Satellite and unmanned aerial vehicle remote sensing monitoring prevention and control system based on insect behavior regulation and control

Abstract

The invention discloses a satellite and unmanned aerial vehicle remote sensing monitoring control system based on insect behavior regulation and control, which relates to the technical field of remote sensing monitoring and comprises an insect monitoring regulation and control system, an unmanned aerial vehicle remote sensing comprehensive processing system, a satellite remote sensing comprehensive processing system and a visual monitoring command platform. The insect monitoring regulation and control system carries out insect behavior regulation and control on a target area and collects an environment monitoring data set, and ground insect information feature data is obtained through analysis; the unmanned aerial vehicle remote sensing comprehensive processing system collects unmanned aerial vehicle multispectral images of the region after insect behavior regulation and control, and analyzes the images to obtain insect generation types, insect distribution thermodynamic diagrams and insect density change trends; a satellite remote sensing comprehensive processing system collects satellite remote sensing data of the region after insect behavior regulation and control, and a wide-area insect pest thermodynamic diagram is obtained through analysis; and the visual monitoring command platform is used for providing a visual operation interface for a user, presenting an insect pest analysis result to the user, and supporting the user to carry out remote operation and implement a real-time insect pest control scheme.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing monitoring, and in particular to a satellite and UAV remote sensing monitoring and control system based on insect behavior regulation. Background Art

[0002] In recent years, the application of insect behavioral manipulation in agricultural pest control has become increasingly widespread. Biomimetic technologies, using chemical pheromones and physical signals, can disrupt insects' natural behavioral patterns, reducing pesticide use and achieving pest control. At the same time, when pest infestations are severe and behavioral manipulation is ineffective, active pesticide spraying is still necessary. Therefore, monitoring the effectiveness of behavioral manipulation is also a crucial step. Rapid advances in communications, circuit chip, Internet of Things, and computer vision technologies have reduced the cost of hardware such as multispectral imaging and thermal infrared sensors, while improving the efficiency of software-side data processing algorithms. Remote sensing technology is gradually moving from specialized fields to large-scale applications. UAV remote sensing offers high-resolution bird's-eye views of crop areas, enabling assessment of pest control effectiveness. Satellite remote sensing, with its wide-area coverage, can analyze vegetation health over larger areas and indirectly assess macro-infestation conditions, providing a comprehensive reference for regional pest control. Therefore, integrating insect behavioral manipulation with drone and satellite remote sensing technologies to build a coordinated pest monitoring and control system for efficient early warning and precise control is a promising research direction.

[0003] At present, the Chinese invention patent application with application number CN202410201845.0 discloses a long-distance trajectory simulation method for insect migration takeoff and cruise. The application includes: obtaining background atmospheric field data of the insect migration takeoff area through remote sensing technology, GIS technology and other means, setting reasonable conditions based on the acquired data to determine the insect takeoff area. These conditions include the species, number, gender, and migration purpose of the insect, and using the Lagrangian particle diffusion model to simulate the insect migration trajectory in the background atmospheric field. This method determines the takeoff area of ​​insect migration based on the background atmospheric field and incorporates more accurate insect flight characteristics into the cruise trajectory simulation, so that it can more accurately simulate the insect migration trajectory and predict possible migration areas. It comprehensively considers the influence of individual insect behavior and environmental factors on the migration trajectory, so as to more comprehensively understand the behavior and laws of insect migration. However, this application focuses on the migration trajectory of insects, and its premise is that a large-scale insect swarm has been formed. This application cannot identify insect pests in the early stages of generation and achieve early intervention and control. Summary of the Invention

[0004] The technical problem solved by the present invention is that in the early stages of insect pest development, small-scale insect swarms are active under the cover of crop leaves, and it is difficult for drones / satellites to capture insect images from overhead. Existing technologies usually avoid this problem and rely on passive monitoring by photographing the insect's living environment, such as vegetation. There is a lack of combining insect behavioral regulation with satellite / remote sensing images to actively monitor the insects themselves directly after interfering with their behavior. Therefore, existing technologies are mostly suitable for inspections after insect pests have formed a scale, and lack the ability to conduct efficient detection and intervention control in the early stages of pest generation. In addition, existing technologies usually use satellite remote sensing alone or drone remote sensing alone for pest monitoring, lack the linkage and integration of satellite and drone remote sensing, and have a single monitoring dimension.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] Satellite and UAV remote sensing monitoring and control system based on insect behavior regulation, including insect monitoring and control system, UAV remote sensing integrated processing system, satellite remote sensing integrated processing system and visual monitoring and command platform;

[0007] The insect monitoring and control system is used to collect environmental monitoring data sets, analyze ground insect information feature data, and perform insect behavioral control;

[0008] The UAV remote sensing integrated processing system is used to collect UAV multispectral images of the target area after insect behavior control, and analyze them to obtain insect distribution heat maps and insect density change trends;

[0009] The satellite remote sensing integrated processing system is used to collect and analyze satellite remote sensing data to obtain wide-area pest heat maps;

[0010] The visual monitoring and command platform is used to provide users with a visual operation interface, presenting insect distribution heat maps, insect density change trends and wide-area pest heat maps to users, supporting users to remotely operate the system and implement real-time pest control plans.

