects, features and advantages of the present invention more obvious and understandable, the specific implementation manners of the present invention will be described in detail below with reference to the accompanying drawings of the specification.

[0023] Many specific details are set forth in the following description in order to fully understand the present invention, but the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0024] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure or characteristic that can be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude other embodiments.

[0025] Refer to Figures 1 to 4 , which is an embodiment of the present invention. This embodiment provides an information-based comprehensive management method for forestry pests, including the following steps: S1. Divide the forest farm into multiple forest areas, collect multi-source data of each forest area and align them in time; The multi-source data of the forest area includes NDVI vegetation index, soil temperature, leaf surface conductivity and soil water content; Furthermore, the NDVI vegetation index is obtained by aerial survey with a multi-spectral sensor carried by an unmanned aerial vehicle; the soil temperature and water content are monitored in real time by an embedded soil sensor; the leaf surface conductivity is measured at fixed points in typical sample plots by a handheld plant physiological monitor.

[0026] Divide the forest farm into multiple forest areas by the spatial clustering analysis method; Furthermore, the DBSCAN density clustering algorithm is used to perform spatial clustering on the multi-source data of each forest area to form forest area units with similarity; subsequently, the silhouette coefficient is used to evaluate the clustering effect, and the forest area units with fuzzy boundaries are re-divided; finally, the mapping relationship between the forest area units and the geographical coordinates is established through the GIS spatial index to complete the division of the forest farm into multiple forest areas. Among them, the silhouette coefficient is obtained from the ratio of the average distance (cohesion) between each spatial coordinate point in the forest area and other spatial coordinate points in the same cluster and the average distance (separation) from the nearest other cluster points, and its value range is [-1, 1]. The closer the value is to 1, the better the clustering effect.

[0027] The GIS spatial index is used to spatially align the multi-source data of each discrete forest area, and clock drift compensation is used to perform dynamic time calibration on the space to generate the multi-source data of the forest area after spatio-temporal alignment.

[0028] Furthermore, first, a spatial location index of the multi-source data of the forest area is established based on the R-tree index structure, and the multi-source data such as the NDVI vegetation index, soil temperature, leaf surface conductivity, and soil water content are matched according to the geographical coordinates through spatial join operations; subsequently, the timestamps of each data acquisition device are synchronized using the NTP protocol, and the linear interpolation compensation algorithm is applied to the data records with clock drift for time calibration; finally, the spatio-temporal association rules are used to fuse the spatially aligned multi-source data with the time calibration results to generate the multi-source data of the forest area after spatio-temporal alignment.

[0029] S2. Construct a fuzzy cognitive map model, predict the governance priority scores of each forest area based on the multi-source data of the forest area after time alignment, and generate a governance priority heat map; Based on the multi-source data of the aligned forest area, define the pest risk index and ecological sensitivity; Furthermore, the pest risk index is calculated by the weighted combination of the normalized NDVI vegetation index outliers, soil temperature deviation values, and leaf surface conductivity sudden increase values, and its value range is [0, 1]. The larger the value, the higher the pest risk; the ecological sensitivity is determined by the ratio of the soil water content gradient change rate to the historical maximum photochemical efficiency, and its value range is [0.1, 0.9]. The larger the value, the stronger the ecological sensitivity.

