Grassland ecological key parameter cooperative monitoring method and system based on unmanned aerial vehicle

Abstract

The present application relates to a method and system for monitoring grassland ecological key parameters based on unmanned aerial vehicles, comprising: selecting a monitoring area; dividing the monitoring area adaptively to obtain coarse-level monitoring plots and fine-level monitoring plots, and determining a preset flight path; setting automatic cruising parameters based on the preset flight path; automatically cruising and collecting data based on the preset flight path and the automatic cruising parameters; and transmitting and storing the collected data. The present application can realize collaborative monitoring, providing a new means for efficient and quantitative monitoring and research of grassland ecological key parameters across regions, and providing the possibility of depicting grassland ecological key parameters at the same time. Large-scale, simultaneous, and long-time-series grassland ecological key parameters can be obtained, which, combined with environmental, climatic, and human impacts, can provide a theoretical and practical basis for further quantitative understanding of the evolution characteristics and mechanisms of grassland structure and function, as well as the development of strategies for pest and disease control and management and utilization measures.

Description

Technical Field

[0001] This invention relates to the field of grassland ecological monitoring technology, and in particular to a method and system for collaborative monitoring of key grassland ecological parameters based on unmanned aerial vehicles (UAVs). Background Technology

[0002] Grassland ecosystems are important ecological barriers and natural resources in my country. Accurate and efficient measurement of grassland plant community structure, landscape characteristics, and the occurrence and development of biological disasters are prerequisites for assessing grassland ecosystem service functions and achieving scientific management and ecological security. However, existing monitoring technologies have significant limitations: 1. Traditional ground-based observation methods: These primarily rely on manual quadrat surveys. While this method offers high accuracy, it suffers from drawbacks such as low efficiency, high cost, the need for professional personnel for on-site operation, destructive sampling, and poor adaptability to harsh environments. Therefore, it is difficult to achieve large-scale, high-frequency repeated monitoring, and even more difficult to construct a large-scale observation network covering different grassland types.

[0003] 2. Satellite remote sensing technology: While enabling large-scale, high-frequency monitoring, its spatial resolution is relatively low (typically ranging from meters to kilometers), making it difficult to identify fine features such as grassland community species, rodent and insect pests, diseases, and grassland fragmentation. Furthermore, satellite remote sensing is susceptible to interference from weather conditions such as clouds and rain, compromising the timeliness and continuity of data acquisition and hindering the achievement of high-precision characterization and long-term, fixed-point monitoring of vegetation, rodent and insect pests, diseases, and landscape features.

[0004] Currently, my country lacks a monitoring technology for key grassland ecological parameters that can balance large-scale coverage with high-precision identification and standardized long-term fixed-point monitoring. Grassland exhibits significant spatial heterogeneity on a large scale; to extend station-based cognitive patterns to regional scales, a new method suitable for networked observation must be developed. Therefore, there is an urgent need for an innovative technological solution that can overcome the aforementioned shortcomings and achieve efficient, high-precision, standardized long-term monitoring. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method and system for collaborative monitoring of key grassland ecological parameters based on unmanned aerial vehicles (UAVs). This method and system aim to effectively integrate UAV aerial photography technology with ecological monitoring needs, solving the problems of low efficiency in traditional ground observation and insufficient resolution in satellite remote sensing. Ultimately, it achieves efficient, high-precision, standardized, long-term fixed-point monitoring of key parameters such as grassland vegetation community structure and landscape characteristics, providing support for large-scale grassland ecological monitoring, research, utilization, and protection.

[0006] To achieve the above objectives, the present invention provides the following solution: A collaborative monitoring method for key grassland ecological parameters based on unmanned aerial vehicles (UAVs) includes: Select monitoring areas; The monitoring area is adaptively divided to obtain coarse-level and fine-level monitoring plots, and a preset flight path is determined. Based on the preset route, set the automatic cruise parameters; Based on the preset route and automatic cruise parameters, automatic cruise and data collection are performed; The collected data is transmitted and stored.

[0007] Optionally, the monitoring area may include: Location marking is performed based on high-resolution maps and BeiDou / GPS positioning technology. The monitoring area is selected by comprehensively considering representativeness, area, topography, and geomorphological features, and the monitoring method is specified.

[0008] Optionally, the coarse-level monitoring plots are divided into: Within the monitoring area, the real-time location of the staff is determined. Based on the real-time location, the staff walks a preset distance in a certain direction, and the direction is identified based on preset points that are temporarily recorded. The preset points are recorded once per second, and the recorded points are automatically deleted after this setting. Based on the identified location, a first preset area with flat terrain is selected as the coarse-level monitoring sample plot; In the coarse-level monitoring sample plot, a first flight route covering the first preset area is planned, and several waypoints are set up evenly distributed on the first flight route.