[0011] Preferably, the insect monitoring and control system is used to collect environmental monitoring data sets, analyze the environmental monitoring data sets to obtain ground insect information feature data and perform insect behavior control;

[0012] The insect behavior regulation includes non-single insect behavior regulation and single insect behavior regulation;

[0013] The insect monitoring and control system includes a cloud-based control and analysis module and a monitoring and control execution module;

[0014] The monitoring and control execution module includes a power supply device, a data remote transmission device, a camera, a temperature and humidity sensor, a soil sensor, a wind direction sensor, a light source device, a pheromone propagation device and an insect sound / vibration signal collection and processing device;

[0015] The insect sound / vibration signal acquisition and processing device comprises a data acquisition card, a microphone probe, a microphone probe, a sound playing device and a vibration exciter;

[0016] When the pest occurrence category in the monitoring area is not single and the monitoring pest target is not directional, the insect monitoring and control system collects real-time environmental data to obtain an environmental monitoring data set, processes the environmental monitoring data set to obtain ground insect information feature data, and performs non-single insect behavioral control based on the ground insect information feature data and the first rule knowledge graph;

[0017] When the type of insect pests in the monitoring area is single and the monitored pest targets are directed, the insect monitoring and control system performs behavioral control of a single insect based on the directed pest target type and the second rule knowledge graph.

[0018] Preferably, the ground insect information characteristic data includes insect occurrence species, insect distribution density at monitoring points and insect age data;

[0019] The processing logic for obtaining ground insect information feature data includes:

[0020] Collect data through sensor arrays to obtain environmental monitoring data sets, and upload the environmental monitoring data sets to the cloud;

[0021] The environmental monitoring data set includes sound / vibration signals, image data, temperature and humidity data, soil data and wind direction data;

[0022] The sensor array includes a microphone probe, a microphone probe, a camera, a temperature and humidity sensor, a soil sensor and a wind direction sensor;

[0023] The cloud-based control and analysis module analyzes the sound / vibration signals in the environmental monitoring dataset to determine the insect species. The module performs noise reduction and reverberation on the sound / vibration signals to obtain vibration time-domain data. Fast Fourier transform (FFT) is then used to obtain vibration spectrum data. Mel-frequency cepstral coefficients, spectrograms, zero-crossing rates, energy, and fundamental frequencies are then calculated based on the vibration time-domain and spectrum data to obtain a time-frequency domain feature set. The pre-trained LSTM recurrent neural network is then used to match the time-frequency domain feature set to determine the insect species.

[0024] The insect occurrence types include insect types and probability values ​​corresponding to each insect type;

[0025] The CNN convolutional neural network model is used to perform target detection on image data in the environmental monitoring dataset to obtain an insect frame selection map. The Deep SORT target tracking algorithm is used to process the insect frame selection maps of consecutive frames to obtain the tracked insect trajectories. The spatial kernel density estimation algorithm is used to calculate the superimposed probability distribution of the tracked insect trajectories to obtain a pixel-level insect swarm density map. The pixel-level insect swarm density map is then integrated to obtain the insect distribution density at the monitoring point.

[0026] Based on the insect occurrence species, insect population estimation, temperature and humidity data and preset insect biological characteristic data, the insect age data is obtained by judgment through the accumulated temperature model. The insect biological characteristic data include the development starting temperature, upper limit temperature, effective accumulated temperature, development threshold, generation cycle information, diapause characteristics, humidity sensitivity, reproduction rate and survival rate of each insect species.

[0027] Preferably, the processing logic for performing non-single insect behavioral regulation includes:

[0028] The cloud-based control analysis module uses the first-rule knowledge graph to match ground insect information feature data with expected behaviors to generate the optimal control method;

[0029] The desired behaviors include aggregation, avoidance, migration, immobility and killing;

[0030] The optimal control method includes control behavior and control parameters;

[0031] The regulation behavior includes outputting sound / vibration signals, optical signals, chemical odor pheromones, and sexual attractant pheromones;

[0032] The control parameters include the frequency of the sound / vibration signal, the wavelength of the optical signal, the release concentration and rate of the chemical odor pheromone, and the release concentration and rate of the sex attractant pheromone.

[0033] Preferably, the processing logic for performing single insect behavioral regulation includes:

[0034] The cloud-based control analysis module uses the second rule knowledge graph to match the targeted pest target types and expected behaviors to generate inherent control methods. The inherent control methods include targeted control behaviors and targeted control parameters.

[0035] The directional control behavior includes outputting sound / vibration signals, optical signals, chemical odor pheromones, and sexual attractant pheromones;

[0036] The directional control parameters include the frequency of the sound / vibration signal, the wavelength of the optical signal, the release concentration and rate of the chemical odor pheromone, and the release concentration and rate of the sex attractant pheromone.

[0037] Preferably, the UAV remote sensing integrated processing system is used to collect and obtain multispectral images from the UAV, analyze the multispectral images from the UAV to obtain a thermodynamic map of insect distribution, and compare the insect density observed by the UAV with the insect distribution density at the monitoring point output by the insect monitoring and control system to generate an insect density change trend;

[0038] The pest distribution includes the pest occurrence degree and occurrence probability value corresponding to each pixel point. The pest occurrence degree includes no pest, mild pest and significant pest.

[0039] The UAV remote sensing integrated processing system includes a UAV remote sensing platform and a UAV image processing module;

[0040] The UAV remote sensing platform includes a payload UAV, a multispectral remote sensing sensor, a GNSS positioning device, a UAV measurement and control device, and a data remote transmission device;

[0041] The UAV image processing module includes a hardware computer and a hyperspectral image analysis program carried by it.

