Soil environment monitoring method and system based on multistage sampling

Abstract

The invention relates to the field of environment monitoring, in particular to a soil environment monitoring method and system based on multistage sampling. The soil environment monitoring method provided by the invention comprises the following steps: acquiring a GIS model of a target area; based on the GIS model of the target area, arranging a first-level data acquisition module in the target area from a macroscopic scale, and identifying at least one abnormal sub-area in the target area by using first-level data acquired by the first-level data acquisition module; based on the GIS model of the abnormal sub-region, second-level data acquisition modules are arranged in the abnormal sub-region from the mesoscopic scale, and at least one abnormal point location in the abnormal sub-region is identified by using second-level data acquired by the second-level data acquisition modules; based on the GIS model of the abnormal point locations, third-level data acquisition modules are arranged at all the abnormal point locations from the microscopic scale, and third-level data acquired by all the third-level data acquisition modules are used for recognizing the soil condition of all the abnormal point locations.

Description

Technical Field

[0001] The present invention relates to the field of monitoring, and in particular, to a soil environment monitoring method and system based on multi-level sampling. Background Art

[0002] With the accelerating advancement of urbanization and industrialization, heavy metal pollution, residues of organic persistent poisons, and accumulation of excessive nutrients such as nitrogen and phosphorus are continuously eroding the soil environment of farmland, park green spaces, construction land, etc., bringing potential ecological and health risks. Therefore, there is an urgent need for a multi-level sampling soil environment monitoring method and system that can balance large-scale screening and local precise diagnosis to achieve efficient, intelligent, and dynamic pollution identification and assessment. Summary of the Invention

[0003] The present invention provides a soil environment monitoring method based on multi-level sampling in one aspect. The soil environment monitoring method based on multi-level sampling includes the following steps: Obtain the GIS model of the target area; Based on the GIS model of the target area, deploy the first-level data acquisition module in the target area from a macroscopic scale, and use the first-level data collected by the first-level data acquisition module to identify at least one abnormal sub-area in the target area; Based on the GIS model of the abnormal sub-area, deploy the second-level data acquisition module in the abnormal sub-area from a mesoscopic scale respectively, and use the second-level data collected by the second-level data acquisition module to identify at least one abnormal point in the abnormal sub-area; Based on the GIS model of the abnormal point, deploy the third-level data acquisition module at each abnormal point from a microscopic scale respectively, and use the third-level data collected by each third-level data acquisition module to identify the soil conditions at each abnormal point.

[0004] In this embodiment or some other embodiments, for the soil environment monitoring method based on multi-level sampling provided by the present invention, The carrier of the first-level data acquisition module is a space platform, and the first-level data is used to evaluate the soil environment situation of the target area; The carrier of the second-level data acquisition module is an aerial platform, and the second-level data is used to evaluate the soil environment situation of the abnormal sub-area; The carrier of the third-level data acquisition module is a ground platform, and the third-level data is used to evaluate the soil conditions at the abnormal point.

[0005] In this embodiment or some other embodiments, based on the GIS model, deploying the first-level data acquisition module in the target area from a macroscopic scale includes the following steps: Based on the GIS model of the target area, divide its horizontal spatial area at the first resolution to obtain a number of first-level sub-areas.

[0006] In this embodiment or some other embodiments, identifying at least one abnormal sub-area in the target area by using the first-level data collected by the first-level data collection module includes the following steps: Based on the single-source first-level data, identify at least one abnormal sub-area in the target area, where the single-source means that the first-level data is the only type of first-level target data; or Based on the multi-source first-level data, identify at least one abnormal sub-area in the target area, where the multi-source means that the first-level data includes at least two types of first-level target data.

[0007] In this embodiment or some other embodiments, identifying at least one abnormal sub-area in the target area based on the multi-source first-level data includes the following steps: Respectively use each type of first-level target data in the multi-source first-level data to identify the abnormal sub-areas in the target area, and in the GIS model of the target area, merge the abnormal sub-areas identified based on each type of first-level target data.