[0009] Optionally, the fine-level monitoring plots are divided into: Within the coarse-level monitoring plots, several second-preset area regions are randomly selected as the fine-level monitoring plots; In the fine-level monitoring sample plot, a second flight route covering the second preset area is planned, and several waypoints are set evenly distributed on the second flight route.

[0010] Optionally, setting the automatic cruise parameters includes: Depending on the obstacles on site, the flight altitude is appropriately increased by zooming to avoid risks, while ensuring that the ground resolution of the acquired images is higher than the preset resolution to meet the needs of species-level identification. The camera on the drone has a pixel value no lower than the preset pixel value.

[0011] Optionally, autonomous cruise and data collection include: After the drone takes off, it automatically flies to each waypoint in sequence according to the preset route and automatic cruise parameters, and automatically performs vertical downward shooting action at each waypoint; After taking photos of all waypoints by pausing before and after taking photos, the drone automatically returns to the takeoff point and lands. In the coarse-level monitoring plots, standardized, long-term aerial image data is acquired at a preset flight altitude to monitor vegetation cover, bare soil patches, and rodent burrows / mounds within the monitoring range. In the fine-level monitoring plots, standardized, long-term aerial image data is acquired at a preset flight altitude to achieve refined monitoring of grassland plant species and pest and disease characteristics.

[0012] A drone-based collaborative monitoring system for key grassland ecological parameters includes: The monitoring area planning module is used to select the monitoring area; The area division module is used to adaptively divide the monitoring area, obtain coarse-level monitoring plots and fine-level monitoring plots, and determine the preset flight path; The parameter setting module is used to set the automatic cruise parameters based on the preset route. The cruise and data collection module is used to perform automatic cruise and data collection based on the preset route and automatic cruise parameters. The data transmission and storage module is used to transmit and store the collected data.

[0013] Optionally, the monitoring area planning module: Location marking is performed based on high-resolution maps and BeiDou / GPS positioning technology. The monitoring area is selected by comprehensively considering representativeness, area, topography, and geomorphological features, and the monitoring method is specified.

[0014] Optionally, the region division module: Within the monitoring area, the real-time location of the staff is determined. Based on the real-time location, the staff walks a preset distance in a certain direction, and the direction is identified based on preset points that are temporarily recorded. The preset points are recorded once per second, and the recorded points are automatically deleted after this setting. Based on the identified location, a first preset area with flat terrain is selected as the coarse-level monitoring sample plot; In the coarse-level monitoring sample plot, a first flight route covering the first preset area is planned, and several waypoints are set up evenly distributed on the first flight route. Within the coarse-level monitoring plots, several second-preset area regions are randomly selected as the fine-level monitoring plots; In the fine-level monitoring sample plot, a second flight route covering the second preset area is planned, and several waypoints are set evenly distributed on the second flight route.

[0015] Optionally, the cruise and data acquisition module: After the drone takes off, it automatically flies to each waypoint in sequence according to the preset route and automatic cruise parameters, and automatically performs vertical downward shooting action at each waypoint; After taking photos of all waypoints by pausing before and after taking photos, the drone automatically returns to the takeoff point and lands. In the coarse-level monitoring plots, standardized, long-term aerial image data is acquired at a preset flight altitude to monitor vegetation cover, bare soil patches, and rodent burrows / mounds within the monitoring range. In the fine-level monitoring plots, standardized, long-term aerial image data is acquired at a preset flight altitude to achieve refined monitoring of grassland plant species and pest and disease characteristics.