[0042] Preferably, the processing logic for analyzing the multispectral image of the UAV includes:

[0043] The UAV image acquisition platform is used to complete data collection. A multi-rotor UAV equipped with a hyperspectral sensor acquires images of the target monitoring area to obtain UAV multispectral images, and a GNSS positioning device records the current image acquisition position to obtain UAV GNSS positioning data.

[0044] The data transmission and communication module transmits the drone's multispectral image and drone's GNSS positioning data to a mobile ground data station, and after the drone returns, the local data is synchronously transmitted back to the drone's image processing module;

[0045] The hyperspectral image analysis program carried by the UAV image processing module performs radiation correction, geometric orthogonality and noise suppression on the synchronously transmitted UAV multispectral image to obtain the UAV pre-processed image;

[0046] Principal component analysis was used to screen the spectral channels of the drone pre-processed images to obtain key feature dimensions. The pre-trained support vector machine model was then used to perform inversion processing in combination with the key feature dimensions to obtain the pest distribution and pixel confidence. Based on the pixel confidence, noise suppression and uncertainty correction were performed on the pest distribution to obtain an insect distribution heat map.

[0047] Based on the insect distribution heat map, pixel value integral calculation is performed to obtain the insect density observed by the drone. The insect density observed by the drone is then compared with the insect distribution density at the monitoring point output by the insect monitoring and control system to generate the insect density change trend.

[0048] The pest distribution includes the pest occurrence degree and occurrence probability value corresponding to each pixel point. The pest occurrence degree includes no pest, mild pest and significant pest.

[0049] Preferably, the satellite remote sensing integrated processing system is used to collect satellite remote sensing data and analyze and generate a wide-area insect pest heat map;

[0050] The satellite remote sensing integrated processing system includes a satellite data interface module, a satellite data pre-processing module and a satellite image analysis module;

[0051] Obtain satellite remote sensing data of the monitoring area through the satellite data interface module;

[0052] The satellite remote sensing data includes optical remote sensing image data, atmospheric upwelling radiation data and meteorological data;

[0053] The spatial resolution of the optical remote sensing image data is not less than 5 meters, and the temporal resolution is not less than 3 days, including blue light, green light, red light, near-infrared light and short-wave infrared light bands.

[0054] Preferably, the satellite data preprocessing module performs radiation correction, geometric orthogonality, noise suppression, rain cloud removal and data interpolation processing on the optical remote sensing image data to obtain the satellite image data to be analyzed, and performs inversion calculation on the satellite image data to be analyzed to obtain the satellite remote sensing derivative data;

[0055] The satellite remote sensing derived data include land surface temperature, soil surface moisture, vegetation canopy temperature, surface leaf area index and primary productivity;

[0056] The satellite image analysis module generates remote sensing feature raster layers based on satellite remote sensing derived data, and performs ROI extraction on the optical remote sensing image data based on the UAV GNSS positioning data generated by the UAV remote sensing integrated processing system to obtain vegetation area images;

[0057] The DeeplabV3+ semantic segmentation network is used to analyze remote sensing feature raster layers, vegetation area images, and satellite remote sensing derived data to obtain vegetation impact results. Based on the vegetation impact results, a wide-area pest heat map is generated.

[0058] The vegetation impact results include the severity of the pests and the scope of impact.

[0059] Preferably, the visual monitoring and command platform includes a visual display module, a system control module and a command and dispatch module;

[0060] The visualization display module is used to integrate all output data from the insect monitoring and control system, satellite remote sensing integrated processing system, and UAV remote sensing integrated processing system. It presents insect distribution heat maps, insect density change trends, and wide-area insect pest heat maps on the client through interactive geographic information system (GIS) maps, trend charts, and real-time status panels, supporting users to query data and conduct historical review.

[0061] The system control module is used to support users to remotely operate the system, receive start commands, stop commands, parameter adjustment commands and mode switching commands issued by users to the system, and support users to accurately control software and hardware devices;

[0062] The command and dispatch module is used to receive real-time pest control plans and supports users to send operation instructions corresponding to the real-time pest control plans to the pest control drones for flight control operations. The pest control plans include the type and number of drones, the type of pesticide, the dosage, the application method and the application cycle.

[0063] The beneficial effects of this invention are as follows: The image data collected by drones and satellites in this application is not from a natural environment, but rather from a human-intervention environment, after insect behavior has been manipulated. In the early stages of an infestation, when small insect swarms operate beneath the cover of crop leaves, this application uses an insect monitoring and control system to induce insects to engage in specific behaviors such as aggregation and migration, overcoming the problem of crop obstruction often encountered by drone and satellite remote sensing. This allows for efficient monitoring of insect behavior, rather than relying solely on vegetation conditions to indirectly assess infestations.

[0064] This application first pre-deploys monitoring points and control equipment in target monitoring areas prone to insect infestations, completing advance preparations for protecting important economic crops. In the early stages of an infestation, ground environmental data is collected and a dual-pathway insect behavioral control plan is generated based on the type of insect pest. This facilitates more accurate and efficient control of insect behavior, thereby protecting economic crops.

[0065] This application also uses drone-based remote sensing to analyze insect density trends across entire crop areas, overcoming the limited range of ground-based monitoring points. Analyzing insect density trends overcomes the scale discrepancy between ground-based monitoring data and drone remote sensing data, and verifies the effectiveness of behavioral manipulation of insects, facilitating additional pest identification in crop areas beyond monitoring points.