[0030] Through partial least squares path analysis, the causal relationship between the pest risk index and ecological sensitivity and the multi-source data of the forest area is established, and the Bayesian network inference algorithm is used to construct the fuzzy cognitive map weight matrix; Furthermore, first, the multi-source data of the forest area after spatio-temporal alignment are used as exogenous latent variables, and the pest risk index and ecological sensitivity are used as endogenous latent variables; the iterative weight algorithm is used to calculate the path coefficients between the latent variables, and the Bootstrap sampling is used to verify the path significance; for the significant paths that pass the test, the standardized regression coefficient is used to quantify the positive impact degree of the NDVI vegetation index on the pest risk index, and the negative impact degree of soil water content on ecological sensitivity and other causal relationships. Based on the causal relationship, the process of constructing the weight matrix of the fuzzy cognitive map using the Bayesian network inference algorithm is as follows: the verified path coefficients are converted into the conditional probability table of the Bayesian network, and the transition probability between nodes is estimated by Markov chain Monte Carlo sampling; the expectation maximization algorithm is applied to optimize the network parameters, and finally a weight matrix of the fuzzy cognitive map reflecting the non-linear relationship between multi-source data and evaluation indicators is generated, and the value range of the matrix elements is limited to the interval [-1, 1], where the positive value represents a promoting relationship and the negative value represents an inhibitory relationship.

[0031] Based on the variation law of historical multi-source data of the forest area, the state transition rules between the pest risk index and ecological sensitivity are defined, and the Sigmoid function is used for iterative update to generate a dynamic state transition mechanism. Furthermore, first, analyze the temporal and spatial variation patterns of the pest risk index and ecological sensitivity in historical data, identify the typical scenarios where the pest risk index increases when the NDVI vegetation index continuously decreases and the soil temperature continuously rises, and the key conditions for the improvement of ecological sensitivity when the soil water content rebounds and the leaf surface conductivity is stable (for example, when the NDVI continuously decreases for 3 months and the monthly average value of the soil temperature rises by 0.5 °C, the pest risk is upgraded); convert the key conditions into state transition rules, and establish a rule base composed of "IF-THEN" conditional statements. The process of iteratively updating the rule base through the Sigmoid function is as follows: input the conditional part of each rule into the Sigmoid function to calculate the activation value, and adjust the slope parameter of the Sigmoid function according to the error between the actual observed value of the current multi-source data of the forest area and the predicted value of the rule; after multiple iterations, output a dynamic state transition mechanism with adaptive ability.

[0032] It should be noted that the variation law of historical multi-source data of the forest area refers to the typical scenarios and correlation patterns of multi-source data of the forest area before and after the occurrence of pests. For example, the NDVI vegetation index will show a continuous decrease of more than 0.15 in the 3 weeks before the outbreak of pests, the soil temperature will remain above 28 °C for 5 consecutive days during the active period of pests, and the fluctuation range of the leaf surface conductivity in the ecologically sensitive area usually exceeds 2 times the standard deviation of the historical average.

[0033] Through the fuzzy dynamic cooperation algorithm, the weight matrix of the fuzzy cognitive map and the dynamic state transition mechanism are combined to form a fuzzy cognitive map model.

[0034] Furthermore, the fuzzy association mapping method is used to establish the mapping relationship between the fuzzy cognitive map weight matrix and the dynamic state transition mechanism. The causal relationship nodes in the fuzzy cognitive map weight matrix are corresponded to the conditional variables of the state transition rules. Then, the fuzzy inference engine is used to synchronously process the causal influence of the weight matrix and the timing constraints of the state transition rules, and the activation degree of each element is quantified through the membership function. Finally, the cooperative optimization algorithm is applied to balance the action intensity of the static weight relationship and the dynamic transition rules. When the difference between the predicted value output by the weight matrix and the result generated by the state transition mechanism exceeds the critical value (the value range is usually [0, 1]), the contribution weights of both are automatically adjusted, so that the fuzzy cognitive map model can not only reflect the causal relationship between multi-source data, but also adapt to the dynamic characteristics of environmental changes.

[0035] The fuzzy cognitive map model predicts the governance priority scores of each forest area based on the aligned multi-source data, pest risk index and ecological sensitivity in the forest area. The expression is: ; where, is the governance priority score of the th forest area, is the Sigmoid function, is the pest risk index, is the ecological sensitivity index, is the multi-source data of the th forest area, is the index variable of the forest area, is the total number of forest areas; Furthermore, first, the interaction between the pest risk index and the ecological sensitivity index reflects the basic threat degree through product normalization. For the multi-source data of each forest area, the dynamic matching degree between the pest risk and the multi-source data of the forest area and the coordination between the ecological sensitivity and the multi-source data of the forest area are respectively evaluated. These two complementary indicators quantify the actual impact through the non-linear adjustment function. Finally, the local impact assessments of all forest areas are integrated to form a comprehensive governance priority judgment. Finally, through the Sigmoid function for normalization processing, the governance priority score of the th forest area is output. The score value range is from 0 to 1, and the larger the value, the higher the governance priority.