[0016] The beneficial effects of this invention are as follows: This invention realizes a four-in-one monitoring system that is long-term, fixed-point, standardized, and collaborative. It utilizes unmanned aerial vehicles (UAVs) for aerial photography to acquire key ecological parameters such as plant species, grassland rodent and insect pests and diseases, fragmentation, vegetation cover, and biomass. UAVs are highly efficient, easy to operate, and provide high-resolution images, overcoming many difficulties of traditional quadrat surveys and saving significant manpower, resources, and time while increasing the scope and number of sampling points for grassland ecological parameters. This invention enables collaborative monitoring, providing a new method for efficient, quantitative monitoring and research of grassland ecological parameters across regions, and offering the possibility of characterizing grassland ecological parameters simultaneously over a given time period. Acquiring grassland ecological parameters on a large scale, simultaneously over a long period, and in conjunction with environmental, climatic, and anthropogenic influences, provides a theoretical and practical foundation for further quantitative understanding of the evolutionary characteristics and mechanisms of grassland structure and function, as well as for formulating rodent and insect pest control strategies and management and utilization measures. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the collaborative monitoring method for key grassland ecological parameters based on unmanned aerial vehicles (UAVs) according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the multi-level nested design of monitoring plots according to an embodiment of the present invention; Figure 3 This is a system composition block diagram according to an embodiment of the present invention; Figure 4 This is a drone aerial photography data acquisition and processing flow according to an embodiment of the present invention; wherein, (a) is a physical image of the Phantom3 quadcopter drone, (b) is a first-phase fixed-point, fixed-altitude vertical downward aerial image, (c) is a second-phase aerial image of the same location, (d) is a first-phase plateau pika burrow extraction image, (e) is a first-phase vegetation and bare spot extraction image, (f) is a second-phase vegetation and bare spot extraction image, (g) is a second-phase plateau pika burrow extraction image, (h) is a first-phase pika burrow buffer zone image, (i) is a first-phase vegetation and bare spot distribution image within the buffer zone, (j) is a second-phase vegetation and bare spot distribution image within the buffer zone, (k) is a second-phase pika burrow buffer zone image, (l) is a first-phase vegetation and bare spot change image within the buffer zone, (m) is an overall change image of vegetation and bare spots between the two phases, and (n) is a second-phase vegetation and bare spot change image within the buffer zone. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] like Figure 1 As shown, this embodiment proposes a collaborative monitoring method for key grassland ecological parameters based on unmanned aerial vehicles (UAVs), including: S1. Select the monitoring area; S2. Adaptively divide the monitoring area to obtain coarse-level and fine-level monitoring plots, and determine the preset flight path; S3. Based on the preset route, set the automatic cruise parameters; S4. Based on the preset route and automatic cruise parameters, perform automatic cruise and data collection; S5. Transmit and store the collected data.

[0022] Furthermore, the selected monitoring area includes: Location marking is performed based on high-resolution maps and BeiDou / GPS positioning technology. The monitoring area is selected by comprehensively considering representativeness, area, topography, and geomorphological features, and the monitoring method is specified.

[0023] Specifically, in this embodiment, S1. Monitoring sample point planning and setting: Based on research or production needs, the location is marked using a collaborative monitoring method of key grassland ecological parameters by UAVs and high-resolution maps and BeiDou / GPS positioning technology loaded on the application system terminal. The monitoring area is selected by comprehensively considering characteristics such as representativeness, area, terrain, and landform (Setter APP), and the monitoring method is specified (Flighter APP). The selected monitoring area can be named according to the needs to facilitate data management, information extraction, and navigation guidance during subsequent high-frequency monitoring.

[0024] Furthermore, the division of the coarse-level monitoring plots includes: Within the monitoring area, the real-time location of the staff is determined. Based on the real-time location, the staff walks a preset distance in a certain direction, and the direction is identified based on preset points that are temporarily recorded. The preset points are recorded once per second, and the recorded points are automatically deleted after this setting. Based on the identified location, a first preset area with flat terrain is selected as the coarse-level monitoring sample plot; In the coarse-level monitoring sample plot, a first flight route covering the first preset area is planned, and several waypoints are set up evenly distributed on the first flight route.

[0025] The detailed hierarchical classification of monitoring plots includes: Within the coarse-level monitoring plots, several second-preset area regions are randomly selected as the fine-level monitoring plots; In the fine-level monitoring sample plot, a second flight route covering the second preset area is planned, and several waypoints are set evenly distributed on the second flight route.

[0026] Specifically, in this embodiment, S2. Make adaptive divisions and settings (Setter APP and Flighter APP) based on the problem to be solved (local scale) or network construction (regional or larger scale): like Figure 2 As shown, the adaptive division and settings include: Coarse-level monitoring plots: In the selected area, the real-time location of the staff is determined using the GPS / BeiDou-based "Positioning" button on the Setter APP. Then, the "Walking" button is clicked, and the staff walks 5-10m in a certain direction. The location is accurately identified by using the temporarily recorded red points (recorded once per second, but automatically deleted after this setting to save space). Based on this, a relatively flat area of ​​200m × 200m is selected to ensure safe drone flight. A fixed flight altitude (usually set to 20m to ensure flight safety, obtain aerial photo resolution and standardized monitoring, and comparability of long-term series; with the improvement of drone onboard camera zoom technology, the altitude can be adjusted, but the coverage and resolution of the aerial photos must be consistent) is obtained to acquire standardized, long-term aerial photos. This achieves efficient and high-precision monitoring of targets such as vegetation cover, bare soil patches, rodent burrows / mounds within the monitoring range, while also laying the foundation for matching with satellite MODIS data and building an integrated air-space-ground monitoring system.