[0066] This application makes full use of the advantages of the large range of satellite remote sensing to analyze the health status of vegetation in a wide area outside the economic crop farmland to indirectly judge the insect pest situation in the macro region, providing a comprehensive reference for regional insect pest control, which is conducive to improving the accuracy of monitoring and control and making timely preparations for control. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 A schematic diagram of the basic framework of a satellite and UAV remote sensing monitoring and control system based on insect behavior regulation provided by one embodiment of the present invention;

[0068] Figure 2 A schematic structural diagram of an insect monitoring and control system provided by one embodiment of the present invention;

[0069] Figure 3 A schematic diagram of the structure of a UAV remote sensing integrated processing system provided by one embodiment of the present invention;

[0070] Figure 4 A schematic diagram of the structure of a satellite remote sensing integrated processing system provided by one embodiment of the present invention;

[0071] Figure 5 A schematic diagram of the structure of a visual monitoring and command platform provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0072] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0073] Reference Figure 1-5 , which is one embodiment of the present invention, provides a satellite and UAV remote sensing monitoring and control system based on insect behavior regulation, including: an insect monitoring and control system, a UAV remote sensing integrated processing system, a satellite remote sensing integrated processing system, and a visual monitoring and command platform;

[0074] The insect monitoring and control system is used to regulate the behavior of insects in the target area, collect environmental monitoring data sets, and comprehensively analyze the obtained insect information feature data;

[0075] The UAV remote sensing integrated processing system and the satellite remote sensing integrated processing system combine insect information characteristic data from ground monitoring points to collect UAV multispectral images and satellite remote sensing data related to insect regulatory behavior in the target area, thereby analyzing and obtaining insect occurrence species, insect distribution heat maps, insect density change trends and the extent of insect pest occurrence;

[0076] The visual monitoring and command platform is used to provide users with a visual operation interface, presenting information such as insect occurrence types, insect distribution heat maps, wide-area insect and pest distribution heat maps, insect density change trends, and the degree of pest occurrence to users, supporting users to remotely operate the system and implement real-time pest control plans.

[0077] The insect monitoring and control system is used to collect environmental monitoring data sets, analyze ground insect information feature data, and perform insect behavioral control;

[0078] The UAV remote sensing integrated processing system is used to collect UAV multispectral images of the target area after insect behavior control, and analyze them to obtain insect distribution heat maps and insect density change trends;

[0079] The satellite remote sensing integrated processing system is used to collect and analyze satellite remote sensing data to obtain wide-area pest heat maps;

[0080] The visual monitoring and command platform is used to provide users with a visual operation interface, presenting insect distribution heat maps, insect density change trends and wide-area pest heat maps to users, supporting users to remotely operate the system and implement real-time pest control plans.

[0081] In this embodiment, the insect monitoring and control system is used to collect environmental monitoring data sets, analyze the environmental monitoring data sets to obtain ground insect information feature data and perform insect behavior control;

[0082] The insect behavior regulation includes non-single insect behavior regulation and single insect behavior regulation;

[0083] The insect monitoring and control system includes a cloud-based control and analysis module and a monitoring and control execution module;

[0084] The monitoring and control execution module includes a power supply device, a data remote transmission device, a camera, a temperature and humidity sensor, a soil sensor, a wind direction sensor, a light source device, a pheromone propagation device and an insect sound / vibration signal collection and processing device;

[0085] The insect sound / vibration signal acquisition and processing device comprises a data acquisition card, a microphone probe, a microphone probe, a sound playing device and a vibration exciter;

[0086] When the pest occurrence category in the monitoring area is not single and the monitoring pest target is not directional, the insect monitoring and control system collects real-time environmental data to obtain an environmental monitoring data set, processes the environmental monitoring data set to obtain ground insect information feature data, and performs non-single insect behavioral control based on the ground insect information feature data and the first rule knowledge graph;

[0087] When the type of insect pests in the monitoring area is single and the monitored pest targets are directed, the insect monitoring and control system performs behavioral control of a single insect based on the directed pest target type and the second rule knowledge graph.

[0088] In this embodiment, the ground insect information characteristic data includes insect occurrence species, insect distribution density at monitoring points, and insect age data;

[0089] The processing logic for obtaining ground insect information feature data includes:

[0090] Collect data through sensor arrays to obtain environmental monitoring data sets, and upload the environmental monitoring data sets to the cloud;

[0091] The environmental monitoring data set includes sound / vibration signals, image data, temperature and humidity data, soil data and wind direction data;

[0092] The sensor array includes a microphone probe, a microphone probe, a camera, a temperature and humidity sensor, a soil sensor and a wind direction sensor;

[0093] The cloud-based control and analysis module analyzes the sound / vibration signals in the environmental monitoring dataset to determine the insect species. The module performs noise reduction and reverberation on the sound / vibration signals to obtain vibration time-domain data. Fast Fourier transform (FFT) is then used to obtain vibration spectrum data. Mel-frequency cepstral coefficients, spectrograms, zero-crossing rates, energy, and fundamental frequencies are then calculated based on the vibration time-domain and spectrum data to obtain a time-frequency domain feature set. The pre-trained LSTM recurrent neural network is then used to match the time-frequency domain feature set to determine the insect species.

[0094] The insect occurrence types include insect types and probability values ​​corresponding to each insect type;

[0095] The CNN convolutional neural network model is used to perform target detection on image data in the environmental monitoring dataset to obtain an insect frame selection map. The Deep SORT target tracking algorithm is used to process the insect frame selection maps of consecutive frames to obtain the tracked insect trajectories. The spatial kernel density estimation algorithm is used to calculate the superimposed probability distribution of the tracked insect trajectories to obtain a pixel-level insect swarm density map. The pixel-level insect swarm density map is then integrated to obtain the insect distribution density at the monitoring point.

