Scene-adaptive bird disease directional management and control method and platform

Abstract

The invention discloses a scene self-adaption bird disease directional management and control method and platform, and relates to the technical field of data processing, and the method comprises the steps: deploying a bird monitoring and sensing network in a target scene region, and collecting and obtaining a target bird disease distribution data set; performing trend prediction on the target bird disease distribution data set, and outputting bird disease trend prediction parameters; obtaining a bird disease ecological management strategy library; determining target bird disease treatment strategy parameters; obtaining a bird disease treatment effect coefficient; and the target bird disease treatment strategy parameters are optimized and updated to obtain bird disease treatment strategy optimization parameters, and bird disease treatment management and control are carried out. The technical problems that in the prior art, real-time monitoring and sensing of bird activities in a target area are lacked, the bird outbreak situation cannot be predicted and dealt with in advance, the bird control pertinence is poor, and the control result is hysteretic are solved, and the technical effect of improving bird control scientificity, timeliness and stability is achieved.

Description

Technical Field

[0001] The present application relates to the technical field related to data processing, and specifically to a method and platform for targeted bird control under scenario adaptation. Background Art

[0002] In many fields such as agricultural production, airport operation, and power facility maintenance, bird activities often cause a series of safety and economic problems, such as crop damage, increased risk of flight accidents, and power line failures. With the development of technologies such as the Internet of Things and artificial intelligence, the use of intelligent means to accurately monitor, predict trends, and conduct targeted governance of bird infestations has become an important part. However, existing governance solutions still have significant deficiencies in scene adaptability, data-driven prediction capabilities, and governance strategy optimization. Existing bird infestation governance usually relies on manual inspections and simple driving devices, which are difficult to adapt to complex and diverse scene requirements. Fixed strategies are difficult to cope with dynamically changing bird infestation risks, which is not only inefficient, but also difficult to sustain governance effects. Especially in complex scenes, bird activity patterns are affected by multiple factors such as the environment, seasons, and climate, and may even cause additional negative impacts on the ecological environment.

[0003] Therefore, at the current stage, the technologies related to bird control and management lack real-time monitoring and perception of bird activities in the target area, and are unable to predict trends in advance to respond to bird outbreaks, resulting in technical problems such as poor targeted bird control and delayed control results. Summary of the invention

[0004] This application solves the technical problems in the prior art of lack of real-time monitoring and perception of bird activities in the target area, inability to make early trend forecasts to respond to bird infestation outbreaks, resulting in poor targeted bird infestation management and delayed control results by providing a scene-adaptive targeted bird infestation management and control method and platform, thereby achieving the technical effect of improving the scientificity, timeliness and stability of bird infestation management.

[0005] The present application provides a method for directional control of bird infestation under scene adaptation, the method comprising: deploying a bird monitoring and sensing network in a target scene area, collecting and acquiring a target bird infestation distribution data set through the bird monitoring and sensing network; calling a bird activity adaptive prediction network set to perform trend prediction on the target bird infestation distribution data set, and outputting bird infestation trend prediction parameters; if the bird infestation trend prediction parameters exceed a preset bird infestation risk threshold, a bird infestation emergency response mechanism is triggered, and a bird infestation ecological control strategy library is obtained according to the bird infestation emergency response mechanism; based on the bird infestation ecological control strategy library, a directional strategy analysis is performed on the bird infestation trend prediction parameters to determine target bird infestation control strategy parameters; according to the target bird infestation control strategy parameters, directional control of bird infestation and effect evaluation are performed on the target scene area to obtain a bird infestation control effect coefficient; based on the bird infestation control effect coefficient, the target bird infestation control strategy parameters are tuned and updated to obtain bird infestation control strategy optimization parameters, and bird infestation control is performed through the bird infestation control strategy optimization parameters.

[0006] In a possible implementation, the output bird infestation trend prediction parameters perform the following processing: utilize big data technology to mine and obtain a bird infestation distribution data set, classify and train the bird infestation distribution data set based on bird infestation scene factor information, and establish a bird activity prediction network library; classify and mark the target scene area according to the bird infestation scene factor information, and determine a target scene calibration parameter combination; match and call the bird activity prediction network library based on the target scene calibration parameter combination to obtain a bird activity adaptive prediction network set; perform trend prediction on the target bird infestation distribution data set based on the bird activity adaptive prediction network set, and output the bird infestation trend prediction parameters.

[0007] In a possible implementation, the establishment of a bird activity prediction network library performs the following processing: clustering and marking the bird infestation scene factor information to determine the bird infestation scene clustering factors, wherein the bird infestation scene clustering factors include terrain environment, climate characteristics, bird species, and activity patterns; extracting and integrating parameters of the bird infestation distribution data set based on the bird infestation scene clustering factors to obtain a scene clustering factor parameter set; associating and dividing the bird infestation distribution data set according to the scene clustering factor parameter set to obtain a bird infestation scene parameter distribution data set; performing prediction training on the bird infestation scene parameter distribution data sets respectively to obtain a bird activity prediction network set, and identifying and establishing the bird activity prediction network library based on the bird activity prediction network set.

[0008] In a possible implementation, the bird infestation scene parameter distribution data set is obtained, and the following processing is further performed: the bird infestation distribution data set is extracted and divided according to the scene clustering factor parameter set to obtain the scene factor parameter distribution data set; the bird infestation impact degree is analyzed for the scene clustering factor parameter set respectively to determine the scene factor parameter impact coefficient set; the scene parameter association data depth is determined according to the scene factor parameter impact coefficient set; the scene factor parameter distribution data set is associated with the depth of the scene parameter association data and expanded to obtain the bird infestation scene parameter distribution data set.

[0009] In a possible implementation, the determination of the target bird infestation control strategy parameters also performs the following processing: based on the bird infestation ecological control strategy library, the bird infestation trend prediction parameters are matched with a directional strategy to obtain a target bird infestation control directional strategy; according to the target bird infestation control directional strategy, a bird infestation control strategy parameter space is constructed; using the bird infestation trend prediction parameters as constraint parameters, strategy analysis is performed within the bird infestation control strategy parameter space to generate a control strategy parameter selection threshold; constructing a bird infestation control effect fitness function, using the bird infestation control effect fitness function to perform global optimization within the control strategy parameter selection threshold to determine the target bird infestation control strategy parameters.

