A water quality monitoring method and system based on the impact of water source risk sources
Through the combination of multispectral remote sensing images and geospatial information, the risk degree coefficient is calculated, and the problems of low automation degree of water quality monitoring methods in existing water sources are solved, and more accurate and comprehensive water quality monitoring is achieved.
Patent Information
- Application Number
- CN202410888378.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-04
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-07-04
AI Technical Summary
The existing water quality monitoring methods in water sources have low degree of automation, low efficiency, high cost, and difficult to achieve comprehensive, real-time and continuous monitoring. The impact of risk sources around the water sources on water quality has not been fully considered, resulting in low monitoring accuracy and reliability.
Multi-spectral remote sensing images are used for spatial downscale processing, combined with geospatial information to obtain risk source information, calculate the risk degree coefficient of grid units, and monitor according to the preset water quality parameter monitoring model, and divide the water source area for more accurate monitoring Water quality parameters.
It improves the reliability and accuracy of water quality monitoring in water sources, achieves more comprehensive water quality monitoring, and can more effectively consider the impact of risk sources on water quality in water sources and surrounding areas.
Smart Images

Figure CN118518602B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of water quality monitoring at water sources, and in particular to a water quality monitoring method and system based on the influence of risk sources at water sources. Background Art
[0002] In the field of water quality monitoring at water sources, traditional monitoring methods mainly rely on manual carrying of portable instruments to the site for single-point sampling and measurement, or continuous data collection through fixed sites. Although these methods can provide a certain degree of water quality information, they have problems such as low automation, low efficiency, high cost, and long data update cycle. In addition, due to manpower limitations, traditional methods are difficult to achieve comprehensive, real-time and continuous monitoring of water sources.
[0003] In recent years, with the rapid development of remote sensing technology, methods for monitoring water quality parameters in combination with remote sensing images have gradually attracted attention. However, most of the existing water quality monitoring methods based on machine learning of remote sensing images focus on algorithm optimization and model construction, and rarely consider the potential impact of risk sources around water sources on water quality parameters. These risk sources include, but are not limited to, sewage outlets, enterprise land, agricultural non-point sources, etc. The sewage they discharge often has a direct or indirect impact on the water quality of the water source. In particular, the closer the water source is to these risk sources, the higher the possibility that its water quality parameters will be polluted. Therefore, when conducting water quality monitoring, the existing water quality monitoring methods do not fully consider the impact of these risk sources, resulting in low monitoring accuracy and reliability. Summary of the invention
[0004] In response to the problems in the prior art, the present application provides a water quality parameter monitoring method that fully considers the risk source information of water sources and can monitor water quality more accurately and comprehensively by combining multispectral remote sensing images with high spatial resolution, wide coverage and short update cycle.
[0005] A water quality monitoring method based on the impact of water source risk sources includes the following steps:
[0006] Acquire a first multispectral remote sensing image of a target water source area;
[0007] Performing spatial downscaling processing on the spectral bands in the first multispectral remote sensing image to obtain a first high-resolution multispectral remote sensing image;
[0008] Obtaining geospatial information of the target water source and its surrounding areas, and obtaining information on a number of risk sources based on the first high-resolution multispectral remote sensing image and the geospatial information;
[0009] Obtain a vector boundary layer of the target water source, and perform grid processing to obtain a plurality of grid units, and calculate the risk degree coefficient of each grid unit affected by the risk source according to the risk source information;
[0010] The water quality parameters of the grid unit are obtained according to the first high-resolution multispectral remote sensing image and the risk degree coefficient of the grid unit in accordance with a preset water quality parameter monitoring model.
[0011] The present application also provides a water quality monitoring system based on the impact of water source risk sources, including:
[0012] Multispectral remote sensing image acquisition module: used to acquire a first multispectral remote sensing image of a target water source;
[0013] A multispectral remote sensing image processing module: used for performing spatial downscaling processing on the spectral bands in the first multispectral remote sensing image to obtain a first high-resolution multispectral remote sensing image;
[0014] Risk source acquisition module: used to acquire geographic space information of the target water source and its surrounding areas, and acquire some risk source information according to the first high-resolution multi-spectral remote sensing image and the geographic space information;
[0015] Risk degree coefficient acquisition module: used to obtain the vector boundary layer of the target water source, and perform grid processing to obtain a number of grid units, and calculate the risk degree coefficient of each grid unit affected by the risk source according to the risk source information;
[0016] Water quality parameter distribution result acquisition module: used to obtain the water quality parameters of the grid unit according to the first high-resolution multispectral remote sensing image and the risk degree coefficient of the grid unit and according to a preset water quality parameter monitoring model.