[0096] Based on the insect occurrence species, insect population estimation, temperature and humidity data and preset insect biological characteristic data, the insect age data is obtained by judgment through the accumulated temperature model. The insect biological characteristic data include the development starting temperature, upper limit temperature, effective accumulated temperature, development threshold, generation cycle information, diapause characteristics, humidity sensitivity, reproduction rate and survival rate of each insect species.

[0097] In this embodiment, the processing logic for regulating non-single insect behavior includes:

[0098] The cloud-based control analysis module uses the first-rule knowledge graph to match ground insect information feature data with expected behaviors to generate the optimal control method;

[0099] The desired behaviors include aggregation, avoidance, migration, immobility and killing;

[0100] The optimal control method includes control behavior and control parameters;

[0101] The regulation behavior includes outputting sound / vibration signals, optical signals, chemical odor pheromones, and sexual attractant pheromones;

[0102] The control parameters include the frequency of the sound / vibration signal, the wavelength of the optical signal, the release concentration and rate of the chemical odor pheromone, and the release concentration and rate of the sex attractant pheromone.

[0103] In this embodiment, the processing logic for regulating the behavior of a single insect includes:

[0104] The cloud-based control analysis module uses the second rule knowledge graph to match the targeted pest target types and expected behaviors to generate inherent control methods. The inherent control methods include targeted control behaviors and targeted control parameters.

[0105] The directional control behavior includes outputting sound / vibration signals, optical signals, chemical odor pheromones, and sexual attractant pheromones;

[0106] The directional control parameters include the frequency of the sound / vibration signal, the wavelength of the optical signal, the release concentration and rate of the chemical odor pheromone, and the release concentration and rate of the sex attractant pheromone.

[0107] Among them, the insect sound / vibration signal acquisition and processing device includes: monitoring function, signal noise reduction function and real-time automatic signal analysis function. It can automatically edit the input signal and complete time domain and spectrum analysis.

[0108] The insect sound / vibration signal acquisition and processing device comprises a microphone and a probe. The microphone system is a multi-point microphone array that collects the relative coordinate characteristics of the sound signal. The probe can collect vibration signals on plants. The device has over 20 sound / vibration acquisition and processing channels, each capable of simultaneous acquisition, with a sampling frequency range of 0 to 80,000 kHz. The direction of the signal source is calculated by comparing the data collected at each array point.

[0109] The cloud-based data processing system includes cloud-based comparison capabilities, including similarity comparison, waveform addition, subtraction, and multiplication, and artificial simulation of signals. This system compares the sound with a sound library to determine species and corresponding behavioral and physiological activities. It also automatically selects and plays corresponding control sound / vibration signals to regulate insect behavior.

[0110] Among them, the insect monitoring and control system of the present application can actively intervene in and control the behavior of target insects through ground monitoring points, overcoming the problem of crops blocking target pests in traditional remote sensing monitoring. By pre-deploying multifunctional sensors in vulnerable areas to monitor and analyze the ground environment in real time, and based on whether the pest category is directional, a two-way generation and matching of specific insect behavioral control plans is achieved. The target insects are induced to produce preset specific behaviors such as aggregation, avoidance, migration or killing, so that the pest targets that were originally difficult to capture by remote sensing under the cover of crop leaves can be displayed from a bird's-eye view, improving the efficiency and accuracy of subsequent drone and satellite remote sensing monitoring, completing the advance preparation for protecting important economic crops, and providing a basis for early warning and accurate identification of pests.

[0111] In this embodiment, the UAV remote sensing integrated processing system is used to collect and obtain multispectral images from the UAV, analyze the multispectral images from the UAV to obtain an insect distribution heat map, and compare the insect density observed by the UAV with the insect distribution density at the monitoring point output by the insect monitoring and control system to generate an insect density change trend;

[0112] The UAV remote sensing integrated processing system includes a UAV remote sensing platform and a UAV image processing module;

[0113] The UAV remote sensing platform includes a payload UAV, a multispectral remote sensing sensor, a GNSS positioning device, a UAV measurement and control device, and a data remote transmission device;

[0114] The UAV image processing module includes a hardware computer and a hyperspectral image analysis program carried by it.

[0115] In this embodiment, the processing logic for analyzing the multispectral image of the drone includes:

[0116] The UAV image acquisition platform is used to complete data collection. A multi-rotor UAV equipped with a hyperspectral sensor acquires images of the target monitoring area to obtain UAV multispectral images, and a GNSS positioning device records the current image acquisition position to obtain UAV GNSS positioning data.

[0117] The data transmission and communication module transmits the drone's multispectral image and drone's GNSS positioning data to a mobile ground data station, and after the drone returns, the local data is synchronously transmitted back to the drone's image processing module;

[0118] The hyperspectral image analysis program carried by the UAV image processing module performs radiation correction, geometric orthogonality and noise suppression on the synchronously transmitted UAV multispectral image to obtain the UAV pre-processed image;

[0119] Principal component analysis was used to screen the spectral channels of the drone pre-processed images to obtain key feature dimensions. The pre-trained support vector machine model was then used to perform inversion processing in combination with the key feature dimensions to obtain the pest distribution and pixel confidence. Based on the pixel confidence, noise suppression and uncertainty correction were performed on the pest distribution to obtain an insect distribution heat map.

[0120] The pixel value integral calculation based on the insect distribution heat map is used to obtain the insect density observed by the drone, and the insect density observed by the drone is compared with the insect distribution density at the monitoring points output by the insect monitoring and control system to generate the insect density change trend.