[0027] Fine-scale monitoring plots: Within the coarse-scale plots, 1-5 areas of 50m × 50m are randomly selected. The real-time location of the staff is determined using the GPS / BeiDou-based "location" button on the Setter APP, and the fine-scale monitoring plots are deployed with reference to the relative positions of the coarse-scale plots. Standardized, long-term aerial photographs are acquired at a fixed flight altitude (usually set to 2m to ensure flight safety, obtain aerial photograph resolution and standardized monitoring, and ensure comparability of long-term series; with the improvement of drone-borne camera zoom technology, the altitude can be adjusted, but the consistency of aerial photograph coverage and resolution must be ensured) to achieve efficient and high-precision monitoring of fine-scale targets such as grassland plant species and pest and disease characteristics, while also matching with satellite Landsat data.

[0028] The selection and layout of sample plots are based on needs and do not necessarily require both methods to be used simultaneously. Typically, for coarse-level sample plots, a flight path covering an area of ​​200m × 200m is planned, containing 16 pre-set, evenly distributed waypoints, which can be set up with a single click on the operating terminal. For fine-level sample plots, a flight path covering an area of ​​40m × 40m is planned, containing 16 pre-set, evenly distributed waypoints, which can also be set up with a single click on the operating terminal.

[0029] Furthermore, setting the automatic cruise parameters includes: Depending on the obstacles on site, the flight altitude is appropriately increased by zooming to avoid risks, while ensuring that the ground resolution of the acquired images is higher than the preset resolution to meet the needs of species-level identification. The camera on the drone has a pixel value no lower than the preset pixel value.

[0030] Specifically, in this embodiment, S3. Standardization setting of automatic cruise parameters: Set fixed flight parameters for the flight path on the terminal to ensure the consistency and comparability of ground resolution and quality of aerial imagery, thereby achieving standardized monitoring. Preferably, depending on the obstacles (such as fences) on site, the flight altitude can be appropriately increased by zooming to avoid risks, but at the same time, it is necessary to ensure that the ground resolution of the acquired images is higher than 1mm to meet the needs of species-level identification.

[0031] The camera on the drone has a resolution of no less than 12 megapixels. Furthermore, the above aerial photography parameters can be adjusted for altitude and speed based on the drone's performance and specific monitoring needs, but must meet the requirements for resolution and standardized monitoring.

[0032] Furthermore, autonomous navigation and data collection include: After the drone takes off, it automatically flies to each waypoint in sequence according to the preset route and automatic cruise parameters, and automatically performs vertical downward shooting action at each waypoint; After taking photos of all waypoints by pausing before and after taking photos, the drone automatically returns to the takeoff point and lands. In the coarse-level monitoring plots, standardized, long-term aerial image data is acquired at a preset flight altitude to monitor vegetation cover, bare soil patches, and rodent burrows / mounds within the monitoring range. In the fine-level monitoring plots, standardized, long-term aerial image data is acquired at a preset flight altitude to achieve refined monitoring of grassland plant species and pest and disease characteristics.

[0033] Specifically, in this embodiment, S4. Automatic cruise and data acquisition: After takeoff, the drone automatically flies to each waypoint in sequence according to the preset routes and flight parameters in S2 and S3, and automatically performs vertical shooting at each waypoint. In order to prevent wind resistance and shaking from causing blurry photos, it is set to pause for 1 second before taking a picture and pause for another second after taking a picture. After taking pictures of all waypoints, the drone automatically returns to the takeoff point and lands.

[0034] S5. Data Transmission and Storage: Standardized aerial imagery data collected by the drone is automatically saved in the drone's storage location (default setting is the memory card). When the "Take Off" button is pressed in the Flighter app on the control terminal, the system automatically generates folders based on time and sorts them numerically. Clicking "Complete" after the flight completes the flight and displays the completion time of the current route. Upon drone return, the aerial photos within the corresponding time period are wirelessly transmitted to the remote controller via the "Download" button. Once network connectivity is available, the aerial photos and corresponding flight path information can be uploaded to the cloud or a local data management platform via the "Upload" button for centralized storage, management, and backup.

