Underground water replenishment amount dynamic evaluation method and system based on unmanned aerial vehicle monitoring

Abstract

The invention discloses a groundwater replenishment amount dynamic evaluation method and system based on unmanned aerial vehicle monitoring, and relates to the field of ecological monitoring, and the method comprises a planning module which is used for uploading a target region land surface three-dimensional model, and creating an unmanned aerial vehicle flight path on the target region land surface three-dimensional model; the acquisition module is used for carrying a multispectral camera and a thermal infrared sensor and acquiring visible light, multispectral and thermal infrared data of the earth surface; according to the invention, the unmanned aerial vehicle carries the multispectral camera and the thermal infrared sensor, so that high-resolution earth surface multi-source data can be dynamically obtained, accurate planning of a flight path is realized through innovative axis network node design, and comprehensiveness, pertinence and accuracy controllability of data acquisition are ensured.

Description

Technical Field

[0001] This invention relates to the field of ecological monitoring technology, specifically to a method and system for dynamic assessment of groundwater recharge based on unmanned aerial vehicle (UAV) monitoring. Background Technology

[0002] At the soil level, groundwater refers to the water existing between soil particles and in pores, and is an important component of soil water. It participates in the water cycle between soil, plants, and the atmosphere, is absorbed and utilized by plant roots, and affects the physicochemical properties of soil and microbial activity. Its content changes are closely related to precipitation, evaporation, and vegetation transpiration.

[0003] Patent application number 202510588110.2 discloses a method for calculating the groundwater recharge from rainfall infiltration under varying groundwater depth scenarios, comprising the following steps: Step 1, segmented impact judgment: Selecting rainfall infiltration events to recharge groundwater, analyzing the combined impact of rainfall and pre-rainfall groundwater depth on the total groundwater recharge, determining whether segmentation exists, and constructing corresponding segmented scenarios; Step 2, total amount model construction: Establishing a total groundwater recharge model for rainfall infiltration under a constant groundwater depth scenario, making the parameters of the total groundwater recharge model for rainfall infiltration under a constant groundwater depth scenario dynamically change with groundwater depth, and constructing a total groundwater recharge model for rainfall infiltration under varying groundwater depth scenarios; Step 3, allocation model construction: Based on the lag allocation effect of the rainfall infiltration groundwater recharge process, constructing a total groundwater recharge model for rainfall infiltration. The application proposes a method to construct an allocation model based on the superposition of groundwater recharge processes, using a lagged allocation weight characteristic function. Step 4 involves parameter calibration of the allocation model: selecting rainfall infiltration recharge events with rainfall occurring only at time step 0, and calibrating the parameters of the lagged allocation weight characteristic function using the 0th time step. Step 5 involves parameter calibration of the total amount model: substituting the calibrated parameter values ​​of the lagged allocation weight characteristic function into the allocation model, and calibrating the parameters of the total groundwater recharge model based on the segmented scenarios in Step 1, thus simulating the rainfall infiltration recharge amount. This application aims to address the problem that "existing technologies do not consider the dynamic relationship between groundwater depth and infiltration recharge coefficient, nor do they consider the lagged allocation process of groundwater depth and rainfall on groundwater recharge, resulting in poor accuracy in estimating infiltration recharge amounts over short periods."

[0004] However, with the increasing demand for water resource management, traditional groundwater recharge assessment methods have limitations such as low data acquisition efficiency, insufficient spatial coverage, and lagging dynamic response, making it difficult to meet the needs of accurate monitoring in complex surface environments. To address this, a dynamic assessment method and system for groundwater recharge based on UAV monitoring is proposed. Summary of the Invention

[0005] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a method and system for dynamic assessment of groundwater recharge based on UAV monitoring, which can effectively solve the problems of the existing technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions; This invention discloses a method and system for dynamic assessment of groundwater recharge based on unmanned aerial vehicle (UAV) monitoring, comprising: The planning module uploads a 3D model of the target area's land surface and creates a UAV flight path on that model. The acquisition module carries a multispectral camera and a thermal infrared sensor to acquire visible light, multispectral, and thermal infrared data of the land surface. The data processing module integrates the visible light, multispectral, and thermal infrared data and identifies water body boundaries, vegetation distribution, and soil moisture. The assessment module establishes a groundwater recharge assessment model to evaluate groundwater recharge. The prediction module receives the groundwater recharge assessment results from the assessment module and predicts the groundwater recharge based on these results.