[0036] It should be noted that the pest risk index H is a quantified value obtained by weighted summation of historical pest occurrence frequencies, pest population density monitoring data, and host plant distributions; the ecological sensitivity index is a quantified value obtained by the analytic hierarchy process of vegetation type rarity, species diversity, and soil erosion sensitivity, representing the vulnerability of the forest ecosystem to disturbances; both are subject to 0-1 standardization processing, and the larger the value, the higher the pest risk or the stronger the ecological sensitivity respectively.

[0037] Cluster analysis is performed on the historical pest diffusion rate in the forest area through the K-means algorithm, and a low-priority threshold P1 and a high-priority threshold P2 are defined; Furthermore, after the K-means algorithm performs three-class clustering on the historical pest diffusion rate in the forest area, the upper quartile of the lowest-risk category in the clustering results is selected as P1 (value range 0.2 - 0.4), representing the safety critical value of the pest diffusion rate; the lower quartile of the highest-risk category is selected as P2 (value range 0.6 - 0.8), marking the danger critical value of the pest diffusion rate.

[0038] When <P1, mark the current forest area as a low-hazard forest area; for example: the governance priority score of a certain forest area = 0.35, lower than P1 = 0.4.

[0039] When P1 ≤ <P2, mark the current forest area as a medium-hazard forest area; for example: the governance priority score of a certain forest area = 0.55, between P1 = 0.4 and P2 = 0.7.

[0040] When ≥P2, mark the current forest area as a high-hazard forest area; for example: the governance priority score of a certain forest area = 0.8, exceeding P2 = 0.7.

[0041] Probability density interpolation is performed on each marked forest area through spatial kernel density estimation to generate a governance priority heat map.

[0042] Furthermore, based on the spatial distribution coordinates of the marked low-hazard forest areas, medium-hazard forest areas, and high-hazard forest areas, the Gaussian kernel function is used to calculate the hazard level probability density of each geographical location; a regular grid is established within the forest farm range, and the kernel function contribution values of the marked results of the surrounding forest areas are superimposed at each regular grid node; the smoothing degree of the density estimation is controlled by the bandwidth parameter, so that the low-hazard forest areas form a blue cold area, the high-hazard forest areas form a red hot area, and the medium-hazard forest areas form a yellow transition area; finally, a continuous probability density surface reflecting the spatial distribution characteristics of the governance priority is output, that is, the governance priority heat map.

[0043] S3. Based on the governance priority heat map, obtain the maximum photochemical efficiency within the forest farm and record the distribution coordinates, and generate an ecological baseline dataset through spatio-temporal matching; Through the inRange function of the OpenCV library, obtain the spatial mask of the high-hazard forest area in the governance priority heat map; It should be noted that the spatial mask refers to a binary image matrix generated based on the coordinates of the marked high-hazard forest areas and used to identify the spatial distribution range of the high-hazard forest areas; for example, all the grid cells corresponding to the high-hazard forest areas within the coordinates (X1, Y1) to (X2, Y2) in the northwest region of a certain forest farm are assigned a value of 1 in the mask, and the remaining areas are assigned a value of 0, forming a high-hazard area identification map with clear spatial boundaries.