[0008] In this embodiment or some other embodiments, identifying at least one abnormal sub-area in the target area based on the single-source first-level data includes the following steps: Preprocess the single-source data of each first-level sub-area at each sampling moment to obtain a first-level data set, where one data in the first-level data set is the standardized instantaneous value corresponding to the single-source data of a first-level sub-area at a sampling moment; Use the first-level data set to obtain at least one time feature vector and at least one spatial feature vector of each data; Respectively use the time feature vector and spatial feature vector of each data to generate the abnormal score of the corresponding data, and respectively obtain the average abnormal score of the corresponding first-level sub-area according to the abnormal scores of the single-source data of each first-level sub-area at each sampling moment; Set a sliding window and a moving step size, and based on the sliding window and the moving step size, obtain a number of first-level sub-area clustering groups; Respectively generate the score threshold of the corresponding clustering group based on the average abnormal scores of multiple first-level sub-areas in each clustering group, where the score threshold minimizes the binary classification entropy of the corresponding clustering group, and obtain the global score threshold according to the score threshold of each clustering group; Based on the global score threshold, compare the average abnormal score of each first-level sub-area one by one, and mark the first-level sub-areas with the average abnormal score exceeding the global score threshold as abnormal sub-areas.

[0009] In this embodiment or some other embodiments, based on the GIS model, a second-level data acquisition module is arranged in each abnormal sub-region at the meso scale, including the following steps: Based on the GIS model corresponding to the abnormal sub-region, divide its horizontal spatial region at the second resolution to obtain a number of second-level sub-regions, and the second resolution is higher than the first resolution; Arrange a second-level data acquisition module in each second-level sub-region to collect at least one type of second-level target data in the corresponding sub-region respectively.

[0010] In this embodiment or some other embodiments, based on the second-level data collected by the second-level data acquisition module, identify at least one abnormal point in each abnormal sub-region, including the following steps: According to the second-level target data and its corresponding abnormal threshold, mark the abnormal second-level sub-regions from the several second-level sub-regions, and set the central position of each abnormal second-level sub-region as the abnormal point.

[0011] In this embodiment or some other embodiments, based on the GIS model, a third-level data acquisition module is arranged at the micro scale in each abnormal point, including the following steps: Based on the GIS model corresponding to the abnormal point, divide its vertical spatial region at the third resolution to obtain a number of soil sampling points, and at each soil sampling point, use the third-level data acquisition module to collect soil samples, and the soil samples are used to evaluate the soil condition.

[0012] On the other hand, the soil environment monitoring system based on multi-level sampling provided by the present invention includes a processor and a memory in an embodiment provided by the present invention. A computer program is stored in the memory, and when the computer program is executed by the processor, the soil environment monitoring method based on multi-level sampling provided in any of the above embodiments is implemented.

[0013] The gain of the soil environment monitoring method and system based on multi-level sampling provided by the present invention is that through the coordination of multi-level sampling from macro to meso to micro and the fusion of multi-level data, the present invention realizes the rapid screening, precise positioning and high-precision profile soil data monitoring of soil environment anomalies in the target area. Description of the Drawings

[0014] Figure 1 Flowchart of the soil environment monitoring method based on multi-level sampling provided by the embodiment of the present invention; Figure 2 Flowchart of identifying abnormal sub-regions based on single-source data provided by the embodiment of the present invention; Figure 3Schematic diagram of the system structure of the soil environment monitoring method based on multi-level sampling provided by the embodiments of the present invention. Detailed implementation manners

[0015] In the following description, specific details such as systems, structures, and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present application. Those skilled in the art should understand that the present application can also be implemented in other embodiments without these specific details.