[0007] Furthermore, during the planning module's operation phase, the system user manually uploads a 3D model of the target area's land surface, and simultaneously selects 3D coordinates above the 3D model of the target area's land surface as nodes for the UAV's flight path. The UAV's flight path is then created by connecting these flight path nodes together. When selecting three-dimensional coordinates above the three-dimensional model of the land surface of the target area, a planar grid is configured above the three-dimensional model of the land surface of the target area from a top-down perspective. The two-dimensional coordinates of the center point of each grid are used as the reference coordinates of the UAV flight path nodes. Then, z-axis parameters are configured for each reference coordinate. The three-dimensional coordinates used as the UAV flight path nodes are obtained by combining the two-dimensional coordinates and z-axis parameters. In this system, the distance between adjacent axes in the grid is equal. The distance between adjacent axes in the grid follows the principle that the higher the system evaluation accuracy requirement, the smaller the distance between adjacent axes, and vice versa. When configuring z-axis parameters for each reference coordinate, the vertical distance between each three-dimensional coordinate obtained by configuring z-axis parameters and the land surface of the target area's three-dimensional land surface model is equal.

[0008] Furthermore, the multispectral camera and thermal infrared sensor mounted on the acquisition module are detachably connected to the UAV platform, and the multispectral camera covers a band range of 400-2500nm, including the red-edge band and the near-infrared band. The thermal infrared sensor has a temperature resolution of no more than 0.1℃ and a spatial resolution of no less than 10m. The acquisition module is also equipped with a GPS positioning module and an inertial measurement module. The GPS positioning module and the inertial measurement module are used to acquire the source geographical location and attitude information of the acquired data, respectively. The geographic location information includes absolute coordinates, and the attitude information includes heading angle, pitch angle, roll angle, flight speed, and acceleration.

[0009] Furthermore, the multispectral camera and thermal infrared sensor mounted on the acquisition module are detachably connected to the UAV platform, and the multispectral camera covers a band range of 400-2500nm, including the red-edge band and the near-infrared band. The thermal infrared sensor has a temperature resolution of no more than 0.1℃ and a spatial resolution of no less than 10m. The acquisition module is also equipped with a GPS positioning module and an inertial measurement module. The GPS positioning module and the inertial measurement module are used to acquire the source geographical location and attitude information of the acquired data, respectively. The geographic location information includes absolute coordinates, and the attitude information includes heading angle, pitch angle, roll angle, flight speed, and acceleration.

[0010] Furthermore, when the data processing module integrates visible light, multispectral, and thermal infrared data, it performs integration operations based on a multi-source data fusion algorithm to identify water body boundaries, vegetation distribution, and soil moisture stages, and performs automatic identification based on an improved image segmentation algorithm. The integration operation execution phase of the multi-source data fusion algorithm performs registration, correction, and fusion processing on visible light, multispectral, and thermal infrared data. Specifically, it includes: performing geometric correction on multispectral and visible light data based on ground control points, performing atmospheric correction on thermal infrared data using a radiative transfer model, achieving spatial registration of multi-source data through a feature point matching algorithm, and fusing the registered data through a weighted fusion algorithm. The improved image segmentation algorithm uses a deep learning semantic segmentation model to classify and segment surface information based on training samples. The training samples include surface images and their annotation information for different land use types, different soil moisture, and different vegetation coverage. Among them, the ground control points are defined by the system user in the three-dimensional model of the land surface of the target area.

[0011] Furthermore, the groundwater recharge assessment model of the assessment module is calculated using hydrogeological parameters and key surface parameters; The hydrogeological parameters include permeability coefficient, effective porosity, and hydraulic gradient, which are obtained by collecting geological survey data and long-term monitoring data of the target area; the key surface parameters include surface temperature, vegetation index, soil moisture, and water area, which are obtained by the processing results of the data processing module. The groundwater recharge assessment results in the aforementioned assessment model are as follows: Q = PETR; In the formula: Q is the groundwater recharge; P is the precipitation; E is the evaporation; T is the vegetation transpiration; R is the surface runoff; Evaporation E and vegetation transpiration T are obtained by using surface temperature and vegetation index, while surface runoff R is obtained by using soil moisture and water area.