[0044] Through the stratified random sampling method, measure the initial chlorophyll fluorescence and maximum chlorophyll fluorescence of the dark-adapted leaves with an AM fluorometer within the spatial mask of the high-hazard forest area, and use the Genty formula to obtain the maximum photochemical efficiency, and at the same time record the distribution coordinates through GPS; Furthermore, first divide the spatial mask into several sampling units with equal areas, and randomly select 3 - 5 sampling points within each unit; use an AM fluorometer to measure the initial chlorophyll fluorescence and maximum chlorophyll fluorescence of the dark-adapted leaves at the sampling points; calculate the maximum photochemical efficiency of each sampling point according to Genty; at the same time, use a GPS device to record the precise coordinates of each sampling point; finally, obtain a data set containing the measured values of the maximum photochemical efficiency and their corresponding spatial positions.

[0045] It should be noted that the dark-adapted leaves refer to the plant leaves that have reached a stable physiological state after sufficient dark treatment; for example, the chlorophyll fluorescence measured after covering the leaves to be measured with a light-shielding clip for 30 minutes before measurement to completely relax their photosynthetic mechanisms.

[0046] Use the Kriging spatial interpolation method to perform spatial matching on the NDVI vegetation index, maximum photochemical efficiency, and distribution coordinates within the forest farm, and combine the improved dynamic time warping algorithm for time series alignment to generate an ecological baseline dataset; It should be noted that the improvement process of the dynamic time warping algorithm is as follows: First, expand the single time warping path of the dynamic time warping algorithm into a spatio-temporal coupled warping path (spatially adjacent nodes are forced to share > 60% of the time offset); secondly, design a two-channel distance metric function, the LSTM channel calculates the cosine similarity of the seasonal trend, and the Transformer channel detects local mutation points (such as pest outbreak weeks) through self-attention, and the two are dynamically fused through spatial entropy weights; finally, an improved dynamic time warping algorithm is formed to improve the spatio-temporal alignment accuracy and ecological event detection ability.

[0047] S4. Based on the priority heat map, a multi-criteria decision-making algorithm is used to generate spraying instructions, and the UAV is scheduled according to the spraying instructions to perform differential spraying; The spatial mask of the high-hazard forest area and the ecological baseline dataset are registered in resolution through the GDAL library to generate a multi-variable raster stack; Furthermore, first, the raster data of the spatial mask of the high-hazard forest area and raster layers such as the NDVI vegetation index and the maximum photochemical efficiency in the ecological baseline dataset are read through the Warper tool of the GDAL library; the nearest neighbor resampling method is used to unify each layer to the same spatial resolution; the pixel positions of all layers are strictly aligned through geographic coordinate transformation; finally, the registered spatial mask and each layer of the ecological baseline data are stacked in band order to form a multi-variable raster stack containing spatial position information and multi-dimensional ecological parameters.

[0048] The band splitting method is used to separate three criterion layers from the multi-variable raster stack; the three criterion layers include the hazard degree criterion layer, the ecological sensitivity criterion layer, and the spraying feasibility criterion layer; Furthermore, first, the maximum photochemical efficiency is used as the first band, the governance priority heat map is used as the second band, the surface slope data obtained based on the DEM is used as the third band, and the distance between each pixel and the water source in each multi-variable raster stack obtained through hydrological analysis is used as the fourth band; then, the R-channel intensity value data (in the range of 0-255) in the governance priority heat map of the second band is extracted through the band index extraction method as the hazard degree criterion layer; then, based on the maximum photochemical efficiency data of the first band, an ecological sensitivity threshold (with a value range of 0.1-0.64) is defined based on the historical maximum photochemical efficiency and binary segmentation is performed to generate the ecological sensitivity criterion layer; finally, a dual conditional judgment is performed on the third-band surface slope data and the fourth-band distance between the pixel and the water source (example: slope < 25° and water source distance > 50m), and the spraying feasibility criterion layer is generated through pixel-by-pixel logical AND Boolean operation.