[0016] It should be noted that in the description of the present application, the detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application; in addition, terms such as "first" and "second" are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0017] In one embodiment, please refer to Figure 1 , Figure 1 is the flowchart of the soil environment monitoring method based on multi-level sampling provided by the embodiments of the present invention; as Figure 1 shown, the soil environment monitoring method based on multi-level sampling provided by the embodiments of the present invention includes the following steps: S01. Obtain the GIS model of the target area.

[0018] It can be understood that the target area described in the present invention refers to the geographical range that needs to be monitored for the soil environment, which can be delimited by factors such as administrative boundaries, terrain features, or pollution suspected areas.

[0019] Furthermore, the GIS model of the target area described in the present invention is a comprehensive data structure for digitally expressing the target area; it includes multi-source spatial information such as digital elevation model (DEM), soil attribute layers, land use, and hydrographic water systems and ground obstacles; it forms a unified spatial reference framework through geocoding and vector / raster data fusion to support the path planning, positioning navigation, and data analysis of the subsequent multi-level data acquisition module.

[0020] Specifically, for the GIS model of any target area in step S01, it can be obtained through the following steps: obtain high-precision terrain data through unmanned aerial vehicle aerial survey or satellite remote sensing, and combine the existing cadastral database to import information such as soil texture, soil organic matter, land use type, and hydrographic geographical elements to complete the construction of the regional digital elevation model and attribute layers.

[0021] In some other embodiments, the GIS model obtained in step S01 is also marked with obstacles (such as buildings, dense vegetation or water bodies) that may affect the monitoring, so as to avoid or focus on them during path planning and sampling point selection in the meso and micro stages, thereby laying a spatial foundation for subsequent macro screening and dynamic scheduling.

[0022] S02. Based on the GIS model of the target area, deploy the first-level data acquisition module within the target area at the macro scale, and use the first-level data collected by the first-level data acquisition module to identify at least one abnormal sub-area within the target area.

[0023] It should be noted that the first-level data acquisition module of the present invention is deployed relying on a space platform, and the space platform includes but is not limited to spaceborne platforms such as remote sensing satellites, high-altitude balloons or suborbital unmanned aerial vehicles.

[0024] It can be understood that the space platform relied on by the present invention can conduct all-weather and large-range continuous monitoring of a target area of dozens to hundreds of square kilometers in the high-altitude or near-space environment to ensure the real-time mastery and dynamic update of the overall soil environment situation of the area.

[0025] Further, relying on the above space platform, in this embodiment, at least one first-level data acquisition module is carried to obtain at least one type of first-level target data within the target area.

[0026] Specifically, the first-level data acquisition module of the present invention includes but is not limited to one or a combination of multiple sensing devices such as multi-spectral / hyperspectral imagers, infrared thermal radiation sensors, synthetic aperture radars (SAR) and lidars (LiDAR).

[0027] Among them, the multi-spectral / hyperspectral imager is used to obtain the reflectance information of the ground surface in visible light, near-infrared and short-wave infrared bands; the infrared thermal radiation sensor is used to measure the surface thermal radiation temperature distribution; the synthetic aperture radar is used to penetrate clouds and vegetation to obtain radar echo profile signals; the lidar is used to obtain high-precision digital elevation models and vegetation canopy structure data.

[0028] Furthermore, based on the above sensing devices, first-level target data such as multi-band reflectance data, surface temperature field data, radar scattering coefficient and echo profile, digital elevation model and canopy height / density, and quantitative inversion of soil surface moisture content through multi-source fusion can be obtained respectively, providing a reliable spatial and temporal basis for subsequent grid-based fine layout and abnormal sub-area positioning at the meso level.

[0029] In this embodiment, to better identify the abnormal sub-areas within the target area, step S02 also implements the following steps: Based on the GIS model of the target area, divide its horizontal spatial area at the first resolution to obtain a number of first-level sub-areas.

[0030] Specifically, in step S02, the target area is gridded on the horizontal plane according to a fixed size, that is, the target area is divided into a number of non-overlapping first grid cells, and any one first grid cell corresponds to a first-level sub-area; further, the first resolution can be equivalently the geometric parameters of the first grid cell, such as side length, area, etc.