[0012] Furthermore, the evaluation module is equipped with a control unit at its lower level, which is used to control the continuous operation of the evaluation module; The control unit has several triggering conditions that can be manually set by the system user. The control unit controls the evaluation module to run continuously based on the triggering conditions. The triggering conditions include: preset time interval, precipitation exceeding a preset threshold, and soil moisture change rate exceeding a preset threshold.

[0013] Furthermore, the prediction logic for groundwater recharge based on the assessment results in the prediction module is expressed as follows: ; In the formula: This represents the predicted groundwater recharge for the nth period. This is the current estimated value for groundwater recharge. This is a comprehensive correction factor; These are the weighting coefficients; For the current period, the change rate of soil moisture, the change rate of vegetation index, the change rate of surface temperature, the change rate of precipitation, and the change in recharge in period i; This is the trend memory coefficient; in, The values ​​are user-defined on the system side, and all are positive numbers. The sum is 1. The initial default value is 0.25.

[0014] Furthermore, the planning module interacts with the acquisition module and the data processing module via a wireless network. The acquisition module is internally connected to a cloud database via a wireless network. The data processing module interacts with the evaluation module via a wireless network. The evaluation module has a control unit connected to it via a wireless network. The evaluation module interacts with the prediction module via a wireless network.

[0015] On the other hand, a dynamic assessment method for groundwater recharge based on UAV monitoring includes: Upload a 3D model of the target area's land surface, determine and connect the drone's flight path nodes above the model to form a flight path; acquire visible light, multispectral, and thermal infrared data of the land surface, along with corresponding geographical location and attitude information, using a drone equipped with a multispectral camera, thermal infrared sensor, positioning module, and inertial measurement module; register, correct, and fuse the acquired data using a multi-source data fusion algorithm, and then identify water body boundaries, vegetation distribution, and soil moisture using an improved image segmentation algorithm; assess groundwater recharge using collected hydrogeological parameters and key surface parameters obtained from data processing; continuously execute groundwater recharge assessments according to preset triggering conditions; and predict future groundwater recharge by combining the current groundwater recharge assessment value, comprehensive correction coefficient, rate of change of various factors, and trend memory coefficient.

[0016] Compared with the known prior art, the technical solution provided by this invention has the following beneficial effects: This invention provides a method and system for dynamic assessment of groundwater recharge based on UAV monitoring. During the execution of this method and system, a multispectral camera and a thermal infrared sensor are mounted on the UAV to dynamically acquire high-resolution multi-source surface data. The innovative axis network node design enables precise planning of the flight path, ensuring the comprehensiveness, relevance, and controllability of data collection. Meanwhile, by utilizing multi-source data fusion algorithms and improved image segmentation algorithms, data can be efficiently integrated and analyzed, and key information such as water body boundaries and vegetation distribution can be accurately identified, providing a reliable basis for groundwater recharge assessment. The recharge amount can be calculated by combining the assessment model with hydrogeological parameters and key surface parameters, and continuous dynamic assessment can be achieved based on preset triggering conditions. Furthermore, the prediction logic comprehensively considers the rate of change of multiple factors and the trend memory coefficient, which can accurately predict the future trend of recharge, providing intelligent, efficient and precise management and protection services for groundwater. Attached Figure Description

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

[0018] Figure 1 This is a schematic diagram of a dynamic assessment system for groundwater recharge based on UAV monitoring. Figure 2 This is a flowchart illustrating a method for dynamically assessing groundwater recharge based on drone monitoring. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0020] The present invention will be further described below with reference to embodiments.