[0035] like Figure 3 As shown, this embodiment also proposes a collaborative monitoring system for key grassland ecological parameters based on unmanned aerial vehicles (UAVs), including: The monitoring area planning module is used to select the monitoring area; The area division module is used to adaptively divide the monitoring area, obtain coarse-level monitoring plots and fine-level monitoring plots, and determine the preset flight path; The parameter setting module is used to set the automatic cruise parameters based on the preset route. The cruise and data collection module is used to perform automatic cruise and data collection based on the preset route and automatic cruise parameters. The data transmission and storage module is used to transmit and store the collected data.

[0036] Furthermore, the monitoring area planning module: Location marking is performed based on high-resolution maps and BeiDou / GPS positioning technology. The monitoring area is selected by comprehensively considering representativeness, area, topography, and geomorphological features, and the monitoring method is specified.

[0037] Furthermore, the region division module: Within the monitoring area, the real-time location of the staff is determined. Based on the real-time location, the staff walks a preset distance in a certain direction, and the direction is identified based on preset points that are temporarily recorded. The preset points are recorded once per second, and the recorded points are automatically deleted after this setting. Based on the identified location, a first preset area with flat terrain is selected as the coarse-level monitoring sample plot; In the coarse-level monitoring sample plot, a first flight route covering the first preset area is planned, and several waypoints are set up evenly distributed on the first flight route. Within the coarse-level monitoring plots, several second-preset area regions are randomly selected as the fine-level monitoring plots; In the fine-level monitoring sample plot, a second flight route covering the second preset area is planned, and several waypoints are set evenly distributed on the second flight route.

[0038] Furthermore, the cruise and data acquisition module: After the drone takes off, it automatically flies to each waypoint in sequence according to the preset route and automatic cruise parameters, and automatically performs vertical downward shooting action at each waypoint; After taking photos of all waypoints by pausing before and after taking photos, the drone automatically returns to the takeoff point and lands. In the coarse-level monitoring plots, standardized, long-term aerial image data is acquired at a preset flight altitude to monitor vegetation cover, bare soil patches, and rodent burrows / mounds within the monitoring range. In the fine-level monitoring plots, standardized, long-term aerial image data is acquired at a preset flight altitude to achieve refined monitoring of grassland plant species and pest and disease characteristics.

[0039] A preferred embodiment of this invention is described below: Preparation: Select a typical grassland monitoring area and ensure safe airspace use. Prepare a multi-rotor drone with a software development interface and a visible light camera with a resolution of >12 megapixels. Install the professional drone path planning application of this invention on a built-in monitor or a compatible tablet computer, and load a satellite map of the monitoring area.

[0040] Sample point and route settings: In the application, based on GPS coordinates, a monitoring project named "GS-MQ-01(Gansu-Maqu-01)" is created, and a rectangular area of ​​250m×250m is designated as plot G.

[0041] Within plot G, use the application's "one-click add" function to randomly generate three 50m×50m plots B, named "B01", "B02", and "B03" respectively.

[0042] Plan flight routes for sample plot G: Set a flight range of 200m × 200m, and evenly generate 16 waypoints. Set the flight altitude to 20m, speed to 8m / s, and camera vertically downward. Check whether the flight route conflicts with obstacles such as high-voltage power lines or transmission towers. If there is a conflict, adjust the flight altitude by zooming the camera to confirm that the resolution and monitored area meet the requirements for spatiotemporal comparability.

[0043] Plan flight paths for each Plot B: Set a flight area of ​​40m × 40m and evenly generate 16 waypoints. Set the flight altitude to 2m, speed to 4m / s, and camera vertically downward. Check for conflicts between the flight path and ground obstacles such as fences. If conflicts are found, fine-tune the flight altitude by zooming the camera, while confirming that the resolution and monitored area still meet the requirements for spatiotemporal comparability.

[0044] Monitoring Execution: Connect the drone's built-in display or tablet to the drone's remote controller. At the takeoff point, click the "One-Click Takeoff" button in the application. The drone will automatically execute the flight and photography tasks for plot G and the three plots B in sequence. The entire process requires no manual intervention.

[0045] Data Collection: After the mission is completed, the drone automatically lands. Aerial photos are transmitted to the remote controller's display or tablet, and are automatically divided into folders based on the aerial photography time. The photos are then uploaded to a cloud server via 4G / 5G network. On the server side, image processing software can be used to stitch and correct the photos, and extract ecological parameters such as vegetation cover, species classification, biomass estimation, and landscape pattern index.

[0046] This embodiment has the following significant advantages: Efficiency and economy: The use of drones for automated operations greatly improves monitoring efficiency, reduces manpower, time and economic costs, and makes large-scale, high-frequency monitoring possible.