[0121] The UAV remote sensing platform includes a multi-rotor payload drone, a multispectral remote sensing sensor, an onboard GNSS positioning device, a flight control communication module, a data transmission communication module, and a payload power supply interface. The large multi-rotor drone provides basic payload carrying capacity. The onboard GNSS positioning device provides geographic coordinates for the hyperspectral images collected by the payload to generate GeoTIff data. The flight control communication module receives flight control commands to execute data collection tasks. The data transmission communication module transmits the collected data to the next communication node. The payload power supply interface provides power to the onboard hyperspectral sensor to ensure its normal operation.

[0122] Among them, the control range of the multi-rotor payload drone is not less than 5km, the flight time is not less than 30min, the cruising altitude is not less than 400m, the payload weight is not less than 3kg, and the power supply time for the payload is not less than 15min.

[0123] The sensing range of the multispectral remote sensing sensor should cover the electromagnetic wave bands of 400-1000nm and 2400-2600nm, with a spectral resolution of not less than 10nm, a spatial resolution of not less than 1m, and an imaging time of not longer than 100ms.

[0124] The GNSS positioning device should be positioned through the three-difference conditions, and its positioning accuracy should be achieved: the spatial positioning accuracy error is no higher than 1m in open scenes; the spatial positioning accuracy error is no higher than 5m in complex scenes with high canopy vegetation coverage.

[0125] Remote data transmission devices should include GPRS and 5G remote transmission systems, with an uplink speed of no less than 1Mbps.

[0126] The insect monitoring and control system's ground-based detection points have a limited control area. Integrating drone remote sensing overcomes this limitation, enabling drones to collect and analyze detailed image data down to the centimeter level. Specifically, this application compares the insect density observed by drones with the density of monitoring points output by the insect monitoring and control system, effectively verifying the effectiveness of behavioral insect control. This facilitates accurate identification and assessment of insect pest occurrences, improving the timeliness and effectiveness of monitoring and control efforts.

[0127] In this embodiment, the satellite remote sensing integrated processing system is used to collect satellite remote sensing data and analyze and generate a wide-area insect pest heat map;

[0128] The satellite remote sensing integrated processing system includes a satellite data interface module, a satellite data pre-processing module and a satellite image analysis module;

[0129] Obtain satellite remote sensing data of the monitoring area through the satellite data interface module;

[0130] The satellite remote sensing data includes optical remote sensing image data, atmospheric upwelling radiation data and meteorological data;

[0131] The spatial resolution of the optical remote sensing image data is not less than 5 meters, and the temporal resolution is not less than 3 days, including blue light, green light, red light, near-infrared light and short-wave infrared light bands.

[0132] In this embodiment, the satellite data preprocessing module performs radiation correction, geometric orthogonality, noise suppression, rain cloud removal, and data interpolation processing on the optical remote sensing image data to obtain the satellite image data to be analyzed, and performs inversion calculation on the satellite image data to be analyzed to obtain the satellite remote sensing derivative data;

[0133] The satellite remote sensing derived data include land surface temperature, soil surface moisture, vegetation canopy temperature, surface leaf area index and primary productivity;

[0134] The satellite image analysis module generates remote sensing feature raster layers based on satellite remote sensing derived data, and performs ROI extraction on the optical remote sensing image data based on the UAV GNSS positioning data generated by the UAV remote sensing integrated processing system to obtain vegetation area images;

[0135] The DeeplabV3+ semantic segmentation network is used to analyze remote sensing feature raster layers, vegetation area images, and satellite remote sensing derived data to obtain vegetation impact results. Based on the vegetation impact results, a wide-area pest heat map is generated.

[0136] The vegetation impact results include the severity of the pests and the scope of impact.

[0137] Among them, the spatial resolution of satellite optical remote sensing data shall not be less than 5m, and the temporal resolution shall not be less than 3 days, including at least blue light, green light, red light, near-infrared light and short-wave infrared light bands; the spatial resolution of the inverted meteorological reference data shall not be less than 1km, the temporal resolution shall not be less than 1 hour, and the average confidence level shall not be less than 80%.

[0138] The downlink speed of the satellite remote sensing data interface should be no less than 5Mbps.

[0139] After identifying pests at ground-based monitoring points, this application uses a satellite remote sensing integrated processing system to monitor a wide area surrounding the target cash crop fields. This analysis indirectly determines macro-regional pest conditions by analyzing the vegetation environment, analyzing the severity and impact of pests, and generating a wide-area pest heat map. This provides early warning of nearby pest intrusions and overcomes the coverage limitations of ground-based and drone-based monitoring.

[0140] In this embodiment, the visual monitoring and command platform includes a visual display module, a system control module and a command and dispatch module;

[0141] The visualization display module is used to integrate all output data from the insect monitoring and control system, satellite remote sensing integrated processing system, and UAV remote sensing integrated processing system. It presents insect distribution heat maps, insect density change trends, and wide-area insect pest heat maps on the client through interactive geographic information system (GIS) maps, trend charts, and real-time status panels, supporting users to query data and conduct historical review.

[0142] The system control module is used to support users to remotely operate the system, receive start commands, stop commands, parameter adjustment commands and mode switching commands issued by users to the system, and support users to accurately control software and hardware devices;

[0143] The command and dispatch module is used to receive real-time pest control plans and supports users to send operation instructions corresponding to the real-time pest control plans to the pest control drones for flight control operations. The pest control plans include the type and number of drones, the type of pesticide, the dosage, the application method and the application cycle.