[0031] In a specific embodiment, for an agricultural experimental field with a total area of about 12 km 2 Based on its GIS model, the first grid cells are divided on the horizontal plane in the corresponding area at a fixed resolution of 50 m × 50 m. The internal spatial characteristics of each first grid cell are relatively uniform, which can provide a clear geometric reference for arranging the data acquisition module and avoid blind spots and redundant coverage.

[0032] It can be understood that in some other specific embodiments, the setting of the first resolution can also be flexibly adjusted according to other factors, such as cost, monitoring accuracy requirements, road network conditions, etc., to adapt to different application scenarios.

[0033] In the embodiment, the identification of any abnormal sub-area is achieved by analyzing the first-level data collected by the first-level data acquisition module in a sampling period; further, the identification of at least one abnormal sub-area in the target area described in step S02 includes the following steps: Based on the single-source first-level data, identify at least one abnormal sub-area in the target area, where the single-source means that the first-level data is the only type of first-level target data; or Based on the multi-source first-level data, identify at least one abnormal sub-area in the target area, where the multi-source means that the first-level data includes at least two types of first-level target data; The identification of at least one abnormal sub-area in the target area based on the multi-source first-level data includes the following steps: Respectively use each type of first-level target data in the multi-source first-level data to identify the abnormal sub-areas in the target area, and in the GIS model of the target area, merge the abnormal sub-areas identified based on each type of first-level target data.

[0034] In this embodiment, based on any type of first-level target data, the identification process of the abnormal sub-area is as Figure 2 shown: S021. Preprocess the single-source data of each first-level sub-area at each sampling moment to obtain a first-level data set, where one data in the first-level data set is the standardized instantaneous value corresponding to the single-source data of a first-level sub-area at a sampling moment.

[0035] It is understandable that the single-source data described in step S021 is a type of first-level target data.

[0036] Furthermore, the preprocessing described in step S021 includes but is not limited to steps such as noise filtering, missing value step size, and standardization.

[0037] Specifically, for the standardization of the single-source data of any first-level sub-region, the following formula is satisfied: , where is the standardized instantaneous value of the first-level sub-region i at the sampling time t, is the single-source data of the first-level sub-region i at the sampling time t, represents the average value of the single-source data of the first-level sub-region i within the sampling period, represents the standard deviation of the single-source data of the first-level sub-region i within the sampling period.

[0038] S022. Use the first-level data set to obtain at least one time feature vector and at least one space feature vector for each data.

[0039] Furthermore, the time feature vector based on each data includes the standardized instantaneous value (i.e., the data itself), the moving average value of the standardized instantaneous value within the sampling period, the difference value between the standardized instantaneous value at the previous sampling time, and the fluctuation amplitude of the standardized instantaneous value within the sampling period; it should be noted that in some other embodiments, the time feature vector may also include other time series feature statistics.

[0040] Furthermore, the space feature vector based on each data includes the domain average standardized instantaneous value and the maximum difference of the domain standardized instantaneous value; it should be noted that in some other embodiments, the space feature vector may also include other space feature statistics.

[0041] S023. Respectively use the time feature vector and the space feature vector of each data to generate the anomaly score corresponding to the data, and respectively obtain the average anomaly score of the corresponding first-level sub-region according to the anomaly scores of the single-source data of each first-level sub-region at each sampling time.

[0042] Furthermore, the anomaly score of any data is obtained by weighted summation of the multi-dimensional Mahalanobis distance score and the vector autoregressive residual score, that is , , is the fusion weight, used to weigh the weights of the multi-dimensional Mahalanobis distance score and the vector autoregressive residual score, and its specific parameter values can be adjusted according to the actual situation, is the multi-dimensional Mahalanobis distance score, is the vector autoregressive residual score.