[0010] In a possible implementation, the determination of the target bird infestation control strategy parameters further performs the following processing: randomly selecting multiple control strategy parameters within the control strategy parameter selection threshold, using the bird infestation control effect fitness function to evaluate the fitness of the multiple control strategy parameters to obtain multiple strategy parameter fitnesses; initializing particle swarm parameters, the particle swarm parameters including particle position and particle velocity; performing gradient ascent calculation on the fitness of the multiple strategy parameters to obtain multi-parameter gradient ascent calculation results, and determining the strategy parameter optimization direction based on the multi-parameter gradient ascent calculation results; evaluating and iteratively updating the particle swarm parameters based on the strategy parameter optimization direction until a preset termination condition is met, and optimizing and determining the particle with the largest fitness as the target bird infestation control strategy parameter.

[0011] In a possible implementation, the bird control strategy optimization parameters are obtained, and the following processing is also performed: based on the bird control effect coefficient, the target bird control strategy parameters are fine-tuned and analyzed for variation to obtain parameter variation optimization rules; according to the parameter variation optimization rules, the target bird control strategy parameters are mutated, optimized and expanded to obtain a bird control strategy variation parameter set; based on the bird control effect fitness function, the bird control strategy variation parameter set is compared and optimized to obtain the bird control strategy optimization parameters.

[0012] The present application also provides a bird infestation directional management and control platform under scene adaptation, including: a target bird infestation distribution data set acquisition module, which is used to deploy a bird monitoring perception network in the target scene area, and collect and acquire the target bird infestation distribution data set through the bird monitoring perception network; a bird infestation trend prediction parameter output module, which is used to call the bird activity adaptive prediction network set to perform trend prediction on the target bird infestation distribution data set, and output the bird infestation trend prediction parameters; a bird infestation ecological management strategy library acquisition module, which is used to trigger the bird infestation emergency response mechanism if the bird infestation trend prediction parameter exceeds the preset bird infestation risk threshold, and obtain the bird infestation ecological management strategy according to the bird infestation emergency response mechanism. a bird infestation ecological management strategy library; a target bird infestation management strategy parameter determination module, which is used to perform directional strategy analysis on the bird infestation trend prediction parameters based on the bird infestation ecological management strategy library, and determine the target bird infestation management strategy parameters; a bird infestation management effect coefficient acquisition module, which is used to perform directional bird infestation management and effect evaluation on the target scene area according to the target bird infestation management strategy parameters, and obtain the bird infestation management effect coefficient; a bird infestation management strategy optimization parameter acquisition module, which is used to tune and update the target bird infestation management strategy parameters based on the bird infestation management effect coefficient, obtain the bird infestation management strategy optimization parameters, and perform bird infestation management and control through the bird infestation management strategy optimization parameters.

[0013] Through the scene-adaptive targeted bird control method and platform proposed in this application, a bird monitoring and sensing network is deployed in the target scene area to collect and obtain the target bird distribution data set; trend prediction is performed on the target bird distribution data set, and bird trend prediction parameters are output; a bird ecological control strategy library is obtained; target bird control strategy parameters are determined; bird control effect coefficients are obtained; target bird control strategy parameters are tuned and updated to obtain optimized bird control strategy parameters, and bird control is performed. This solves the technical problems in the prior art of lack of real-time monitoring and perception of bird activities in the target area, inability to make advance trend predictions to respond to bird outbreaks, resulting in poor targeting of bird control and delayed control results, and achieves the technical effect of improving the scientificity, timeliness and stability of bird control. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure are briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the platform according to the embodiment of the present application. It should be understood that the previous or following operations are not necessarily performed precisely in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0015] Figure 1A schematic flow chart of a method for targeted bird infestation control and management under scenario adaptation provided in an embodiment of the present application.

[0016] Figure 2 A schematic diagram of the structure of a bird infestation directional control and management platform under scenario adaptation provided in an embodiment of the present application.

[0017] Explanation of the accompanying reference numerals: target bird infestation distribution data set acquisition module 10, bird infestation trend prediction parameter output module 20, bird infestation ecological management strategy library acquisition module 30, target bird infestation management strategy parameter determination module 40, bird infestation management effect coefficient acquisition module 50, bird infestation management strategy optimization parameter acquisition module 60. DETAILED DESCRIPTION

[0018] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0019] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0020] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, platform, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0021] The present application embodiment provides a method for directional bird control under scene adaptation, such as Figure 1 As shown, the method includes: Step S100: deploy a bird monitoring and sensing network in a target scene area, and collect a target bird infestation distribution data set through the bird monitoring and sensing network.

[0022] Preferably, an intelligent bird activity monitoring and sensing network is established in the scene that needs to be governed (such as agricultural areas, airport periphery, power facilities, etc.), wherein the bird monitoring and sensing network is composed of a variety of sensor devices and data transmission modules, which are used to collect data related to bird activities in real time. Specifically, the sensor devices may include high-definition cameras for real-time capture of bird activity images, combined with computer vision algorithms to identify bird species, quantity and activity trajectories; sound sensors for capturing bird calls, and analyzing bird species and behavior patterns through sound recognition technology; environmental sensors for detecting environmental factors in the area (such as temperature, humidity, wind speed, light intensity, etc.) to assist in analyzing the environmental inducements of bird activities. The data collected by the deployment monitoring and sensing network are collected and processed to generate a target bird disease distribution data set, which may include a bird activity heat map, indicating the concentrated area and distribution of bird activities; time dimension distribution, statistics of bird activity patterns at different time periods (morning, daytime, dusk); bird species distribution, identification of the main bird species and quantity in the target area; dynamic behavior characteristics, recording key activity characteristics such as bird flight trajectories, clustering behaviors, and foraging points. Through the collaborative work of sensors, a complete bird monitoring and sensing system is formed to provide accurate data for subsequent trend predictions.

[0023] Step S200, calling the bird activity adaptive prediction network set to perform trend prediction on the target bird infestation distribution data set, and outputting bird infestation trend prediction parameters.