[0049] The analytic hierarchy process is used to assign weights to the hazard degree criterion layer, the ecological sensitivity criterion layer, and the spraying feasibility criterion layer; Furthermore, first, a judgment matrix containing the hazard degree criterion layer, the ecological sensitivity criterion layer, and the spraying feasibility criterion layer is constructed, and the relative importance degree between each criterion layer is determined through pairwise comparison; the eigenvector method is used to calculate the maximum eigenvalue of the judgment matrix and its corresponding normalized eigenvector to obtain the initial weights of each criterion layer; a consistency test is performed to ensure the logical rationality of the judgment matrix, and the comparison scale is adjusted if necessary; finally, the weight assignment result of the criterion layer that meets the consistency requirement is output, and the sum of the weights of the three criterion layers is strictly equal to 1. For example, the weight of the hazard degree criterion layer is 0.5, the weight of the ecological sensitivity criterion layer is 0.3, and the weight of the spraying feasibility criterion layer is 0.2.

[0050] According to the weights of each criterion layer, MCDA is used for spatial weighted overlay analysis to obtain the coordinates of the forest areas to be sprayed and the amount of pesticides to be applied, and the treatment path is generated based on the governance priority scores of each forest area. Furthermore, based on the weights of the hazard level criterion layer, ecological sensitivity criterion layer, and pesticide application feasibility criterion layer determined by the analytic hierarchy process, standardize each ecological parameter in the multi-variable raster stack to eliminate the dimension difference. Using the weighted linear combination method, multiply the spatial data corresponding to each criterion layer by its weight and then overlay them to generate the pesticide application priority distribution surface. Divide the continuous priority values into several pesticide application amount levels through the natural breaks classification method. At the same time, considering the temporal dimension characteristics of the governance priority score, use the shortest path algorithm to plan a governance path network that takes into account both spatial efficiency and temporal benefits. Finally, output a decision result set containing the geographical coordinates of the forest areas to be sprayed, the pesticide application amount level, and the path order. For example, a forest area with high hazard (coordinates X: 123.45, Y: 67.89) is analyzed to obtain a pesticide application amount level of 3 and is arranged to be executed at the third operation point of the second treatment path.

[0051] Map the coordinates of the forest areas to be sprayed, the amount of pesticides to be applied, and the treatment path to pesticide application instructions through multi-criteria decision mapping, and dispatch drones to perform differential pesticide application according to the pesticide application instructions.

[0052] Furthermore, convert the set of coordinates of the forest areas to be sprayed into a sequence of waypoints recognizable by the drone, and establish a mapping relationship between the spatial position and the operation parameters. The pesticide application amount level is mapped to specific operation parameters such as spraying flow rate, flight altitude, and spraying width through a preset conversion rule, and the priority of the treatment path is converted into the waypoint access order and time scheduling plan. Integrate the spatial position information, operation parameter settings, and time scheduling requirements to generate a structured pesticide application instruction set to ensure that the drone can accurately perform differential pesticide application operations. For example, the pesticide application instructions corresponding to the waypoint (X: 123.45, Y: 67.89) include parameters such as a spraying flow rate of 250 ml / min, a flight altitude of 4.5 m, and a spraying width of 3.2 m, and are marked as the third operation task to be executed at 10:00 am the next day.

[0053] S5. Obtain the pesticide application amount correction coefficient and combine it with the ecological baseline data set, calculate the ecological restoration score through the ecological resonance index method, and adjust the pesticide application instructions in real time according to the ecological restoration score.

[0054] Based on the pesticide application amount in the pesticide application instructions and the maximum photochemical efficiency in the ecological baseline data set, calculate the pesticide application amount correction coefficient through the dose-response surface method. The expression is: ; where is the pesticide application amount correction coefficient for the forest area in the time period. is the dose-response curvature parameter, is the threshold of photochemical efficiency (the value range is 0.75 - 0.85), is the maximum photochemical efficiency, is the application rate of the forest area during the time period; Furthermore, based on the maximum photochemical efficiency in the ecological baseline dataset and the application rate in the dosing instruction, an S-shaped dose-response function relationship is constructed; the maximum photochemical efficiency is compared with the preset photochemical efficiency threshold ; the steepness of the response curve is controlled by the curvature parameter ; when the maximum photochemical efficiency deviates from the threshold , the influence of the application rate on the correction coefficient shows a non-linear change characteristic; the finally output correction coefficient has a value range between 0 and 1, reflecting the dosing adjustment ratio at the current photosynthesis efficiency level. For example, when the maximum photochemical efficiency