[0021] Example 1: This embodiment presents a dynamic assessment system for groundwater recharge based on unmanned aerial vehicle (UAV) monitoring, such as... Figure 1 As shown, it includes: The planning module is used to upload a 3D model of the target area's land surface and create a drone flight path on the 3D model of the target area's land surface. During the planning module operation phase, the system user manually uploads a 3D model of the target area's land surface and simultaneously selects 3D coordinates above the 3D model of the target area's land surface as nodes for the UAV's flight path. The UAV's flight path is created by connecting the UAV's flight path nodes to each other. When selecting three-dimensional coordinates above the three-dimensional model of the land surface of the target area, a planar grid is configured above the three-dimensional model of the land surface of the target area from a top-down perspective. The two-dimensional coordinates of the center point of each grid are used as the reference coordinates of the UAV flight path nodes. Then, z-axis parameters are configured for each reference coordinate. The three-dimensional coordinates used as the UAV flight path nodes are obtained by combining the two-dimensional coordinates and z-axis parameters. In this system, the spacing between adjacent axes in the grid is equal. The spacing between adjacent axes in the grid follows the principle that the higher the system evaluation accuracy requirement, the smaller the spacing between adjacent axes, and vice versa. When configuring z-axis parameters for each reference coordinate, the principle is that the vertical distance between each three-dimensional coordinate obtained by configuring z-axis parameters and the land surface of the three-dimensional model of the target area is equal. The acquisition module is used to carry a multispectral camera and a thermal infrared sensor to acquire visible light, multispectral and thermal infrared data of the Earth's surface; The acquisition module is equipped with a multispectral camera and a thermal infrared sensor that can be detachably connected to the UAV platform. The multispectral camera covers a band range of 400-2500nm, including the red-edge band and the near-infrared band. The thermal infrared sensor has a temperature resolution of no more than 0.1℃ and a spatial resolution of no less than 10m. The acquisition module is also equipped with a GPS positioning module and an inertial measurement module. The GPS positioning module and the inertial measurement module are used to acquire the source geographical location and attitude information of the acquired data, respectively. The geographic location information includes absolute coordinates, and the attitude information includes heading angle, pitch angle, roll angle, flight speed, and acceleration. The data acquisition module has an internal cloud database, which is used to store the data collected and the processing results of the acquisition module. The cloud database adopts a distributed storage architecture, including multiple storage nodes. Data is stored between the storage nodes through data redundancy and load balancing mechanisms. The cloud database is equipped with a data backup module and a recovery module. The data backup module is used to perform incremental backups and full backups of data according to preset periods. The recovery module is used to restore system data by backing up data when the system fails or malfunctions. The data processing module is used to integrate visible light, multispectral and thermal infrared data, and to identify water body boundaries, vegetation distribution and soil moisture; When the data processing module integrates visible light, multispectral and thermal infrared data, it performs integration operations based on multi-source data fusion algorithms, identifies water body boundaries, vegetation distribution and soil moisture stages, and performs automatic identification based on improved image segmentation algorithms. The integration operation execution phase of the multi-source data fusion algorithm involves registration, correction, and fusion processing of visible light, multispectral, and thermal infrared data. Specifically, this includes: geometric correction of multispectral and visible light data based on ground control points; atmospheric correction of thermal infrared data using a radiative transfer model; spatial registration of multi-source data using a feature point matching algorithm; and fusion of the registered data using a weighted fusion algorithm. The improved image segmentation algorithm uses a deep learning semantic segmentation model to classify and segment surface information based on training samples. The training samples include surface images and their annotation information for different land use types, different soil moisture, and different vegetation cover. Among them, the ground control points are defined by the system user in the three-dimensional model of the land surface of the target area; It should be noted that: Multi-source data fusion algorithm operation: Geometric correction: Based on user-defined ground control points (such as road intersections and other obvious feature points), geometric correction is performed on multispectral and visible light data. By establishing a transformation model, geometric distortion caused by factors such as UAV flight attitude is eliminated, ensuring the accurate location of ground features in the image.

[0022] Atmospheric correction: Thermal infrared data is processed using a radiative transfer model to simulate the atmospheric transmission of thermal infrared radiation, remove the absorption and scattering effects of gases such as water vapor, and restore the true surface temperature.

[0023] Spatial registration: Using feature point matching algorithms, feature points such as edges and corners of images from different data sources are extracted and matched. The data are then unified to the same coordinate system through a transformation matrix to achieve spatial alignment.

[0024] Data fusion: Weights are assigned based on the characteristics of each data source (such as high spatial resolution of visible light and rich spectral information of multispectral sources), and an image with the advantages of multiple sources is generated through a weighted fusion algorithm.

[0025] Improved image segmentation algorithm for recognition: A deep learning semantic segmentation model, such as U-Net, was employed and trained on labeled samples containing different land use types, soil moisture, and vegetation cover. After learning features such as color and texture, the model classifies new images pixel by pixel. Water body boundaries are identified based on their low reflectivity in the near-infrared band and thermal infrared temperature characteristics. Vegetation distribution: extracted using high reflectivity in the red-edge and near-infrared bands, and coverage assessed using NDVI; Soil moisture: inferred from a combination of multispectral reflectance and thermal infrared temperature.