[0043] Specifically, the multi-dimensional Mahalanobis distance score of any data is obtained through the following calculation model: , where is the multi-dimensional Mahalanobis distance score corresponding to the standardized instantaneous value of the lower-level sub-region i at the sampling time t, is the total feature vector composed of the time feature vector and the space feature vector of the lower-level sub-region i at the sampling time t, , m takes the total number of the time feature vector and the space feature vector, , represents the vector composed of the empirical means corresponding to the elements in the total feature vector, is the vector transpose, , represents the vector covariance matrix.

[0044] It should be noted that the larger the multi-dimensional Mahalanobis distance score, the more the overall performance of the data (including trend, fluctuation, and neighborhood difference) in the corresponding first-level sub-region deviates from the normal performance in history.

[0045] Specifically, the vector autoregressive residual score of any data is obtained through the following calculation model: , , where , , and are the prediction parameter matrix and the prediction parameter vector fitted by the least squares method based on the standardized instantaneous values of the first-level sub-region i before the sampling time t, represents the total feature vector composed of the time feature vector and the space feature vector of the first-level sub-region i at the sampling time t - 1, represents the total feature prediction vector composed of the time feature vector and the space feature vector of the first-level sub-region i at the sampling time t, represents the second-order norm between the total feature vector and the total feature prediction vector of the first-level sub-region i at the sampling time t - 1, is the corresponding vector autoregressive residual score.

[0046] It should be noted that the larger the vector autoregressive residual score, the more serious the deviation of the data in the corresponding first-level sub-region from the "normal" evolution model based on the past day, that is, it may correspond to a sudden anomaly.

[0047] Furthermore, the average anomaly score of any first-level sub-region is the mean value of the anomaly scores at different sampling times within the sampling period of the first-level sub-region.

[0048] S024. Set a sliding window and a moving step size, and obtain several first-level sub-region clustering groups based on the sliding window and the moving step size.

[0049] It can be understood that the sliding window is a region with a fixed size, which can include at least one first-level sub-region, and the moving step size is the number of grids that the sliding window moves within the grid region.

[0050] Furthermore, the sliding window and the moving step size can be set according to the actual situation, so as to balance the calculation efficiency and the result fineness while maintaining the local threshold adaptability, thereby efficiently and accurately identifying abnormal first-level sub-regions.

[0051] S025. Generate a score threshold for each corresponding clustering group based on the average abnormal scores of multiple first-level sub-regions in each clustering group. The score threshold minimizes the binary classification entropy of the corresponding clustering group, and obtain a global score threshold according to the score threshold of each clustering group.

[0052] Furthermore, for the average abnormal scores of multiple first-level sub-regions in each clustering group, try different score thresholds one by one until the best score threshold that makes the binary classification entropy of abnormal first-level sub-regions and normal first-level sub-regions the lowest value is found.

[0053] Specifically, the score threshold of any clustering group is obtained through the following objective model: , , , , where Table represents the total number of first-level sub-regions in the k-th clustering group, represents the average abnormal score of the j-th first-level sub-region in the k-th clustering group, and respectively represent the proportions of first-level sub-regions determined as "normal" and "abnormal" according to the score threshold , represents the binary classification entropy obtained based on the score threshold .

[0054] Furthermore, the global score threshold is the mean of the best score thresholds of different clustering groups. While taking into account the local optimal discrimination effect within each clustering group, it can provide a unified abnormal determination standard for the entire target region.

[0055] S026. Based on the global score threshold, compare the average abnormal scores of each first-level sub-region one by one, and mark the first-level sub-regions whose average abnormal scores exceed the global score threshold as abnormal sub-regions.

[0056] Further, when there are multiple abnormal sub-regions that are spatially continuous, the multiple spatially adjacent abnormal sub-regions are aggregated into an integrated abnormal sub-region to ensure that subsequent operations can perform a one-time unified layout and sampling of the complete and connected pollution patches, thereby improving the sampling efficiency, avoiding repeated coverage, and ensuring the integrity of the boundaries of the abnormal regions.