[0047] High precision and high resolution: By taking aerial photos at low altitude (especially at a height of 2m), millimeter-level ground resolution images are obtained, breaking through the precision limitations of satellite remote sensing and enabling precise identification of subtle features such as grassland plant species and pests.

[0048] Standardization and comparability: By using standardized processes such as fixed flight routes, fixed altitudes, and fixed shooting parameters, the consistency and comparability of data obtained at different times and locations were ensured, laying the foundation for long-term sequence studies and networked comparative analysis.

[0049] Collaborative monitoring and multi-scale integration: The innovative G and B dual-level nested design enables the collaborative collection and seamless correlation of key parameters at the landscape scale (G plot) and the community / species scale (B plot), effectively solving the problem of scale extrapolation in ecology.

[0050] Adaptability and Flexibility: Unmanned aerial vehicle (UAV) platforms are far less affected by weather conditions than satellites and adapt well to complex terrain. Flight path planning is flexible, allowing for rapid adjustments to monitoring plans as needed.

[0051] The following describes the collaborative monitoring method and system for key grassland ecological parameters based on unmanned aerial vehicles (UAVs) proposed in this embodiment. Multi-period, multi-degradation gradient monitoring of plateau pika burrow dynamics, bare soil patch erosion, and grassland ecological parameters was conducted on the alpine meadows of the Qinghai-Tibet Plateau in the Yellow River source region. This fully verifies the feasibility and monitoring effectiveness of the technical solution presented in this application. The specific implementation steps are as follows: 1. Selection of monitoring area: Based on high-resolution satellite maps and BeiDou / GPS positioning technology, the alpine meadows of the Yellow River source region on the Qinghai-Tibet Plateau were selected as the core monitoring area. Four types of typical degradation gradient monitoring plots were divided according to the proportion of bare soil patches: almost no bare patches (NBP, bare patch proportion <5%), low bare patch (LBP, bare patch proportion 5%~20%), medium bare patch (MBP, bare patch proportion 20%~60%), and high bare patch (HBP, bare patch proportion >60%). Each monitoring plot was 200m×100m in size. The area was selected by comprehensively considering the flatness of the terrain, the representativeness of the grassland, and the accessibility of the monitoring plots.

[0052] 2. Two-tiered monitoring plot delineation and pre-planned flight route: According to the adaptive classification rules of the technical solution in this embodiment, a coarse-level + fine-level nested classification was carried out for the four types of monitoring sample plots, and the route and waypoint settings were completed: Coarse-level monitoring plots: Each 200m×100m degradation gradient plot is used as a coarse-level monitoring unit. Flat areas are selected, and a first flight route covering the entire plot is planned. Sixteen waypoints are evenly distributed along the flight route to monitor macro-ecological parameters such as vegetation cover, bare soil patches, and plateau pika burrows / mounds. Fine-grained hierarchical monitoring plots: Within each coarse-grained plot, three 50m×50m areas are randomly selected as fine-grained monitoring units. A second flight route covering the area is planned, and 16 waypoints are evenly distributed to accurately monitor fine-grained characteristics such as grassland plant species and grassland erosion caused by pika disturbance.

[0053] 3. Standardized settings for automatic cruise parameters: The DJI Phantom 3 Pro drone equipped with a 12-megapixel visible light camera was selected, and the automatic cruise parameters were configured according to the parameter rules of this application: the coarse-level monitoring flight altitude was fixed at 20m, and the ground resolution reached 1cm, which met the requirements for accurate identification of mouse and rabbit burrow entrances and bare spot boundaries; there were no obstacles such as fences or power lines on site, so there was no need to adjust the flight altitude. Only the anti-shake rule of pausing for 1 second before taking a picture and pausing for 1 second after taking a picture was set to ensure that the aerial image was clear and the ground resolution of the image was higher than 1mm, which fully met the requirements for species-level and mouse burrow identification.