[0017] Compared with the existing technology, this solution monitors the water quality of water sources based on multispectral remote sensing images, divides the multispectral remote sensing images of water sources into several grid units, fully analyzes the risk level of risk sources in the water source and its surrounding areas to each grid unit, and then uses the preset water quality parameter monitoring model to monitor the water quality of water sources more accurately and comprehensively, effectively improving the reliability of water quality monitoring of water sources.
[0018] In order to provide a clearer understanding of the present application, the specific implementation of the present application will be described below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] Figure 1 A flow chart of a water quality monitoring method based on the impact of water source risk sources is provided for this application;
[0020] Figure 2A flow chart of a method for obtaining risk source information of a water quality monitoring method based on the impact of risk sources in water sources in this application;
[0021] Figure 3 A flow chart of a method for calculating a risk degree coefficient of a water quality monitoring method based on the impact of risk sources in a water source area according to the present application;
[0022] Figure 4 A flow chart of a method for constructing a water quality parameter monitoring model for a water quality monitoring method based on the impact of water source risk sources in this application;
[0023] Figure 5 This is a flow chart of a method for preprocessing water quality monitoring data based on a water quality monitoring method influenced by water source risk sources in this application;
[0024] Figure 6 This is a schematic diagram of a water quality monitoring system based on the impact of water source risk sources in this application. DETAILED DESCRIPTION
[0025] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.
[0026] It should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of the application. It should be understood that the operations of the flowcharts may be implemented out of order, and steps without logical contextual relationships may be reversed in order or implemented simultaneously. In addition, those skilled in the art may add one or more other operations to the flowcharts, or may remove one or more operations from the flowcharts, under the guidance of the content of this application.
[0027] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0028] Example 1
[0029] See also Figure 1 , Figure 1This is a flow chart of a water quality monitoring method based on the impact of water source risk sources in this application. This application provides a water quality monitoring method based on the impact of water source risk sources, specifically including the following steps:
[0030] S1: Acquire the first multispectral remote sensing image of the target water source;
[0031] S2: performing spatial downscaling processing on the spectral bands in the first multispectral remote sensing image to obtain a first high-resolution multispectral remote sensing image;
[0032] S3: Obtaining geographic spatial information of the target water source and its surrounding areas, and obtaining information on a number of risk sources based on the first high-resolution multispectral remote sensing image and the geographic spatial information;
[0033] S4: obtaining a vector boundary layer of the target water source, and performing grid processing to obtain a plurality of grid units, and calculating the risk degree coefficient of each grid unit affected by the risk source according to the risk source information;
[0034] S5: According to the first high-resolution multispectral remote sensing image and the risk degree coefficient of the grid unit, the water quality parameter of the grid unit is obtained according to a preset water quality parameter monitoring model.
[0035] Compared with the existing technology, this solution monitors the water quality of water sources based on multispectral remote sensing images with high spatial resolution, wide coverage and short update cycle. The multispectral remote sensing images of water sources are divided into several grid units, and the risk level of multiple risk sources in the water source and its surrounding areas to each grid unit is fully analyzed. Then, according to the preset water quality parameter monitoring model, the water quality of water sources can be monitored more accurately and comprehensively, effectively improving the reliability of water quality monitoring of water sources.
[0036] For step S1: obtain a first multispectral remote sensing image of a target water source.
[0037] In this embodiment, the staff can download the first multispectral remote sensing image covering the target water source in the same time range on the Digital Earth open platform, and the first multispectral remote sensing image records the response characteristics of the water quality parameters of the target water source to multiple spectral bands, including: b1, b2, b3, b4, b5, b6, b7, b8, b8A, b11 and b12 bands. By analyzing the spectral reflectance in different bands, the water quality of the target water source can be reflected. The preprocessing of the multispectral remote sensing image is specifically: performing atmospheric correction and radiation correction on the multispectral image.
[0038] Since the atmosphere will interfere with the multi-spectral image and other remote sensing effects, these interferences are mainly manifested as scattering, absorption, etc., it is necessary to perform atmospheric correction on the multi-spectral remote sensing image. Atmospheric absorption and scattering can be simulated by the atmospheric model method, so as to calculate physical quantities such as energy transfer and emissivity, and provide accurate parameters for atmospheric correction.