[0055] measured in a certain forest area during the monitoring period is = 0.82, the corresponding dosing correction coefficient is = 0.86. It should be noted that the photochemical efficiency threshold is defined by statistical analysis using the percentile method based on the maximum photochemical efficiency of historical vegetation in the forest farm, and the value range is 0.75 - 0.85; Based on the dosing correction coefficient and the ecological baseline dataset, the ecological restoration score of each forest area is calculated by the ecological resonance index method, and the expression is: where is the ecological restoration score of the th forest area, is the value of the th ecological baseline data of the th forest area, is the historical average value (in the same season and the same forest area) of the th ecological baseline data item, is the sign function (used to judge whether the difference is positive, negative or zero).

[0056] Furthermore, obtain the ecological baseline data of the nth forest area , and conduct a standardized comparison with the historical average and historical standard deviation in the same forest area in the same season; divide the deviation degree of each ecological baseline data by to obtain the standardized difference, and then multiply it by the corresponding application rate correction coefficient ; retain the deviation direction information through the sign function to distinguish positive improvement and negative degradation; finally, calculate the average value of the adjusted results of all ecological baseline data to obtain the score reflecting the overall ecological restoration status of the forest area . For example, after applying pesticides in a certain forest area, it is measured that the NDVI vegetation index has increased by 1.2 standard deviations compared with the historical average, and the soil moisture content has decreased by 0.8 standard deviations. After calculation, the ecological restoration score =0.45.

[0057] Based on the ecological restoration scores of each forest area, identify and label the ecological restoration deviation status through dynamic clustering analysis; the ecological restoration deviation status includes normal, warning, and abnormal; Furthermore, first collect the ecological restoration scores of each forest area in consecutive time periods , and analyze the score distribution characteristics using the density-based clustering algorithm; determine the optimal number of clusters through the silhouette coefficient, and divide the forest areas into three categories: the "normal" state with concentrated distribution of ecological restoration scores, the "warning" state with high score distribution, and the "abnormal" state with low score distribution; consider the time continuity constraint during the clustering process to ensure the temporal consistency of the state labeling results. For example, the ecological restoration scores of a certain forest area are stable in the interval of 0.4 - 0.6 in three consecutive monitoring periods and are clustered into the "normal" state.

[0058] Based on the ecological restoration deviation status, update the application rate correction coefficient through the incremental learning algorithm, and optimize the pesticide application instructions according to the updated application rate correction coefficient.

[0059] Furthermore, according to the ecological restoration deviation status labeling results obtained from the dynamic clustering analysis, for the forest areas in the "warning" and "abnormal" states, use the online gradient descent method to adjust the curvature parameter in the dose-response surface; calculate the error between the current application rate correction coefficient and the target restoration score, and update the model parameters through backpropagation; the updated dose-response surface recalculates the application rate correction coefficient of each forest area, and then generates optimized pesticide application instructions considering the dynamic characteristics of ecological restoration. For example, the application rate correction coefficient Adjust from 0.7 to 0.8, and correspondingly increase the application rate of pesticides by 15%.