[0024] Preferably, a plurality of different types of prediction models in the adaptive prediction network set for bird activity are called to analyze the target bird infestation distribution data set, explore the patterns therein and predict future bird activity trends and risks. Specifically, the adaptive prediction network set for bird activity usually contains multiple prediction models that can work independently or collaboratively, including but not limited to time series prediction models (such as LSTM, GRU) for analyzing historical bird activity data and predicting short-term or long-term bird activity trends, spatial distribution prediction models (such as CNN, ConvLSTM) for identifying and predicting changes in the spatial distribution of bird activities, such as the transfer or expansion of activity hotspots, and environmental adaptation models (such as adaptive regression models) for combining environmental factors (temperature, humidity, light, etc.) and historical data to predict changes in bird behavior under specific environmental conditions; by continuously receiving new monitoring data, the model will dynamically adjust its own parameters or structure (for example, using transfer learning or incremental learning methods) to adapt to the changing characteristics of the target scene.

[0025] Preferably, the data of the target bird infestation distribution data set are cleaned and normalized, and the noise data is removed as the input of the prediction model. The spatial distribution characteristics are extracted using a convolutional network, and the temporal trend of bird activities is analyzed using a time series network. Environmental factors and historical behavior patterns are used as auxiliary inputs of the model. Combined with the output results of these prediction models, bird infestation trend prediction parameters are generated, including bird activity intensity, such as the changing trend of activity frequency or the number of individuals; changes in activity areas, the diffusion, transfer or shrinkage trend of hot spots; risk level, the bird infestation risk level is calculated based on the predicted data and the preset threshold, thereby significantly improving the accuracy and efficiency of bird infestation management.

[0026] Furthermore, step S200 also includes step S210, using big data technology to mine and obtain a bird infestation distribution data set, classifying and training the bird infestation distribution data set based on the bird infestation scene factor information, and establishing a bird activity prediction network library; step S220, classifying and marking the target scene area according to the bird infestation scene factor information, and determining the target scene calibration parameter combination; step S230, matching and calling the bird activity prediction network library based on the target scene calibration parameter combination to obtain a bird activity adaptive prediction network set; step S240, performing trend prediction on the target bird infestation distribution data set based on the bird activity adaptive prediction network set, and outputting the bird infestation trend prediction parameters.

[0027] Preferably, a big data mining algorithm (such as cluster analysis and pattern recognition) is used to analyze the characteristics of bird activities to generate a bird infestation distribution data set, including multidimensional data such as spatial distribution, temporal characteristics, and species information. Then, the bird infestation distribution data set is classified and trained based on the bird infestation scene factor information, wherein the bird infestation scene factor information includes scene characteristics, such as specific scenes such as airports, farmlands, and power facilities, environmental conditions, temperature, humidity, light, wind speed and other external factors, bird characteristics, species, activity habits, foraging and roosting behaviors. Classification training refers to labeling the bird infestation distribution data set according to the scene factor information (for example, "airport-high-risk area" or "farmland-peak migration period"), using a machine learning algorithm (such as random forest, SVM) or a deep learning algorithm (such as CNN, LSTM) to train the classified data, output a prediction model, and form a bird activity prediction network library, wherein each model is optimized for different scene factors, and then a bird activity prediction network library is constructed, which includes prediction models suitable for multiple scenes.

[0028] Preferably, the actual situation of the target scene is collected, including the geographical location, environmental conditions, monitoring data, etc., and classified according to the scene factor information. The classification marking content includes the scene category, such as "airport-runway periphery" or "farmland-fruit tree area", risk level, high risk, medium risk, low risk, specific environmental parameters, such as wind speed, light intensity and other dynamic environmental information, combined with the above marking information, to generate a scene parameter combination (such as scene type, dynamic environmental conditions, bird species distribution, etc.), according to the factor classification parameter combination of the target scene area, call the prediction network model of the corresponding parameter type, call multiple prediction networks (for example, a model specializing in The adaptive prediction network set of bird activities is composed of two sets (one focuses on spatial distribution prediction, the other focuses on temporal trend prediction). The types and quantities of prediction models can be dynamically adjusted according to scenario requirements to improve prediction accuracy and adaptability. Finally, the target bird distribution data set is input into the adaptive prediction network set of bird activities for analysis and prediction to generate specific bird trend prediction parameters, including activity intensity (such as changing trends in quantity and density), activity range (such as distribution of hot spots), risk level (such as high, medium and low risk areas), and dynamic influencing factors (such as the impact of weather changes on bird activities), so as to realize accurate bird activity trend prediction driven by big data.

[0029] Furthermore, step S210 also includes step S211, clustering and marking the bird-infestation scene factor information to determine the bird-infestation scene clustering factors, wherein the bird-infestation scene clustering factors include terrain environment, climate characteristics, bird species and activity patterns; step S212, extracting and integrating parameters of the bird-infestation distribution data set based on the bird-infestation scene clustering factors to obtain a scene clustering factor parameter set; step S213, associating and dividing the bird-infestation distribution data set according to the scene clustering factor parameter set to obtain a bird-infestation scene parameter distribution data set; step S214, respectively performing predictive training on the bird-infestation scene parameter distribution data sets to obtain a bird activity prediction network set, and identifying and establishing the bird activity prediction network library based on the bird activity prediction network set.

[0030] Preferably, clustering and marking the bird-infested scene factor information refers to using a clustering algorithm (such as K-means, DBSCAN, etc.) to classify a large amount of scene factor information, find out scene categories with similar characteristics, such as "wetland-high humidity-migratory bird migration area" or "mountain-low temperature-resident bird foraging area", and determine the clustering factors of the bird-infested scene, such as terrain environment (plains, mountains, forests, wetlands, etc.), climate characteristics (temperature, humidity, wind speed, rainfall, etc.), bird species (migratory birds, migratory birds, resident birds, etc.), activity patterns (migration time, peak activity time), etc. Then, data related to scene clustering factors are extracted from the bird infestation distribution dataset, such as terrain-related data, GPS location, regional area, and surface type; climate-related data, historical temperature, humidity, and rainfall frequency; bird species-related data, species classification, and migration schedule; activity patterns-related data, activity trajectories, density changes, and time distribution; these parameters are integrated to generate a set of scene clustering factor parameters, with specific classification parameters for each bird infestation scene clustering factor, such as bird species parameters including sparrows, pigeons, etc.