[0144] Those skilled in the art will appreciate that embodiments of the present invention may provide methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0145] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. Satellite and UAV remote sensing monitoring and control system based on insect behavior regulation, characterized by: include: Including insect monitoring and control system, UAV remote sensing integrated processing system, satellite remote sensing integrated processing system and visual monitoring and command platform; The insect monitoring and control system is used to regulate insect behavior in the target area, collect environmental monitoring data sets, and conduct comprehensive analysis to obtain ground insect information feature data; The UAV remote sensing integrated processing system combines ground insect information characteristic data with drone multispectral images of target areas after insect behavior control, thereby analyzing the insect occurrence species, insect distribution heat map and insect density change trends; The satellite remote sensing integrated processing system combines ground insect information characteristic data with satellite remote sensing data collected from target areas after insect behavior regulation, and then analyzes and obtains a wide-area insect pest heat map; The visual monitoring and command platform is used to provide users with a visual operation interface, present pest analysis results to users, and support users to remotely operate the system and implement real-time pest control plans.

2. The satellite and UAV remote sensing monitoring and control system based on insect behavior regulation according to claim 1, characterized in that: The insect monitoring and control system is used to collect environmental monitoring data sets, analyze the environmental monitoring data sets to obtain ground insect information feature data, and perform insect behavior control; The insect behavior regulation includes non-single insect behavior regulation and single insect behavior regulation; The insect monitoring and control system includes a cloud-based control and analysis module and a monitoring and control execution module; The monitoring and control execution module includes a power supply device, a data remote transmission device, a camera, a temperature and humidity sensor, a soil sensor, a wind direction sensor, a light source device, a pheromone propagation device and an insect sound / vibration signal collection and processing device; The insect sound / vibration signal acquisition and processing device comprises a data acquisition card, a microphone probe, a microphone probe, a sound playing device and a vibration exciter; When the pest occurrence category in the monitoring area is not single and the monitoring pest target is not directional, the insect monitoring and control system collects real-time environmental data to obtain an environmental monitoring data set, processes the environmental monitoring data set to obtain ground insect information feature data, and performs non-single insect behavioral control based on the ground insect information feature data and the first rule knowledge graph; When the type of insect pests in the monitoring area is single and the monitored pest targets are directed, the insect monitoring and control system performs behavioral control of a single insect based on the directed pest target type and the second rule knowledge graph.

3. The satellite and UAV remote sensing monitoring and control system based on insect behavior regulation according to claim 2, characterized in that: The ground insect information characteristic data includes insect occurrence species, insect distribution density at monitoring points and insect age data; The processing logic for obtaining ground insect information feature data includes: Collect data through sensor arrays to obtain environmental monitoring data sets, and upload the environmental monitoring data sets to the cloud; The environmental monitoring data set includes sound / vibration signals, image data, temperature and humidity data, soil data and wind direction data; The sensor array includes a microphone probe, a microphone probe, a camera, a temperature and humidity sensor, a soil sensor and a wind direction sensor; The cloud-based control and analysis module analyzes the sound / vibration signals in the environmental monitoring dataset to determine the insect species. The module performs noise reduction and reverberation on the sound / vibration signals to obtain vibration time-domain data. Fast Fourier transform (FFT) is then used to obtain vibration spectrum data. Mel-frequency cepstral coefficients, spectrograms, zero-crossing rates, energy, and fundamental frequencies are then calculated based on the vibration time-domain and spectrum data to obtain a time-frequency domain feature set. The pre-trained LSTM recurrent neural network is then used to match the time-frequency domain feature set to determine the insect species. The insect occurrence types include insect types and probability values ​​corresponding to each insect type; The CNN convolutional neural network model is used to perform target detection on image data in the environmental monitoring dataset to obtain an insect frame selection map. The Deep SORT target tracking algorithm is used to process the insect frame selection maps of consecutive frames to obtain the tracked insect trajectories. The spatial kernel density estimation algorithm is used to calculate the superimposed probability distribution of the tracked insect trajectories to obtain a pixel-level insect swarm density map. The pixel-level insect swarm density map is then integrated to obtain the insect distribution density at the monitoring point. Based on the insect occurrence species, insect population estimation, temperature and humidity data and preset insect biological characteristic data, the insect age data is obtained by judgment through the accumulated temperature model. The insect biological characteristic data include the development starting temperature, upper limit temperature, effective accumulated temperature, development threshold, generation cycle information, diapause characteristics, humidity sensitivity, reproduction rate and survival rate of each insect species.

4. The satellite and UAV remote sensing monitoring and control system based on insect behavior regulation according to claim 2, characterized in that: The processing logic for regulating the behavior of non-single insects includes: The cloud-based control analysis module uses the first-rule knowledge graph to match ground insect information feature data with expected behaviors to generate the optimal control method; The desired behaviors include aggregation, avoidance, migration, immobility and killing; The optimal control method includes control behavior and control parameters; The regulation behavior includes outputting sound / vibration signals, optical signals, chemical odor pheromones, and sexual attractant pheromones; The control parameters include the frequency of the sound / vibration signal, the wavelength of the optical signal, the release concentration and rate of the chemical odor pheromone, and the release concentration and rate of the sex attractant pheromone.