[0060] This embodiment also provides an information-based comprehensive management system for forestry pests, including: a data collection module, a priority division module, a baseline generation module, an instruction generation module, and an ecological restoration evaluation module; The data collection module is used to divide the forest farm into multiple forest areas, collect multi-source data of each forest area, and perform time alignment; The priority division module is used to construct a fuzzy cognitive map model, predict the governance priority scores of each forest area based on the multi-source data of the forest areas after time alignment, and generate a governance priority heat map; The baseline generation module is used to obtain the maximum photochemical efficiency in the forest farm based on the governance priority heat map, record the distribution coordinates, and generate an ecological baseline data set through spatio-temporal matching; The instruction generation module is used to generate pesticide application instructions according to the priority heat map by using a multi-criteria decision-making algorithm, and dispatch drones to perform differential pesticide application according to the pesticide application instructions; The ecological restoration evaluation module is used to obtain the pesticide application rate correction coefficient, combine it with the ecological baseline data set, calculate the ecological restoration score through the ecological resonance index method, and adjust the pesticide application instructions in real time according to the ecological restoration score.

[0061] This embodiment also provides a computer device applicable to the information-based comprehensive management method for forestry pests, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the information-based comprehensive management method for forestry pests proposed in the above embodiment.

[0062] The computer device can be a terminal. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball, or a touchpad set on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0063] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for comprehensive informatization management of forestry pests as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, magnetic disk or optical disc.

[0064] In summary, through the steps of constructing a fuzzy cognitive map model, the present invention realizes the dynamic coupling analysis of pest risk index and ecological sensitivity, and improves the accuracy and adaptability of risk assessment; by calculating the ecological restoration score of each forest area through the ecological resonance index method, the present invention realizes the dynamic correlation assessment of pesticide application effect and ecological restoration status, and enhances the pertinence and timeliness of prevention and control measures.

[0065] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. An information-based integrated management method for forestry pests, characterized in that: including, divide the forest farm into multiple forest areas, collect multi-source data of each forest area and align them in time; the multi-source data of the forest area includes NDVI vegetation index, soil temperature, leaf surface conductivity and soil water content; construct a fuzzy cognitive map model, predict the governance priority scores of each forest area based on the multi-source data of the forest area after time alignment, and generate a governance priority heat map; based on the governance priority heat map, obtain the maximum photochemical efficiency in the forest farm and record the distribution coordinates, and generate an ecological baseline dataset through spatio-temporal matching; according to the priority heat map, use a multi-criteria decision-making algorithm to generate a pesticide application instruction, and dispatch an unmanned aerial vehicle to perform differential pesticide application according to the pesticide application instruction; obtain a pesticide application amount correction coefficient and combine it with the ecological baseline dataset, calculate an ecological restoration score through the ecological resonance index method, and adjust the pesticide application instruction in real time according to the ecological restoration score.

2. The integrated information management method for forestry pests as described in claim 1, characterized in that: divide the forest farm into multiple forest areas by spatial clustering analysis method; use GIS spatial index to spatially align the multi-source data of each discrete forest area, and perform dynamic time calibration on the space by clock drift compensation to generate the multi-source data of the forest area after spatio-temporal alignment.

3. The integrated information management method for forestry pests as described in claim 1, wherein: The steps of constructing the fuzzy cognitive map model are as follows: define the pest risk index and ecological sensitivity; through partial least squares path analysis, establish the causal relationship between the pest risk index and ecological sensitivity and the multi-source data of the forest area, and use the Bayesian network inference algorithm to construct the fuzzy cognitive map weight matrix; define the state transition rule between the pest risk index and ecological sensitivity, and perform iterative update through Sigmoid to generate a dynamic state transition mechanism; through the fuzzy dynamic cooperation algorithm, combine the fuzzy cognitive map weight matrix with the dynamic state transition mechanism to form a fuzzy cognitive map model.

4. The integrated information management method for forestry pests as described in claim 3, characterized in that: The steps of predicting the governance priority scores of each forest area and generating a governance priority heat map are as follows: the fuzzy cognitive map model predicts the governance priority scores of each forest area based on the multi-source data of the forest area after alignment, the pest risk index and ecological sensitivity; perform clustering analysis on the historical pest diffusion rate of the forest area by the K-means algorithm, and define the low-priority threshold and high-priority threshold; mark the hazards of each forest area according to the comparison results of the governance priority scores of each forest area with the low-priority threshold and high-priority threshold; perform probability density interpolation on the marked forest areas through spatial kernel density estimation to generate a governance priority heat map.