[0054] 4. Automatic cruise and multi-period data acquisition: During the peak grassland growth period from 2015 to 2018, automatic cruise and data collection were conducted in two phases: After a single-click takeoff, the drones strictly followed preset routes and cruise parameters, sequentially flying to each waypoint and performing automatic vertical downward shooting. After completing the shooting of all waypoints, they automatically returned to the takeoff point and landed. Simultaneously, first-level aerial data such as vegetation cover, bare soil patch area, and plateau pika burrow density were collected from coarse-level plots. Second-level aerial data such as plant species composition and grassland erosion characteristics caused by pika disturbance were collected from fine-level plots, achieving coordinated collection of key ecological parameters over a large scale and long time period. The drone aerial data collection and processing workflow is as follows: Figure 4 As shown; Figure 4 (a) is a physical image of the Phantom 3 quadcopter intelligent drone. Figure 4 (b) is a vertically downward aerial image taken at a fixed point and altitude during the first phase. Figure 4 (c) is a second aerial image of the same location. Figure 4 (d) is the first image extracted from the entrance of a plateau pika burrow. Figure 4 (e) is the first phase of vegetation and bare spot extraction image. Figure 4 (f) is the second phase vegetation and bare spot extraction image. Figure 4 (g) is the image extracted from the entrance of the plateau pika burrow in the second phase. Figure 4 (h) is a map of the buffer zone at the entrance of the pika burrow in the first phase. Figure 4 (i) is a distribution map of vegetation and bare patches within the first-phase buffer zone. Figure 4 (j) is a distribution map of vegetation and bare patches within the second-phase buffer zone. Figure 4 (k) is a map of the buffer zone at the entrance of the pika burrow in the second phase. Figure 4 (l) is a map showing the changes in vegetation and bare patches within the first phase buffer zone. Figure 4 The figure (m) represents the overall changes in vegetation and bare patches over two periods. Figure 4 (n) represents the vegetation and bare soil patch changes within the second phase buffer zone. The first phase was the initial monitoring stage, conducted during the peak grassland growth period of 2015-2016, with the first round of UAV aerial photography and data collection completed at selected monitoring plots along a preset flight path. The second phase, conducted 2-3 years later during the peak grassland growth period of 2017-2018, involved repeated UAV aerial photography and data collection at the same monitoring plots, along the same flight path, and with the same cruise parameters. The data from both phases were used to compare and analyze the dynamic changes of plateau pika burrows and bare soil patches, as well as the evolution of grassland erosion.

[0055] 5. Data transmission, storage, and parsing: Aerial images collected by the drone are automatically stored on the onboard memory card and wirelessly transmitted to the remote controller terminal after returning to base. The images are automatically categorized into folders according to the monitoring time and uploaded to the cloud platform for backup when the network is available. Subsequently, through geometric correction (correction error <3 pixels), image segmentation and feature extraction, core data such as the density of plateau pika burrows, the proportion of bare soil patches, and the vegetation-bare patch conversion rate of each monitoring plot are accurately obtained, and the grassland ecological parameters are quantitatively analyzed.

[0056] Monitoring results of this embodiment: Achieving dual-level collaborative monitoring: the coarse level completes macroscopic monitoring of large-scale bare patches and rodent burrows, while the fine level captures the microscopic correlation between rodent disturbance and grassland erosion, solving the shortcomings of low efficiency and insufficient resolution of traditional ground monitoring and satellite remote sensing. Quantitative analysis of ecological response patterns: In NBP plots, the density of pika burrows and the proportion of bare patches showed a significant positive correlation, with an increase in burrows directly leading to the expansion of bare patches; in LBP plots, burrows and bare patches significantly decreased within 3 years, demonstrating the high resilience of the grassland; in MBP and HBP plots, there were no significant changes in burrows and bare patches, accurately revealing the response mechanisms of grasslands at different degradation stages to pika disturbance; Supporting management decisions: Long-term monitoring data can be directly used for the prevention and control of pests and diseases in plateau pikas and rabbits, and the formulation of strategies for the restoration of degraded alpine meadows, verifying the practical value of the technical solution of this application in the prevention and control of rodent and insect diseases and grassland ecological management.

[0057] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for collaborative monitoring of key grassland ecological parameters based on unmanned aerial vehicles (UAVs), characterized in that, include: Select monitoring areas; The monitoring area is adaptively divided to obtain coarse-level and fine-level monitoring plots, and a preset flight path is determined. Based on the preset route, set the automatic cruise parameters; Based on the preset route and automatic cruise parameters, automatic cruise and data collection are performed; The collected data is transmitted and stored.

2. The method for collaborative monitoring of key grassland ecological parameters based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The selected monitoring area includes: Location marking is performed based on high-resolution maps and BeiDou / GPS positioning technology. The monitoring area is selected by comprehensively considering representativeness, area, topography, and geomorphological features, and the monitoring method is specified.