[0057] S03. Based on the GIS model of the abnormal sub-region, at the meso-scale, a second-level data acquisition module is deployed within the abnormal sub-region respectively, and at least one abnormal point within the abnormal sub-region is identified by using the second-level data collected by the second-level data acquisition module.

[0058] In this embodiment, the second-level data acquisition module is deployed relying on an aerial platform, and the aerial platform includes but is not limited to multi-rotor drones, fixed-wing drones, or manned helicopters.

[0059] It can be understood that the aerial platform can perform flexible and maneuverable grid flight inspections on the identified abnormal sub-regions within a height range of dozens of meters to hundreds of meters to ensure high-resolution and short-delay monitoring of the near-surface environment of the hotspot areas.

[0060] Further, in this embodiment, at least one second-level data acquisition module is carried on the aerial platform to achieve refined acquisition of key chemical and physical parameters within the abnormal sub-region.

[0061] Specifically, the second-level data acquisition module can be one or a combination of a portable multi-parameter electrochemical sensor array, a portable X-ray fluorescence spectrometer (XRF), a handheld near-infrared spectrometer, or a high-resolution visible light camera: the electrochemical sensor array is used for on-line measurement of pH, conductivity, and oxidation-reduction potential (ORP) of the soil surface layer; the portable XRF is used for on-site qualitative or semi-quantitative analysis of heavy metal elements (such as Cd, Pb, As, Hg); the near-infrared spectrometer is used to capture the reflectance in the 400–2500 nm band to estimate the soil organic matter content and moisture content; the high-resolution camera is used to obtain images of the surface morphology and vegetation cover.

[0062] Based on the above-mentioned sensing devices, this embodiment can respectively obtain second-level target data such as soil surface layer pH / conductivity / ORP data, heavy metal semi-quantitative concentration data, near-infrared spectral reflectance data, and high-precision surface image data, providing accurate positioning and pre-judgment basis for subsequent in-depth profile sampling and risk assessment at the micro level.

[0063] Further, to better implement data acquisition in the abnormal sub-region, step S03 of this embodiment also implements the following steps: Based on the GIS model corresponding to the abnormal sub-region, divide its horizontal spatial region at the second resolution to obtain a number of second-level sub-regions, where the second resolution is higher than the first resolution; Deploy second-level data acquisition modules in each second-level sub-region to respectively collect at least one type of second-level target data within the corresponding sub-region.

[0064] In this embodiment, based on the second-level data collected by the second-level data acquisition modules, identifying at least one abnormal point in each abnormal sub-region includes the following steps: According to the second-level target data and its corresponding abnormal threshold, mark the abnormal second-level sub-regions from the several second-level sub-regions, and set the center position of each abnormal second-level sub-region as an abnormal point.

[0065] It should be noted that the abnormal threshold described in this embodiment refers to the critical value preset for each monitoring index in the second-level data (such as soil surface pH, conductivity, ORP, heavy metal semi-quantitative concentration, near-infrared reflectance, etc.), which can distinguish the normal range from the abnormal range, and is used to mark whether there is an abnormal signal in the second-level sub-region.

[0066] To ensure the scientificity and reliability of the determination, the setting of the abnormal threshold can refer to the limit values and empirical statistical values of specific pollutants or physical and chemical indicators in national or local soil environmental quality regulations and industry technical specifications; for example, the reasonable range of pH (6.5 - 8.5), the content of heavy metal cadmium (Cd) not exceeding 0.3 mg / kg, the conductivity not exceeding 400 μS / cm, etc. can be used as the upper and lower limits of abnormality for the corresponding indicators: when the measured value in a certain second-level sub-region exceeds the above standards, it can be determined that there is a pollution risk in this region.

[0067] S04. Based on the GIS model of the abnormal points, deploy third-level data acquisition modules at each abnormal point respectively at the micro scale, and use the third-level data collected by each third-level data acquisition module to identify the soil conditions at each abnormal point.