[0026] The assessment module is used to establish a groundwater recharge assessment model and assess the groundwater recharge. The groundwater recharge assessment model in the assessment module is calculated using hydrogeological parameters and key surface parameters; Hydrogeological parameters, including permeability coefficient, effective porosity, and hydraulic gradient, are obtained by collecting geological survey data and long-term monitoring data of the target area; key surface parameters, including surface temperature, vegetation index, soil moisture, and water area, are obtained through the processing results of the data processing module. The assessment results of groundwater recharge in the assessment model are as follows: Q = PETR; In the formula: Q is the groundwater recharge; P is the precipitation; E is the evaporation; T is the vegetation transpiration; R is the surface runoff; Among them, evaporation E and vegetation transpiration T are obtained by using surface temperature and vegetation index, and surface runoff R is obtained by using soil moisture and water area. For evaporation E, vegetation transpiration T, and surface runoff R: ; In the formula: This is a regional correction factor; For reference crop evapotranspiration; This is the surface temperature-vegetation index coupling function; This is a correction term for meteorological factors; Among them, the regional correction coefficient Calibrated using historical evapotranspiration data, the value is set between 0.8 and 1.2. ; In the formula: The surface temperature obtained by the drone's thermal infrared sensor; The normalized vegetation index was calculated using the red and near-infrared bands of a multispectral camera. Ideal temperature for vegetation canopy and extreme high temperature threshold; , The fitting coefficients are set to 0.5 and 0.8. ; In the formula: Wind speed acquired by the drone's inertial measurement module; The wind speed and humidity coupling coefficient; Relative humidity of the air; ; In the formula: The vegetation type correction factor is 0.7 for forest, 0.5 for grassland, and 0.6 for farmland. Leaf area index; This is the temperature-vapor pressure difference stress function; This is a soil moisture correction term; Among them, leaf area index Obtained by inversion of the multispectral red-edge band ; In the formula: , The stress sensitivity coefficient is set to 0.2 or 1.5. This is the optimal transpiration temperature for vegetation; For water vapor pressure difference and drought stress threshold; ; In the formula: Soil moisture retrieved by drone; Field holding capacity; These are permanent wilting points, obtained from geological survey data; ; In the formula: The runoff coefficient is taken as 0.6 for bare land and 0.2 for forest land; This refers to precipitation. The slope-humidity coupling coefficient is set to 0.8. The slope of the ground; This is a dynamic correction term for runoff area minus soil moisture. in, ; In the formula: This is the confluence correction factor, set to 0.02; The catchment area and the saturated hydraulic conductivity of the soil are considered. This represents the maximum saturated hydraulic conductivity. The evaluation module is equipped with a control unit, which is used to control the continuous operation of the evaluation module. The control unit has several triggering operating conditions manually set by the system user. The control unit controls the evaluation module to run continuously based on the triggering operating conditions. The triggering operating conditions include: preset time interval, precipitation exceeding a preset threshold, and soil moisture change rate exceeding a preset threshold. The prediction module is used to receive the groundwater recharge assessment results from the evaluation module and predict the groundwater recharge based on the assessment results. The prediction logic for groundwater recharge based on the assessment results in the prediction module is expressed as follows: ; In the formula: This represents the predicted groundwater recharge for the nth period. This is the current estimated value for groundwater recharge. This is a comprehensive correction factor; These are the weighting coefficients; For the current period, the change rate of soil moisture, the change rate of vegetation index, the change rate of surface temperature, the change rate of precipitation, and the change in recharge in period i; This is the trend memory coefficient; in, The values ​​are user-defined on the system side, and all are positive numbers. The sum is 1. The initial default value is 0.25; For trend memory coefficient : ; In the formula: This is the initial trend memory coefficient, with a default value of 0.3. For historical data periods; This is the time decay factor; This represents the estimated groundwater recharge for period k. This represents the estimated groundwater recharge for period k−1. The average absolute value of historical changes in supply volume; Among them, time decay factor =1 / k, For a sign function, when hour, =1, when hour, =0, when hour, =-1; The planning module interacts with the acquisition module and the data processing module via a wireless network. The acquisition module is internally connected to a cloud database via a wireless network. The data processing module interacts with the evaluation module via a wireless network. The evaluation module has a control unit connected to it via a wireless network. The evaluation module interacts with the prediction module via a wireless network.