[0031] Preferably, the bird infestation distribution data set is associated and divided according to the scene clustering factor parameter set, which means that the bird infestation distribution data set is divided into multiple subsets according to the scene clustering factor parameter set, and each subset corresponds to a specific scene type, so as to obtain a bird infestation scene parameter distribution data set, and each subset contains multidimensional data related to a specific scene, forming a parameter distribution with scene specificity. Finally, for each scene parameter distribution data set, a time series model (such as LSTM, ARIMA), a spatial distribution model (such as CNN, ConvLSTM) or a hybrid model (such as Transformer) is used to train an independent prediction model. Each scene factor parameter is trained to obtain a small network model, such as a sparrow activity prediction network, to form a prediction network set. According to the model characteristics and applicable scenarios of the prediction network set, an identifier is added to each model (such as "wetland-high humidity-migration prediction model"), and all prediction networks are stored according to scene classification to form a structured network library, that is, a bird activity prediction network library is established.

[0032] Furthermore, step S213 also includes step S213a, extracting and dividing the bird infestation distribution data set according to the scene clustering factor parameter set to obtain the scene factor parameter distribution data set; step S213b, performing bird infestation impact analysis on the scene clustering factor parameter set respectively to determine the scene factor parameter impact coefficient set; step S213c, determining the scene parameter association data depth according to the scene factor parameter impact coefficient set; step S213d, performing association mining and expansion on the scene factor parameter distribution data set based on the scene parameter association data depth to obtain the bird infestation scene parameter distribution data set.

[0033] Preferably, according to the scene clustering factor parameters, relevant data are extracted from the bird infestation distribution data set. For example, if the scene clustering parameter is "wetland, high humidity", bird activity data related to wetland areas and high humidity are extracted, and the scene factor parameter distribution data set is divided and generated, which is stored in units of scenes, and all bird infestation distribution data related to specific scenes are marked. The degree of bird infestation impact is analyzed for each scene clustering factor parameter set, that is, a regression analysis model (such as linear regression or multivariate regression) is used to determine the specific impact of scene factors on bird infestation distribution, and an impact coefficient is quantified for each scene factor, thereby obtaining a set of scene factor parameter impact coefficients, such as the impact coefficient of wetland area, 0.85 (high correlation), and the impact coefficient of humidity, 0.70 (medium correlation). According to the scene factor parameter impact coefficient, determine which factors need in-depth correlation mining. The greater the impact of each factor parameter on bird infestation, the greater the model training accuracy requirement, and the corresponding training data requirement. The larger the value is, the deeper the scene parameter association data is, indicating the association strength and mining depth between scene factors or between scene factors and bird infestation distribution data. For example, if the impact coefficient of wetland area is high (such as 0.85), it is necessary to deeply mine its relationship with bird activity intensity in the data set, and use weighted association analysis methods (such as Apriori algorithm or FP-Growth) to calculate the association depth between scene parameters. Then, according to the scene parameter association data depth, the scene factor parameter distribution data set is mined to identify and expand hidden association information. Specifically, frequent association rules between scene parameters are identified, and single-dimensional scene parameters are expanded into multi-dimensional data relationships. Machine learning algorithms (such as decision trees or random forests) are used to generate new predictive variables or expand the dimensions of the data set to obtain the bird infestation scene parameter distribution data set. The expanded data set contains more implicit relationships and association information, providing a more comprehensive basis for bird activity prediction and governance strategies.

[0034] Step S300: If the bird infestation trend prediction parameter exceeds a preset bird infestation risk threshold, a bird infestation emergency response mechanism is triggered, and a bird infestation ecological management strategy library is obtained according to the bird infestation emergency response mechanism.

[0035] Preferably, if the bird infestation risk exceeds a preset warning level (preset bird infestation risk threshold), a pre-designed automated emergency response mechanism is activated, and based on the bird infestation emergency response mechanism, an ecological governance strategy library is formed to generate a governance plan suitable for the current situation. Specifically, the bird infestation risk threshold is a standard value used to measure whether bird activities pose a threat to the target scene. It is usually set based on scene characteristics, governance needs, and historical data statistics, based on parameters such as activity density (the number of individual birds exceeds a certain range), concentration of birds in key locations (such as airport runways, near power towers), high-risk activity periods (such as peak bird migration periods), and trend change rate (the rate of change in the intensity of bird infestation activities exceeds safety limits).

[0036] Preferably, when the prediction parameters exceed the preset threshold, the emergency response mechanism is activated, including generating a bird infestation warning signal, and generating a bird infestation ecological management strategy library according to the scenario and the current bird infestation situation, which includes specific management strategies for various scenarios. The core function of the strategy library is to provide the emergency response mechanism with highly targeted and eco-friendly solutions, which may include simple expulsion (such as playing sounds, starting flashing devices), deploying ecological management measures (such as setting up protective nets or spike devices, drone patrols or odor expulsion), arranging capture nets or blocking key areas, planting or clearing specific vegetation to change the bird foraging environment, setting up induced food or habitats to guide birds to migrate to low-risk areas, etc. Strategies can be selected according to the bird species, for example, short-term measures can be used for migratory birds, and long-term measures can be used for resident birds; the management intensity can be adjusted according to the risk level, such as harmless measures for low risks and multi-level combination measures for high risks.

[0037] Step S400: performing directional strategy analysis on the bird infestation trend prediction parameters based on the bird infestation ecological management strategy library to determine target bird infestation management strategy parameters.