5. The satellite and UAV remote sensing monitoring and control system based on insect behavior regulation according to claim 2, characterized in that: The processing logic for regulating the behavior of a single insect includes: The cloud-based control analysis module uses the second rule knowledge graph to match the targeted pest target types and expected behaviors to generate inherent control methods. The inherent control methods include targeted control behaviors and targeted control parameters. The directional control behavior includes outputting sound / vibration signals, optical signals, chemical odor pheromones, and sexual attractant pheromones; The directional control parameters include the frequency of the sound / vibration signal, the wavelength of the optical signal, the release concentration and rate of the chemical odor pheromone, and the release concentration and rate of the sex attractant pheromone.

6. The satellite and UAV remote sensing monitoring and control system based on insect behavior regulation according to claim 1, characterized in that: The UAV remote sensing integrated processing system is used to collect and obtain UAV multispectral images, analyze the UAV multispectral images to obtain insect distribution heat maps, and compare the insect density observed by the UAV with the insect distribution density at the monitoring points output by the insect monitoring and control system to generate insect density change trends; The UAV remote sensing integrated processing system includes a UAV remote sensing platform and a UAV image processing module; The UAV remote sensing platform includes a payload UAV, a multispectral remote sensing sensor, a GNSS positioning device, a UAV measurement and control device, and a data remote transmission device; The UAV image processing module includes a hardware computer and a hyperspectral image analysis program carried by it.

7. The satellite and UAV remote sensing monitoring and control system based on insect behavior regulation according to claim 6, characterized in that: The processing logic for analyzing UAV multispectral images includes: The UAV image acquisition platform is used to complete data collection. A multi-rotor UAV equipped with a hyperspectral sensor acquires images of the target monitoring area to obtain UAV multispectral images, and a GNSS positioning device records the current image acquisition position to obtain UAV GNSS positioning data. The data transmission and communication module transmits the drone's multispectral image and drone's GNSS positioning data to a mobile ground data station, and after the drone returns, the local data is synchronously transmitted back to the drone's image processing module; The hyperspectral image analysis program carried by the UAV image processing module performs radiation correction, geometric orthogonality and noise suppression on the synchronously transmitted UAV multispectral image to obtain the UAV pre-processed image; Principal component analysis was used to screen the spectral channels of the drone pre-processed images to obtain key feature dimensions. The pre-trained support vector machine model was then used to perform inversion processing in combination with the key feature dimensions to obtain the pest distribution and pixel confidence. Based on the pixel confidence, noise suppression and uncertainty correction were performed on the pest distribution to obtain an insect distribution heat map. Based on the insect distribution heat map, pixel value integral calculation is performed to obtain the insect density observed by the drone. The insect density observed by the drone is then compared with the insect distribution density at the monitoring point output by the insect monitoring and control system to generate the insect density change trend. The pest distribution includes the pest occurrence degree and occurrence probability value corresponding to each pixel point, and the pest occurrence degree includes no pest, mild pest and significant pest.

8. The satellite and UAV remote sensing monitoring and control system based on insect behavior regulation according to claim 1, characterized in that: The satellite remote sensing integrated processing system is used to collect satellite remote sensing data and analyze and generate wide-area pest heat maps; The satellite remote sensing integrated processing system includes a satellite data interface module, a satellite data pre-processing module and a satellite image analysis module; Obtain satellite remote sensing data of the monitoring area through the satellite data interface module; The satellite remote sensing data includes optical remote sensing image data, atmospheric upwelling radiation data and meteorological data; The spatial resolution of the optical remote sensing image data is not less than 5 meters, and the temporal resolution is not less than 3 days, including blue light, green light, red light, near-infrared light and short-wave infrared light bands.

9. The satellite and UAV remote sensing monitoring and control system based on insect behavior regulation according to claim 8, characterized in that: The satellite data preprocessing module performs radiation correction, geometric orthogonality, noise suppression, rain cloud removal and data interpolation on the optical remote sensing image data to obtain the satellite image data to be analyzed, and performs inversion calculation on the satellite image data to be analyzed to obtain satellite remote sensing derivative data; The satellite remote sensing derived data include land surface temperature, soil surface moisture, vegetation canopy temperature, surface leaf area index and primary productivity; The satellite image analysis module generates remote sensing feature raster layers based on satellite remote sensing derived data, and performs ROI extraction on the optical remote sensing image data based on the UAV GNSS positioning data generated by the UAV remote sensing integrated processing system to obtain vegetation area images; The DeeplabV3+ semantic segmentation network is used to analyze remote sensing feature raster layers, vegetation area images, and satellite remote sensing derived data to obtain vegetation impact results. Based on the vegetation impact results, a wide-area pest heat map is generated. The vegetation impact results include the severity of the pests and the scope of impact.

10. The satellite and UAV remote sensing monitoring and control system based on insect behavior regulation according to claim 1, characterized in that: The visual monitoring and command platform includes a visual display module, a system control module, and a command and dispatch module; The visualization display module is used to integrate all output data from the insect monitoring and control system, satellite remote sensing integrated processing system, and UAV remote sensing integrated processing system. It presents insect distribution heat maps, insect density change trends, and wide-area insect pest heat maps on the client through interactive geographic information system (GIS) maps, trend charts, and real-time status panels, supporting users to query data and conduct historical review. The system control module is used to support users to remotely operate the system, receive start commands, stop commands, parameter adjustment commands and mode switching commands issued by users to the system, and support users to accurately control software and hardware devices; The command and dispatch module is used to receive real-time pest control plans and supports users to send operation instructions corresponding to the real-time pest control plans to the pest control drones for flight control operations. The pest control plans include the type and number of drones, the type of pesticide, the dosage, the application method and the application cycle.

Images