3. The method for collaborative monitoring of key grassland ecological parameters based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The coarse-level monitoring plots are divided into the following categories: Within the monitoring area, the real-time location of the staff is determined. Based on the real-time location, the staff walks a preset distance in a certain direction, and the direction is identified based on preset points that are temporarily recorded. The preset points are recorded once per second, and the recorded points are automatically deleted after this setting. Based on the identified location, a first preset area with flat terrain is selected as the coarse-level monitoring sample plot; In the coarse-level monitoring sample plot, a first flight route covering the first preset area is planned, and several waypoints are set up evenly distributed on the first flight route.

4. The method for collaborative monitoring of key grassland ecological parameters based on unmanned aerial vehicles (UAVs) according to claim 3, characterized in that, The detailed hierarchical classification of monitoring plots includes: Within the coarse-level monitoring plots, several second-preset area regions are randomly selected as the fine-level monitoring plots; In the fine-level monitoring sample plot, a second flight route covering the second preset area is planned, and several waypoints are set evenly distributed on the second flight route.

5. The method for collaborative monitoring of key grassland ecological parameters based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, Setting the automatic cruise parameters includes: Depending on the obstacles on site, the flight altitude is appropriately increased by zooming to avoid risks, while ensuring that the ground resolution of the acquired images is higher than the preset resolution to meet the needs of species-level identification. The camera on the drone has a pixel value no lower than the preset pixel value.

6. The method for collaborative monitoring of key grassland ecological parameters based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, Automatic cruise and data collection include: After the drone takes off, it automatically flies to each waypoint in sequence according to the preset route and automatic cruise parameters, and automatically performs vertical downward shooting action at each waypoint; After taking photos of all waypoints by pausing before and after taking photos, the drone automatically returns to the takeoff point and lands. In the coarse-level monitoring plots, standardized, long-term aerial image data is acquired at a preset flight altitude to monitor vegetation cover, bare soil patches, and rodent burrows / mounds within the monitoring range. In the fine-level monitoring plots, standardized, long-term aerial image data is acquired at a preset flight altitude to achieve refined monitoring of grassland plant species and pest and disease characteristics.

7. A collaborative monitoring system for key grassland ecological parameters based on unmanned aerial vehicles (UAVs), characterized in that, The system for implementing the method as described in any one of claims 1-6 includes: The monitoring area planning module is used to select the monitoring area; The area division module is used to adaptively divide the monitoring area, obtain coarse-level monitoring plots and fine-level monitoring plots, and determine the preset flight path; The parameter setting module is used to set the automatic cruise parameters based on the preset route. The cruise and data collection module is used to perform automatic cruise and data collection based on the preset route and automatic cruise parameters. The data transmission and storage module is used to transmit and store the collected data.

8. The collaborative monitoring system for key grassland ecological parameters based on unmanned aerial vehicles (UAVs) according to claim 7, characterized in that, The monitoring area planning module: Location marking is performed based on high-resolution maps and BeiDou / GPS positioning technology. The monitoring area is selected by comprehensively considering representativeness, area, topography, and geomorphological features, and the monitoring method is specified.

9. The collaborative monitoring system for key grassland ecological parameters based on unmanned aerial vehicles (UAVs) according to claim 7, characterized in that, The region division module: Within the monitoring area, the real-time location of the staff is determined. Based on the real-time location, the staff walks a preset distance in a certain direction, and the direction is identified based on preset points that are temporarily recorded. The preset points are recorded once per second, and the recorded points are automatically deleted after this setting. Based on the identified location, a first preset area with flat terrain is selected as the coarse-level monitoring sample plot; In the coarse-level monitoring sample plot, a first flight route covering the first preset area is planned, and several waypoints are set up evenly distributed on the first flight route. Within the coarse-level monitoring plots, several second-preset area regions are randomly selected as the fine-level monitoring plots; In the fine-level monitoring sample plot, a second flight route covering the second preset area is planned, and several waypoints are set evenly distributed on the second flight route.

10. The collaborative monitoring system for key grassland ecological parameters based on unmanned aerial vehicles (UAVs) according to claim 7, characterized in that, The cruise and data acquisition module: After the drone takes off, it automatically flies to each waypoint in sequence according to the preset route and automatic cruise parameters, and automatically performs vertical downward shooting action at each waypoint; After taking photos of all waypoints by pausing before and after taking photos, the drone automatically returns to the takeoff point and lands. In the coarse-level monitoring plots, standardized, long-term aerial image data is acquired at a preset flight altitude to monitor vegetation cover, bare soil patches, and rodent burrows / mounds within the monitoring range. In the fine-level monitoring plots, standardized, long-term aerial image data is acquired at a preset flight altitude to achieve refined monitoring of grassland plant species and pest and disease characteristics.

Images