[0027] In this embodiment, the planning module uploads a 3D model of the target area's land surface and creates a UAV flight path on the model. The acquisition module, equipped with a multispectral camera and thermal infrared sensor, acquires visible light, multispectral, and thermal infrared data of the land surface. The cloud database synchronously stores the data and processing results acquired by the acquisition module. The data processing module then integrates the visible light, multispectral, and thermal infrared data and identifies water body boundaries, vegetation distribution, and soil moisture. The evaluation module further establishes a groundwater recharge assessment model to assess the groundwater recharge. The control unit continuously controls the evaluation module in real time. Finally, the prediction module receives the groundwater recharge assessment results from the evaluation module and predicts the groundwater recharge based on the assessment results.

[0028] The system described in the above embodiments, which uses drones to monitor groundwater recharge, offers several positive benefits. It leverages drone-borne equipment to acquire high-precision surface data, which, after processing and evaluation using a model, enables dynamic assessment and prediction of groundwater recharge. This not only improves data acquisition efficiency and assessment accuracy but also allows for timely responses to environmental changes, providing precise data support for water resource management and assisting relevant departments in making scientific decisions. This is of great significance for the protection and rational utilization of groundwater.

[0029] It is important to note that: Drone hardware adaptation and compatibility: It is recommended to clearly define the model and specifications of the drone platform (such as payload and flight time) to ensure the stability of the multispectral camera, thermal infrared sensor, and positioning module. For example, select a drone with wind resistance of ≥6 levels to adapt to the flight requirements of complex terrain.

[0030] Data acquisition time window optimization: In view of the characteristics of thermal infrared data being affected by diurnal temperature differences, it is recommended to specify the data collection time period (such as 6-8 am or 6-8 pm) to reduce the interference of surface temperature fluctuations on the assessment results, and at the same time, clarify the collection frequency for different seasons (such as once a week in the rainy season and once every two weeks in the dry season).

[0031] Quality control of the data processing module: Before fusing multi-source data, add a data preprocessing step: Denoising filters (such as Gaussian filters) are applied to visible light data to eliminate the effects of cloud cover. Radiometric calibration of multispectral data is performed to ensure the accuracy of band reflectance. Perform defect repair on thermal infrared data to avoid interference from abnormal temperature values.

[0032] Additional boundary conditions for the evaluation model: When the target area has karst landforms or groundwater over-extraction, additional parameters (such as the degree of karst development and the distribution of mining wells) need to be introduced to correct the model. Assessment correction rules under extreme weather conditions (such as heavy rain and drought): For example, when precipitation P exceeds the historical average by 200%, the calculation of surface runoff R needs to take into account the change in soil saturated hydraulic conductivity.

[0033] Security mechanisms for cloud databases: Add a data encryption module (such as AES-256 encryption) to the distributed storage architecture to encrypt GPS coordinates and sensor data during transmission, and set access permission levels (administrator, analyst, visitor) to prevent data leakage.

[0034] Example 2: At the implementation level, based on Example 1, this example refers to... Figure 2 A further detailed description is provided of the groundwater recharge dynamic assessment system based on UAV monitoring in Example 1: A method for dynamic assessment of groundwater recharge based on UAV monitoring includes: Step 1: Upload a 3D model of the target area's land surface, determine and connect the drone's flight path nodes above the model to form a flight path; Step 2: Using a drone equipped with a multispectral camera, thermal infrared sensor, positioning module and inertial measurement module, acquire visible light, multispectral and thermal infrared data of the ground surface and the corresponding geographical location and attitude information; Step 3: The acquired data is registered, corrected and fused according to the multi-source data fusion algorithm, and then the water body boundary, vegetation distribution and soil moisture are identified by the improved image segmentation algorithm; Step 4: Assess groundwater recharge by using the collected hydrogeological parameters and key surface parameters obtained from data processing; Step 5: Control the continuous execution of groundwater recharge assessment according to the preset triggering conditions; Step 6: Combine the current groundwater recharge assessment value, comprehensive correction coefficient, change rate of each factor, and trend memory coefficient to predict the future groundwater recharge.