5. The integrated information management method for forestry pests as claimed in claim 1, characterized in that: The steps of obtaining the maximum photochemical efficiency in the forest farm and recording the distribution coordinates and generating an ecological baseline dataset through spatio-temporal matching are as follows: obtain the spatial mask of the high-hazard forest area in the governance priority heat map, and through the stratified random sampling method, measure the initial chlorophyll fluorescence and maximum chlorophyll fluorescence of the dark-adapted leaves in the spatial mask of the high-hazard forest area by an AM fluorometer, and use the Genty formula to obtain the maximum photochemical efficiency, and record the distribution coordinates through GPS at the same time; use Kriging spatial interpolation method to perform spatial matching on the NDVI vegetation index, maximum photochemical efficiency and distribution coordinates in the forest farm, and combine with the improved dynamic time warping algorithm to perform time series alignment to generate an ecological baseline dataset.

6. The integrated information management method for forestry pests as described in claim 5, characterized in that: The steps of generating spraying instructions using the multi-criteria decision-making algorithm are as follows: Through the GDAL library, the spatial mask of the high-hazard forest area is registered with the ecological baseline dataset in terms of resolution to generate a multi-variable raster stack; The band splitting method is used to separate the three criterion layers from the multi-variable raster stack, and the analytic hierarchy process is used for weight assignment; According to the weights of each criterion layer, MCDA is used for spatial weighted overlay analysis to obtain the coordinates and spraying amounts of the forest areas to be sprayed, and a treatment path is generated based on the governance priority scores of each forest area; The spraying instructions are mapped through multi-criteria decision mapping based on the coordinates, spraying amounts, and treatment paths of the forest areas to be sprayed; 7. The integrated information management method for forestry pests as described in claim 1, characterized in that: The steps of calculating the ecological restoration score by the ecological resonance index method and making real-time adjustments to the spraying instructions according to the ecological restoration score are as follows: Based on the spraying amount in the spraying instructions and the maximum photochemical efficiency in the ecological baseline dataset, the spraying amount correction coefficient is calculated by the dose-response surface method, and combined with the ecological baseline dataset, the ecological restoration scores of each forest area are calculated by the ecological resonance index method; Based on the ecological restoration scores of each forest area, the ecological restoration deviation status is identified and marked through dynamic clustering analysis, and the spraying amount correction coefficient is updated by the incremental learning algorithm, and the spraying instructions are optimized according to the updated spraying amount correction coefficient; 8. An information-based integrated management system for forestry pests, based on the information-based integrated management method for forestry pests according to any one of claims 1 to 7, characterized in that: It includes a data acquisition module, a priority division module, a baseline generation module, an instruction generation module, and an ecological restoration evaluation module; The data acquisition module is used to divide the forest farm into multiple forest areas, collect multi-source data of each forest area, and perform time alignment; The priority division module is used to construct a fuzzy cognitive map model, predict the governance priority scores of each forest area based on the multi-source data of the forest areas after time alignment, and generate a governance priority heat map; The baseline generation module is used to obtain the maximum photochemical efficiency within the forest farm and record the distribution coordinates based on the governance priority heat map, and generate an ecological baseline dataset through spatio-temporal matching; The instruction generation module is used to generate spraying instructions using the multi-criteria decision-making algorithm according to the priority heat map, and dispatch drones to perform differential spraying according to the spraying instructions; The ecological restoration evaluation module is used to obtain the spraying amount correction coefficient, combine it with the ecological baseline dataset, calculate the ecological restoration score by the ecological resonance index method, and make real-time adjustments to the spraying instructions according to the ecological restoration score; 9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the integrated information management method for forest pests and diseases according to any one of claims 1 to 7; 10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the integrated information management method for forest pests and diseases according to any one of claims 1 to 7.

Images