[0038] Preferably, based on the predicted bird infestation trend data, the control measures suitable for the current scenario are screened, combined and analyzed from the bird infestation ecological control strategy library to generate specific executable control parameters to achieve accurate and targeted bird infestation control, wherein the bird infestation trend prediction parameters are the key input for control strategy analysis, including activity intensity, spatial distribution, risk level, time trend, species characteristics, etc. Specifically, targeted strategy analysis is an intelligent process that combines prediction parameters with the strategy library, and selects applicable strategies according to the type of bird infestation, such as adopting short-term expulsion plans for migratory birds and long-term control measures for resident birds; strategies are screened according to the bird infestation risk level, with ecological regulation for low risks and long-term control measures for medium risks. Joint governance, high-risk emergency response; lock the target governance area according to spatial distribution; then consider the impact of environmental conditions (such as weather, time) on the applicability of the strategy, and combine multiple governance measures according to multiple trend parameters, such as using lasers to drive away birds in high-risk hotspots and laying out protective nets; adjust the vegetation layout near the migration route and deploy sound-driven devices; parse out specific control parameters from the selected strategy, such as the start time, frequency, and power of the driving device, the patrol path, time interval, and speed of the drone, the layout area and coverage of the protective net, etc., as the final result of the analysis, that is, the target bird control strategy parameters, to ensure the operability and efficiency of the bird control strategy. Through targeted strategy analysis, the prediction parameters and the strategy library are closely combined to ensure that the governance plan is highly targeted and efficient, while avoiding waste of resources and damage to the ecological environment.

[0039] Furthermore, step S400 also includes step S410, performing directional strategy matching on the bird infestation trend prediction parameters based on the bird infestation ecological management strategy library to obtain a target bird infestation management directional strategy; step S420, constructing a bird infestation management strategy parameter space according to the target bird infestation management directional strategy; step S430, using the bird infestation trend prediction parameters as constraint parameters, performing strategy analysis in the bird infestation management strategy parameter space, and generating a management strategy parameter selection threshold; step S440, constructing a bird infestation management effect fitness function, using the bird infestation management effect fitness function to perform global optimization within the management strategy parameter selection threshold, and determining the target bird infestation management strategy parameters.

[0040] Preferably, directional strategy matching of bird infestation trend prediction parameters based on the bird infestation ecological management strategy library refers to screening management strategies that match the target scenario and trend in the strategy library according to the prediction parameters, and generating a target bird infestation management directional strategy. For example, if the predicted high-risk area is an airport runway, select sound and light repellent and drone patrol strategies; if the predicted activity hotspot is an orchard, select ecological regulation and physical barrier strategies; then, based on the target bird infestation management directional strategy, extract all parameter dimensions and value ranges of the target directional strategy, and construct a bird infestation management strategy parameter space, which includes all parameters of the target management strategy and their possible value ranges, such as the power, frequency, and startup time of the repelling equipment, the path, speed, and time interval of the drone patrol, and the coverage, density, and installation location of the protective net, as a parameter search space for bird infestation management.

[0041] Preferably, bird infestation trend prediction parameters (such as bird activity intensity, distribution area, time trend, etc.) are used as constraints to limit the value range of control strategy parameters, and strategy analysis is performed within the bird infestation control strategy parameter space, that is, parameter combinations that meet the requirements are screened according to the constraints, and control strategy parameter selection thresholds are generated to narrow the search range of bird infestation control parameters and improve the optimization rate; finally, a bird infestation control effect fitness function is constructed to quantify the advantages and disadvantages of different strategy parameter combinations and evaluate the control strategy effects, and then the bird infestation control effect fitness function is used to perform global optimization within the control strategy parameter selection threshold, that is, an optimization algorithm (such as genetic algorithm, particle swarm optimization) is used to find the global optimal solution of the fitness function within the selected threshold range, and then the target bird infestation control strategy parameters (specific implementation plan of the control measures) are determined to ensure that the control plan achieves the best balance between effect, cost and eco-friendliness.

[0042] Furthermore, step S440 also includes step S441, randomly selecting multiple governance strategy parameters within the governance strategy parameter selection threshold, using the bird control effect fitness function to evaluate the fitness of the multiple governance strategy parameters, and obtaining multiple strategy parameter fitness; step S442, initializing particle swarm parameters, the particle swarm parameters including particle position and particle velocity; step S443, performing gradient ascent calculation on the fitness of the multiple strategy parameters to obtain multi-parameter gradient ascent calculation results, and determining the strategy parameter optimization direction according to the multi-parameter gradient ascent calculation results; step S444, evaluating and iteratively updating the particle swarm parameters based on the strategy parameter optimization direction until the preset termination condition is met, and optimizing and determining the particle with the largest fitness as the target bird control strategy parameter.

[0043] Preferably, the particle swarm optimization algorithm (PSO) is combined with the gradient ascent method to determine the optimal bird control strategy parameters. Specifically, multiple control strategy parameters are randomly selected within the control strategy parameter selection threshold as the initial solution set, and the bird control effect fitness function is used to perform fitness evaluation, that is, the fitness value is calculated for each randomly generated parameter combination to obtain the fitness of multiple strategy parameters, and then the particle swarm parameters are initialized, and each particle corresponds to a control strategy parameter combination, wherein the particle position represents the current solution of the particle, that is, the specific value of the parameter combination, and the particle speed represents the moving direction and step size of the particle, and controls the update of the parameter value; the gradient of each parameter is calculated for the fitness of multiple strategy parameters to determine the fitness The direction of improvement, for example, if power increase will improve fitness, the gradient direction is power increase; if frequency decrease will improve fitness, the gradient direction is frequency decrease; the gradient ascent calculation result refers to the gradient direction and size of each particle, and then the gradient calculation results of the particles are combined, and the largest gradient ascent calculation result is used as the optimization direction, and the main direction of global optimization is determined to ensure that the particle swarm moves to areas with higher fitness; finally, based on the strategy parameter optimization direction, the particle swarm parameters are evaluated and iteratively updated until the preset termination conditions are met (for example, the number of iterations reaches the set value or the fitness value meets the preset threshold), and the particle with the highest fitness in the particle swarm is used as the target bird control strategy parameter to ensure that the control effect is maximized.

[0044] Step S500: performing directional bird control and effect evaluation on the target scene area according to the target bird control strategy parameters to obtain a bird control effect coefficient.