[0035] In summary, the methods and systems described in the above embodiments, during execution, utilize a drone equipped with a multispectral camera and thermal infrared sensor to dynamically acquire high-resolution multi-source surface data. Innovative axis network node design enables precise flight path planning, ensuring the comprehensiveness, relevance, and controllable accuracy of data acquisition. Simultaneously, by employing multi-source data fusion algorithms and improved image segmentation algorithms, data can be efficiently integrated and analyzed, accurately identifying key information such as water body boundaries and vegetation distribution, providing a reliable basis for groundwater recharge assessment. The recharge amount is calculated using an assessment model combined with hydrogeological parameters and key surface parameters, while continuous dynamic assessment is achieved based on preset trigger conditions. Furthermore, the prediction logic comprehensively considers the rate of change of multiple factors and the trend memory coefficient, accurately predicting future recharge trends, thus providing intelligent, efficient, and precise management and protection services for groundwater.

[0036] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A dynamic assessment system for groundwater recharge based on unmanned aerial vehicle (UAV) monitoring, characterized in that, include: The planning module is used to upload a 3D model of the target area's land surface and create a drone flight path on the 3D model of the target area's land surface. The acquisition module is used to carry a multispectral camera and a thermal infrared sensor to acquire visible light, multispectral and thermal infrared data of the Earth's surface; The data processing module is used to integrate visible light, multispectral and thermal infrared data, and to identify water body boundaries, vegetation distribution and soil moisture; The assessment module is used to establish a groundwater recharge assessment model and assess the groundwater recharge. The prediction module receives the groundwater recharge assessment results from the evaluation module and predicts the groundwater recharge based on the assessment results.

2. The dynamic assessment system for groundwater recharge based on UAV monitoring according to claim 1, characterized in that, During the planning module's operation phase, the system user manually uploads a 3D model of the target area's land surface and simultaneously selects 3D coordinates above the 3D model of the target area's land surface as nodes for the UAV's flight path. The UAV's flight path is created by connecting these flight path nodes together. When selecting three-dimensional coordinates above the three-dimensional model of the land surface of the target area, a planar grid is configured above the three-dimensional model of the land surface of the target area from a top-down perspective. The two-dimensional coordinates of the center point of each grid are used as the reference coordinates of the UAV flight path nodes. Then, z-axis parameters are configured for each reference coordinate. The three-dimensional coordinates used as the UAV flight path nodes are obtained by combining the two-dimensional coordinates and z-axis parameters. In this system, the distance between adjacent axes in the grid is equal. The distance between adjacent axes in the grid follows the principle that the higher the system evaluation accuracy requirement, the smaller the distance between adjacent axes, and vice versa. When configuring z-axis parameters for each reference coordinate, the vertical distance between each three-dimensional coordinate obtained by configuring z-axis parameters and the land surface of the target area's three-dimensional land surface model is equal.

3. The dynamic assessment system for groundwater recharge based on UAV monitoring according to claim 1, characterized in that, The multispectral camera and thermal infrared sensor mounted on the acquisition module are detachably connected to the UAV platform, and the multispectral camera covers a band range of 400-2500nm, including the red-edge band and the near-infrared band. The thermal infrared sensor has a temperature resolution of no more than 0.1℃ and a spatial resolution of no less than 10m. The acquisition module is also equipped with a GPS positioning module and an inertial measurement module. The GPS positioning module and the inertial measurement module are used to acquire the source geographical location and attitude information of the acquired data, respectively. The geographic location information includes absolute coordinates, and the attitude information includes heading angle, pitch angle, roll angle, flight speed, and acceleration.

4. The dynamic assessment system for groundwater recharge based on UAV monitoring according to claim 3, characterized in that, The acquisition module is equipped with a cloud database, which is used to store the data collected and the processing results of the acquisition module. The cloud database adopts a distributed storage architecture, including multiple storage nodes. The storage nodes store data through data redundancy and load balancing mechanisms. The cloud database is equipped with a data backup module and a recovery module. The data backup module is used to perform incremental backups and full backups of data according to a preset period. The recovery module is used to restore system data by backing up the data when the system experiences operational failures or errors.