[0068] The third-level data acquisition module described in this embodiment is deployed relying on a ground platform, and the ground platform includes but is not limited to an automatic drilling vehicle, a wheeled or tracked unmanned ground vehicle, and the platform is equipped with a core sampling device and a sample preservation and preliminary processing unit.

[0069] It can be understood that the ground platform can perform automated and multi-depth stratified sampling on abnormal points within the range of centimeters to meters to ensure the true restoration and high-precision detection of the soil profile structure and pollutants at different depths.

[0070] Further, in this embodiment, a core sampling system, an automatic stratified sample dispenser, and on-site sensors are mounted on the ground platform: the core sampling system is used to automatically obtain soil core samples at preset depths such as 0–10 cm, 10–30 cm, 30–50 cm, etc.; the stratified sample dispenser is used to immediately number and seal the core samples at different depths; the on-site sensors (such as portable soil moisture and temperature probes) are used to measure the immediate moisture content and temperature of the sampling points.

[0071] Based on the above equipment, this embodiment can obtain stratified soil samples, in-situ moisture / temperature data, and on-site warning indicators respectively, providing high-quality and traceable samples and data support for subsequent heavy metal quantification (ICP-MS), organic pollutant analysis (GC-MS), total organic carbon (TOC) determination, and ecological enzyme activity detection in the laboratory.

[0072] In one embodiment, based on the soil environment monitoring method based on multi-level sampling provided in the above embodiment, a soil environment monitoring system based on multi-level sampling is further provided in the embodiment. The system includes a processor and a memory, and a computer program is stored in the memory. When the computer program is executed by the processor, the soil environment monitoring method based on multi-level sampling provided in any of the above embodiments is implemented.

[0073] Specifically, the processor provided in this embodiment is respectively signal-connected to the first-level data acquisition module, the second-level data acquisition module, and the third-level data acquisition module arranged in the above method to obtain the first-level data, the second-level data, and the third-level data collected by the corresponding data acquisition modules; further, based on the first-level data, the second-level data, and the third-level data respectively, the identification of abnormal sub-regions, the positioning of abnormal points, and the acquisition of soil samples at abnormal points are realized.

[0074] The above is only an exemplary description of the present invention in the specific implementation process, and is not intended to limit the protection scope of the present invention. Any equivalent replacement or equivalent change of the technical solutions described in the spirit of the present invention and the claims shall be included in the protection scope of the present invention.

Claims

1. A soil environment monitoring method based on multi-level sampling, characterized in that, Including the following steps: Obtain the GIS model of the target area; Based on the GIS model of the target area, deploy the first-level data acquisition module in the target area from a macroscopic scale, and use the first-level data collected by the first-level data acquisition module to identify at least one abnormal sub-area in the target area; Based on the GIS model of the abnormal sub-area, deploy the second-level data acquisition module in the abnormal sub-area respectively from a mesoscopic scale, and use the second-level data collected by the second-level data acquisition module to identify at least one abnormal point in the abnormal sub-area; Based on the GIS model of the abnormal point, deploy the third-level data acquisition module at each abnormal point respectively from a microscopic scale, and use the third-level data collected by each third-level data acquisition module to identify the soil condition at each abnormal point.

2. The soil environment monitoring method based on multi-level sampling according to claim 1, characterized in that: The carrier of the first-level data acquisition module is a space platform, and the first-level data is used to evaluate the soil environment situation of the target area; The carrier of the second-level data acquisition module is an aerial platform, and the second-level data is used to evaluate the soil environment situation of the abnormal sub-area; The carrier of the third-level data acquisition module is a ground platform, and the third-level data is used to evaluate the soil condition at the abnormal point.