[0045] Preferably, according to the analyzed control strategy parameters, specific bird control measures are implemented in the target scene area, and the control effect is quantitatively evaluated to measure the actual effectiveness of the control. The result of the effect evaluation is expressed by the bird control effect coefficient. Specifically, targeted bird control refers to the implementation of control measures in the target scene in a targeted manner according to the analyzed target control strategy parameters. The control process includes starting relevant equipment or facilities according to the control parameters, such as playing sound waves to drive away birds or simulating the sounds of natural enemies; using lasers and flash lights for visual repelling; patrolling along designated paths and frequencies to drive away birds; installing protective nets or physical isolation devices; removing vegetation that attracts birds, or planting crops that can limit bird activities; and arranging the execution duration and frequency of control measures according to the time characteristics of bird activities (such as peak hours), such as not starting visual devices at night and increasing the frequency of drone patrols during peak hours during the day; for complex bird infestation scenes, a combination of multiple measures is used for comprehensive control, such as using sound and light repellent and protective net layout at the same time to improve the control effect.

[0046] Preferably, after the implementation of the control measures, the changes in bird infestation in the target area are monitored and analyzed in real time to evaluate the effectiveness of the control. Specifically, through the bird monitoring and sensing network, the changes in the number of birds before and after the control, the changes in the frequency or distribution heat map of bird activities, the transfer or disappearance of bird hotspots, and whether the peak period of bird activities after the control is shortened or shifted are collected in real time. Then the collected data is compared with the baseline data before the control to quantify the control effect. For example, based on the bird population reduction rate, activity density change value, control success area coverage rate and target achievement rate, a weighted calculation is performed to generate a bird infestation control effect coefficient as a quantitative indicator of the comprehensive control effect. Through the closed-loop mechanism of control and evaluation, the control efficiency and accuracy can be continuously improved, and the continuous optimization and long-term management of bird infestation control can be achieved.

[0047] Step S600, based on the bird control effect coefficient, the target bird control strategy parameters are tuned and updated to obtain the bird control strategy optimization parameters, and the bird control is controlled by using the bird control strategy optimization parameters.

[0048] Preferably, the bird control effect coefficient in the bird control effect evaluation result is used to dynamically adjust and optimize the parameters of the currently implemented control strategy to make it more in line with the actual scenario needs, improve the efficiency and effectiveness of control, and implement control again through the optimized parameters to achieve a closed-loop improvement of control effect. Among them, if the bird control effect coefficient is high, it indicates that the current strategy is relatively effective and may only require minor optimization. If the bird control effect coefficient is low, it indicates that the current strategy is insufficient and needs to be adjusted or replaced to a larger extent. Tuning and updating refers to analyzing the performance of each indicator in the effect coefficient, finding out the key factors with a large influencing coefficient (such as insufficient coverage of a certain area or insufficient control intensity in a certain time period), combining the target scenario and the characteristics of bird activities, determining the optimization goal, such as increasing the control intensity of a high-risk area or shortening the response time, and then using Machine learning algorithms (such as reinforcement learning) dynamically adjust governance strategies to cope with complex changes in different scenarios. Combining analysis results and optimization models, new governance strategy parameters are generated, called bird control strategy optimization parameters, which may include improved equipment operating parameters (such as power, startup time), optimized governance area range, and new combinations of governance measures (such as adding or reducing certain governance measures). Finally, bird control is carried out through the bird control strategy optimization parameters, that is, the location and operation mode of the equipment are adjusted according to the optimization parameters, the governance measures are executed according to the optimized schedule and frequency, and the governance strategy is adjusted in time according to the real-time monitoring data and optimization parameters. After governance is carried out through the optimized parameters, the effect is evaluated again to form a closed-loop control mechanism of governance-evaluation-optimization, so as to achieve continuous improvement in bird control and provide efficient and reliable governance solutions for different scenarios.

[0049] Furthermore, step S600 also includes step S610, performing fine-tuning variation analysis on the target bird control strategy parameters based on the bird control effect coefficient to obtain parameter variation optimization rules; step S620, performing variation optimization and expansion on the target bird control strategy parameters according to the parameter variation optimization rules to obtain a bird control strategy variation parameter set; step S630, comparing and optimizing the bird control strategy variation parameter set based on the bird control effect fitness function to obtain the bird control strategy optimization parameters.

[0050] Preferably, fine-tuning variation analysis of target bird control strategy parameters based on the bird control effect coefficient refers to making small adjustments in the vicinity of the target parameters to analyze the impact of the adjustments on the bird control effect coefficient, that is, randomly changing the values ​​of certain parameters to observe their sensitivity to the bird control effect coefficient. For example, fine-tuning variation is performed on the power of 70%, and an attempt is made to increase or decrease it by 5% (i.e., 65% to 75%). According to the results of variation analysis, parameter variation optimization rules are summarized. For example, the power is more effective in the range of 65% to 72%. Through fine-tuning and variation analysis, a potential better parameter range is discovered. The target bird control strategy is optimized according to the parameter variation optimization rules. The parameters are mutated and optimized and expanded, that is, according to the mutation optimization rules, new parameter combinations are generated on the basis of the target parameters to obtain the variation parameter set of the bird control strategy, which includes all the parameter combinations of variation optimization and expansion, forming a new parameter candidate set, increasing the potential optimal solution and improving the comprehensiveness of the optimization; the bird control effect fitness function is then used to evaluate the parameters of the bird control strategy variation parameter set, calculate the fitness and compare the fitness values ​​of all variation parameters, and select the parameters with the highest fitness as the final optimization parameters, that is, the bird control strategy optimization parameters, to ensure the best balance between control effect and resource utilization, so that the bird control effect is more accurate and reliable.

[0051] In the above, refer to Figure 1 The method for controlling bird infestation under scene adaptation according to an embodiment of the present invention is described in detail. Figure 2 A bird infestation directional control and management platform under scenario adaptation according to an embodiment of the present invention is described.

[0052] The bird infestation directional control and management platform under scenario adaptation according to the embodiment of the present invention is used to solve the technical problems existing in the prior art of lack of real-time monitoring and perception of bird activities in the target area, inability to make trend predictions in advance to deal with bird infestation outbreaks, resulting in poor pertinence of bird infestation control and lagging control results, thereby achieving the technical effect of improving the scientificity, timeliness and stability of bird infestation control. The bird infestation directional control and management platform under scenario adaptation includes: a target bird infestation distribution data set acquisition module 10, a bird infestation trend prediction parameter output module 20, a bird infestation ecological control strategy library acquisition module 30, a target bird infestation control strategy parameter determination module 40, a bird infestation control effect coefficient acquisition module 50, and a bird infestation control strategy optimization parameter acquisition module 60.