5. The dynamic assessment system for groundwater recharge based on UAV monitoring according to claim 1, characterized in that, When the data processing module integrates visible light, multispectral and thermal infrared data, it performs integration operations based on a multi-source data fusion algorithm, identifies water body boundaries, vegetation distribution and soil moisture stages, and performs automatic identification based on an improved image segmentation algorithm. The integration operation execution phase of the multi-source data fusion algorithm performs registration, correction, and fusion processing on visible light, multispectral, and thermal infrared data. Specifically, it includes: performing geometric correction on multispectral and visible light data based on ground control points, performing atmospheric correction on thermal infrared data using a radiative transfer model, achieving spatial registration of multi-source data through a feature point matching algorithm, and fusing the registered data through a weighted fusion algorithm. The improved image segmentation algorithm uses a deep learning semantic segmentation model to classify and segment surface information based on training samples. The training samples include surface images and their annotation information for different land use types, different soil moisture, and different vegetation coverage. Among them, the ground control points are defined by the system user in the three-dimensional model of the land surface of the target area.

6. The dynamic assessment system for groundwater recharge based on UAV monitoring according to claim 1, characterized in that, The groundwater recharge assessment model of the assessment module is calculated using hydrogeological parameters and key surface parameters. The hydrogeological parameters include permeability coefficient, effective porosity, and hydraulic gradient, which are obtained by collecting geological survey data and long-term monitoring data of the target area; the key surface parameters include surface temperature, vegetation index, soil moisture, and water area, which are obtained by the processing results of the data processing module. The groundwater recharge assessment results in the aforementioned assessment model are as follows: Q = PETR; In the formula: Q is the groundwater recharge; P is the precipitation; E is the evaporation; T is the vegetation transpiration; R is the surface runoff; Evaporation E and vegetation transpiration T are obtained by using surface temperature and vegetation index, while surface runoff R is obtained by using soil moisture and water area.

7. A dynamic assessment system for groundwater recharge based on UAV monitoring according to claim 1, characterized in that, The evaluation module is equipped with a control unit at its lower level, which is used to control the continuous operation of the evaluation module. The control unit has several triggering conditions that can be manually set by the system user. The control unit controls the evaluation module to run continuously based on the triggering conditions. The triggering conditions include: preset time interval, precipitation exceeding a preset threshold, and soil moisture change rate exceeding a preset threshold.

8. A dynamic assessment system for groundwater recharge based on UAV monitoring according to claim 1, characterized in that, The prediction logic for groundwater recharge based on the assessment results in the prediction module is expressed as follows: ； In the formula: This represents the predicted groundwater recharge for the nth period. This is the current estimated value for groundwater recharge. This is a comprehensive correction factor; These are the weighting coefficients; For the current period, the change rate of soil moisture, the change rate of vegetation index, the change rate of surface temperature, the change rate of precipitation, and the change in recharge in period i; This is the trend memory coefficient; in, The values ​​are user-defined on the system side, and all are positive numbers. The sum is 1. The initial default value is 0.

25.

9. A dynamic assessment system for groundwater recharge based on UAV monitoring according to claim 1, characterized in that, The planning module interacts with the acquisition module and the data processing module via a wireless network. The acquisition module is internally connected to a cloud database via a wireless network. The data processing module interacts with the evaluation module via a wireless network. The evaluation module has a control unit connected to it via a wireless network. The evaluation module interacts with the prediction module via a wireless network.

10. A method for dynamic assessment of groundwater recharge based on UAV monitoring, wherein the method is an implementation method of the dynamic assessment system for groundwater recharge based on UAV monitoring as described in any one of claims 1-9, characterized in that, include: Step 1: Upload a 3D model of the target area's land surface, determine and connect the drone's flight path nodes above the model to form a flight path; Step 2: Using a drone equipped with a multispectral camera, thermal infrared sensor, positioning module and inertial measurement module, acquire visible light, multispectral and thermal infrared data of the ground surface and the corresponding geographical location and attitude information; Step 3: The acquired data is registered, corrected and fused according to the multi-source data fusion algorithm, and then the water body boundary, vegetation distribution and soil moisture are identified by the improved image segmentation algorithm; Step 4: Assess groundwater recharge by using the collected hydrogeological parameters and key surface parameters obtained from data processing; Step 5: Control the continuous execution of groundwater recharge assessment according to the preset triggering conditions; Step 6: Combine the current groundwater recharge assessment value, comprehensive correction coefficient, change rate of each factor, and trend memory coefficient to predict the future groundwater recharge.

Images