3. The soil environment monitoring method based on multi-level sampling according to claim 2, characterized in that Based on the GIS model, deploying the first-level data acquisition module in the target area from a macroscopic scale includes the following steps: Based on the GIS model of the target area, divide its horizontal spatial area at the first resolution to obtain a number of first-level sub-areas.

4. The soil environment monitoring method based on multi-level sampling according to claim 3, wherein Using the first-level data collected by the first-level data acquisition module to identify at least one abnormal sub-area in the target area includes the following steps: Based on the single-source first-level data, identify at least one abnormal sub-area in the target area, where the single-source means that the first-level data is the only type of first-level target data; or Based on the multi-source first-level data, identify at least one abnormal sub-area in the target area, where the multi-source means that the first-level data includes at least two types of first-level target data.

5. The soil environmental monitoring method based on multi-level sampling according to claim 4, wherein Based on the multi-source first-level data, identifying at least one abnormal sub-area in the target area includes the following steps: Respectively use each type of first-level target data in the multi-source first-level data to identify the abnormal sub-areas in the target area, and in the GIS model of the target area, merge the abnormal sub-areas identified based on each type of first-level target data.

6. The soil environmental monitoring method based on multi-level sampling according to any one of claims 4-5, characterized in that, Based on the single-source first-level data, identifying at least one abnormal sub-area in the target area includes the following steps: Preprocess the single-source data of each first-level sub-area at each sampling moment to obtain a first-level data set, where one data in the first-level data set is the standardized instantaneous value corresponding to the single-source data of a first-level sub-area at a sampling moment; Using the first-level data set, obtain at least one time feature vector and at least one spatial feature vector of each data; Respectively use the time feature vector and spatial feature vector of each data to generate the abnormal score of the corresponding data, and respectively obtain the average abnormal score of the corresponding first-level sub-area according to the abnormal scores of the single-source data of each first-level sub-area at each sampling moment. Set a sliding window and a moving step size, and based on the sliding window and the moving step size, obtain several first-level sub-region clustering groups; Based on the average anomaly scores of multiple first-level sub-regions in each clustering group respectively, generate a score threshold for the corresponding clustering group, where the score threshold minimizes the binary classification entropy of the corresponding clustering group, and obtain a global score threshold according to the score threshold of each clustering group; Based on the global score threshold, compare the average anomaly scores of each first-level sub-region one by one, and mark the first-level sub-regions with average anomaly scores exceeding the global score threshold as abnormal sub-regions.

7. The soil environment monitoring method based on multi-level sampling according to claim 3, wherein Based on the GIS model, deploy a second-level data acquisition module in each abnormal sub-region from a meso-scale, including the following steps: Based on the GIS model corresponding to the abnormal sub-region, divide its horizontal spatial region at a second resolution to obtain several second-level sub-regions, where the second resolution is higher than the first resolution; Deploy a second-level data acquisition module in each second-level sub-region to respectively collect at least one type of second-level target data in the corresponding sub-region.

8. The soil environment monitoring method based on multi-level sampling according to claim 7, characterized in that Based on the second-level data collected by the second-level data acquisition module, identify at least one abnormal point in each abnormal sub-region, including the following steps: According to the second-level target data and its corresponding anomaly threshold, mark abnormal second-level sub-regions from the several second-level sub-regions, and set the center position of each abnormal second-level sub-region as an abnormal point.

9. The soil environment monitoring method based on multi-level sampling according to claim 2, characterized in that Based on the GIS model, deploy a third-level data acquisition module at each abnormal point from a micro-scale, including the following steps: Based on the GIS model corresponding to the abnormal point, divide its vertical spatial region at a third resolution to obtain several soil sampling points, and at each soil sampling point, use the third-level data acquisition module to collect soil samples, where the soil samples are used to evaluate the soil condition.

10. A soil environment monitoring system based on multi-level sampling, characterized in that, It includes a processor and a memory, and a computer program is stored in the memory. When the computer program is executed by the processor, it implements the soil environment monitoring method based on multi-level sampling as described in any one of claims 1-9.

Images