[0053] The target bird infestation distribution data set acquisition module 10 is used to deploy a bird monitoring perception network in the target scene area, and collect and acquire the target bird infestation distribution data set through the bird monitoring perception network; the bird infestation trend prediction parameter output module 20 is used to call the bird activity adaptive prediction network set to perform trend prediction on the target bird infestation distribution data set and output the bird infestation trend prediction parameter; the bird infestation ecological management strategy library acquisition module 30 is used to trigger the bird infestation emergency response mechanism if the bird infestation trend prediction parameter exceeds the preset bird infestation risk threshold, and obtain the bird infestation ecological management strategy library according to the bird infestation emergency response mechanism; the target bird infestation management strategy parameter A number determination module 40 is used to perform directional strategy analysis on the bird infestation trend prediction parameters based on the bird infestation ecological management strategy library to determine the target bird infestation management strategy parameters; a bird infestation management effect coefficient acquisition module 50 is used to perform directional bird infestation management and effect evaluation on the target scene area according to the target bird infestation management strategy parameters to obtain the bird infestation management effect coefficient; a bird infestation management strategy optimization parameter acquisition module 60 is used to tune and update the target bird infestation management strategy parameters based on the bird infestation management effect coefficient to obtain the bird infestation management strategy optimization parameters, and perform bird infestation management and control through the bird infestation management strategy optimization parameters.

[0054] The specific configuration of the bird infestation trend prediction parameter output module 20 will be described in detail below. The bird infestation trend prediction parameter output module 20 further includes: using big data technology to mine and obtain a bird infestation distribution data set, classifying and training the bird infestation distribution data set based on the bird infestation scene factor information, and establishing a bird activity prediction network library; classifying and marking the target scene area according to the bird infestation scene factor information, and determining the target scene calibration parameter combination; matching and calling the bird activity prediction network library based on the target scene calibration parameter combination to obtain a bird activity adaptive prediction network set; performing trend prediction on the target bird infestation distribution data set based on the bird activity adaptive prediction network set, and outputting the bird infestation trend prediction parameters.

[0055] The specific configuration of the bird infestation trend prediction parameter output module 20 will be described in detail below. The bird infestation trend prediction parameter output module 20 further includes: clustering and marking the bird infestation scene factor information to determine the bird infestation scene clustering factors, the bird infestation scene clustering factors including terrain environment, climate characteristics, bird species and activity patterns; extracting and integrating the parameters of the bird infestation distribution data set based on the bird infestation scene clustering factors to obtain a scene clustering factor parameter set; performing association division on the bird infestation distribution data set according to the scene clustering factor parameter set to obtain a bird infestation scene parameter distribution data set; performing prediction training on the bird infestation scene parameter distribution data set respectively to obtain a bird activity prediction network set, and based on the bird activity prediction network set, identifying and establishing the bird activity prediction network library.

[0056] The specific configuration of the bird infestation trend prediction parameter output module 20 will be described in detail below. The bird infestation trend prediction parameter output module 20 further includes: extracting and dividing the bird infestation distribution data set according to the scene clustering factor parameter set to obtain a scene factor parameter distribution data set; performing bird infestation impact degree analysis on the scene clustering factor parameter set to determine a scene factor parameter impact coefficient set; determining the scene parameter association data depth according to the scene factor parameter impact coefficient set; performing association mining and expansion on the scene factor parameter distribution data set based on the scene parameter association data depth to obtain the bird infestation scene parameter distribution data set.

[0057] The specific configuration of the target bird control strategy parameter determination module 40 will be described in detail below. The target bird control strategy parameter determination module 40 further includes: performing directional strategy matching on the bird control trend prediction parameters based on the bird control ecological control strategy library to obtain the target bird control directional strategy; constructing a bird control strategy parameter space according to the target bird control directional strategy; using the bird control trend prediction parameters as constraint parameters, performing strategy analysis in the bird control strategy parameter space, and generating a control strategy parameter selection threshold; constructing a bird control effect fitness function, and using the bird control effect fitness function to perform global optimization within the control strategy parameter selection threshold to determine the target bird control strategy parameters.

[0058] The specific configuration of the target bird control strategy parameter determination module 40 will be described in detail below. The target bird control strategy parameter determination module 40 further includes: randomly selecting multiple control strategy parameters within the control strategy parameter selection threshold, using the bird control effect fitness function to evaluate the fitness of the multiple control strategy parameters to obtain multiple strategy parameter fitness; initializing particle swarm parameters, the particle swarm parameters include particle position and particle velocity; performing gradient ascent calculation on the fitness of the multiple strategy parameters to obtain multi-parameter gradient ascent calculation results, and determining the strategy parameter optimization direction based on the multi-parameter gradient ascent calculation results; evaluating and iteratively updating the particle swarm parameters based on the strategy parameter optimization direction until the preset termination condition is met, and optimizing and determining the particle with the largest fitness as the target bird control strategy parameter.

[0059] The specific configuration of the bird control strategy optimization parameter acquisition module 60 will be described in detail below. The bird control strategy optimization parameter acquisition module 60 may further include: performing fine-tuning variation analysis on the target bird control strategy parameters based on the bird control effect coefficient to obtain parameter variation optimization rules; performing variation optimization and expansion on the target bird control strategy parameters according to the parameter variation optimization rules to obtain a bird control strategy variation parameter set; performing comparison and optimization on the bird control strategy variation parameter set based on the bird control effect fitness function to obtain the bird control strategy optimization parameters.

[0060] The bird infestation directional control and management platform under scenario adaptation provided by the embodiments of the present invention can execute the bird infestation directional control and management method under scenario adaptation provided by any embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method.

[0061] Although the present application makes various references to certain modules in the platform according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be implemented; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0062] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A method for directional bird control under scene adaptation, characterized in that: The method comprises: Deploy a bird monitoring and sensing network in the target scene area, and collect and obtain a target bird infestation distribution data set through the bird monitoring and sensing network; Calling the bird activity adaptive prediction network set to perform trend prediction on the target bird infestation distribution data set, and outputting bird infestation trend prediction parameters; If the bird infestation trend prediction parameter exceeds the preset bird infestation risk threshold, the bird infestation emergency response mechanism is triggered, and a bird infestation ecological management strategy library is obtained according to the bird infestation emergency response mechanism; Based on the bird infestation ecological management strategy library, a directional strategy analysis is performed on the bird infestation trend prediction parameters to determine the target bird infestation management strategy parameters; According to the target bird infestation control strategy parameters, the target scene area is subjected to directional bird infestation control and effect evaluation to obtain a bird infestation control effect coefficient; The target bird control strategy parameters are tuned and updated based on the bird control effect coefficient to obtain the bird control strategy optimization parameters, and the bird control is managed and controlled using the bird control strategy optimization parameters.

2. The method for directional bird control under scene adaptation as claimed in claim 1, characterized in that: The output bird infestation trend prediction parameters include: Using big data technology to mine and obtain a bird infestation distribution data set, classify and train the bird infestation distribution data set based on bird infestation scene factor information, and establish a bird activity prediction network library; Classify and mark the target scene area according to the bird-infested scene factor information, and determine a target scene calibration parameter combination; Based on the target scene calibration parameter combination, the bird activity prediction network library is matched and called to obtain a bird activity adaptive prediction network set; Based on the bird activity adaptive prediction network set, trend prediction is performed on the target bird infestation distribution data set, and the bird infestation trend prediction parameters are output.

3. The method for directional bird control under scene adaptation as claimed in claim 2, characterized in that: The method of establishing a bird activity prediction network database comprises: Clustering and marking the bird infestation scene factor information to determine the bird infestation scene clustering factors, wherein the bird infestation scene clustering factors include terrain environment, climate characteristics, bird species and activity patterns; Extracting and integrating parameters of the bird infestation distribution data set based on the bird infestation scene clustering factors to obtain a scene clustering factor parameter set; According to the scene clustering factor parameter set, the bird infestation distribution data set is associated and divided to obtain a bird infestation scene parameter distribution data set; Prediction training is performed on the bird-infestation scene parameter distribution data sets respectively to obtain a bird activity prediction network set, and the bird activity prediction network library is identified and established based on the bird activity prediction network set.

4. The method for directional bird control under scene adaptation as claimed in claim 3 is characterized in that: The step of obtaining a bird infestation scene parameter distribution data set includes: Extracting and dividing the bird infestation distribution data set according to the scene clustering factor parameter set to obtain a scene factor parameter distribution data set; Analyzing the degree of bird infestation impact on the scene clustering factor parameter sets respectively to determine a scene factor parameter impact coefficient set; Determining the scene parameter association data depth according to the scene factor parameter influence coefficient set; Based on the scene parameter association data depth, the scene factor parameter distribution data set is associated with mining and expansion to obtain the bird infestation scene parameter distribution data set.

5. The method for directional bird control under scene adaptation as claimed in claim 1, characterized in that: The determination of target bird pest control strategy parameters includes: Based on the bird infestation ecological management strategy library, the bird infestation trend prediction parameters are matched with a directional strategy to obtain a target bird infestation management directional strategy; According to the target bird control directional strategy, construct a bird control strategy parameter space; Taking the bird infestation trend prediction parameter as a constraint parameter, performing strategy analysis in the bird infestation control strategy parameter space, and generating a control strategy parameter selection threshold; A bird control effect fitness function is constructed, and the bird control effect fitness function is used to perform global optimization within the control strategy parameter selection threshold to determine the target bird control strategy parameters.

6. The method for directional bird control under scene adaptation as claimed in claim 5, characterized in that: The step of determining the target bird pest control strategy parameters includes: Randomly select multiple control strategy parameters within the control strategy parameter selection threshold, and use the bird control effect fitness function to evaluate the fitness of the multiple control strategy parameters to obtain the fitness of multiple strategy parameters; Initializing particle swarm parameters, wherein the particle swarm parameters include particle positions and particle velocities; Performing gradient ascent calculation on the fitness of the multiple strategy parameters to obtain a multi-parameter gradient ascent calculation result, and determining a strategy parameter optimization direction according to the multi-parameter gradient ascent calculation result; Based on the optimization direction of the strategy parameters, the particle swarm parameters are evaluated and iteratively updated until a preset termination condition is met, and the particle with the largest fitness is determined by optimization as the target bird control strategy parameter.

7. The method for directional bird control under scene adaptation as claimed in claim 5, characterized in that: The obtained bird control strategy optimization parameters include: Based on the bird control effect coefficient, fine-tuning variation analysis is performed on the target bird control strategy parameters to obtain parameter variation optimization rules; According to the parameter variation optimization rule, the target bird control strategy parameters are mutated, optimized and expanded to obtain a bird control strategy variation parameter set; Based on the bird control effect fitness function, the bird control strategy variation parameter set is compared and optimized to obtain the bird control strategy optimization parameters.

8. The bird-infestation directional control and management platform under scene adaptation is characterized by: The platform is used to implement the method for directional bird control under scenario adaptation according to any one of claims 1 to 7, and the platform includes: A target bird infestation distribution data set acquisition module is used to deploy a bird monitoring and sensing network in a target scene area, and acquire a target bird infestation distribution data set through the bird monitoring and sensing network; A bird infestation trend prediction parameter output module is used to call a bird activity adaptive prediction network set to perform trend prediction on the target bird infestation distribution data set and output bird infestation trend prediction parameters; A bird infestation ecological management strategy library acquisition module is used to trigger a bird infestation emergency response mechanism if the bird infestation trend prediction parameter exceeds a preset bird infestation risk threshold, and obtain a bird infestation ecological management strategy library according to the bird infestation emergency response mechanism; A target bird infestation control strategy parameter determination module is used to perform a directional strategy analysis on the bird infestation trend prediction parameters based on the bird infestation ecological control strategy library to determine the target bird infestation control strategy parameters; A bird infestation control effect coefficient obtaining module is used to perform directional bird infestation control and effect evaluation on the target scene area according to the target bird infestation control strategy parameters to obtain the bird infestation control effect coefficient; The bird control strategy optimization parameter acquisition module is used to tune and update the target bird control strategy parameters based on the bird control effect coefficient, obtain the bird control strategy optimization parameters, and perform bird control management through the bird control strategy optimization parameters.

Images