[0039] The radiation correction of the multispectral remote sensing image of the UAV can eliminate or correct the image radiation distortion caused by factors such as the multispectral sensor in the UAV, atmospheric conditions and solar altitude angle, so that the pixel value on the multispectral remote sensing image of the UAV can more accurately reflect the reflectivity or radiation brightness of the actual ground object. The radiation correction method includes: empirical model method, atmospheric correction method and surface reflection correction method.
[0040] Of course, in other embodiments, multispectral images can be downloaded from other platforms such as the USGS official website; the preprocessing operations of the multispectral remote sensing images can also be appropriately adjusted, such as performing preprocessing operations such as geometric correction and image mosaicking on the multispectral remote sensing images.
[0041] For step S2: spatial downscaling is performed on the spectral bands in the first multispectral remote sensing image to obtain a first high-resolution multispectral remote sensing image.
[0042] In this embodiment, the spatial resolution of the spectral band of the first multi-spectral remote sensing image is 20 m, and the spatial resolution of the spectral band of the first high-resolution multi-spectral remote sensing image is 10 m.
[0043] The spectral bands in the first multispectral remote sensing image are spatially downscaled to obtain a first high-resolution multispectral remote sensing image, and the method further includes the following steps: using the surface-to-point Kriging interpolation method to spatially downscale the b5, b6, b7, b8A, b11 and b12 bands with a spatial resolution of 20m in the first multispectral remote sensing image, thereby obtaining the first high-resolution multispectral remote sensing image with a spatial resolution of 10m for the b5, b6, b7, b8A, b11 and b12 bands. Of course, the conversion of multispectral remote sensing images from low spatial resolution to high spatial resolution can also be achieved through data fusion techniques such as PANSHARPENING, filter resampling, etc.
[0044] For step S3: obtaining the geospatial information of the target water source and its surrounding areas, obtaining some risk source information based on the first high-resolution multispectral remote sensing image and the geospatial information.
[0045] Among them, in this embodiment, the risk source information includes: risk source type, risk source location information and risk source number. The risk source types include: direct risk source, fixed risk source and agricultural non-point source pollution. The direct risk source includes: sewage outlets such as domestic sewage and industrial sewage directly harm the water quality of the water source; the fixed risk source includes: the enterprise land of high-pollution industries such as petrochemicals, medicine, printing and dyeing, papermaking, fermentation, etc. directly or indirectly pollutes the water source through wastewater discharge, waste gas emissions, improper solid waste treatment, etc.; the agricultural non-point source pollution includes: pollutants generated in the agricultural production process, such as fertilizers, pesticides, livestock and poultry breeding waste, etc., enter the water body through surface runoff, soil erosion, etc., causing pollution to the water source.
[0046] The geospatial information refers to information on the target water source and its surrounding areas in public data platforms such as local departments or environmental monitoring agencies, including: population distribution information, industrial layout information, agricultural activity information and historical pollution information.
[0047] See also Figure 2 , Figure 2 This is a flow chart of a method for obtaining risk source information based on a water quality monitoring method for water source risk sources in this application. The method of obtaining a number of risk source information based on the high-resolution multispectral remote sensing image and the geographic space information also includes the following steps:
[0048] S31: using geographic information system software, utilizing the spectral features of the high-resolution multispectral remote sensing image, and extracting the location information of the risk source through image segmentation, classification and target detection;
[0049] S32: Integrate the extracted risk source location information with the geographic space information to form the risk source information;
[0050] S33: Verify the accuracy and reliability of the risk source information through on-site investigation, comparative analysis and data verification.
[0051] The geographic information system software is ArcGIS. Of course, in other embodiments, other types of geographic information system software may also be used. The risk source types may also be adaptively modified, for example, the risk source types also include urban non-point source pollution and mobile source pollution.
[0052] For step S4: obtain the vector boundary layer of the target water source, and perform grid processing to obtain a number of grid units, and calculate the risk degree coefficient of each grid unit affected by the risk source according to the risk source information.
[0053] The vector boundary layer of the target water source refers to a spatial data of the boundary of the target water source protection area and its related attributes, which is composed of the vector data of the target water source.
[0054] The method of obtaining the vector boundary layer of the target water source and performing gridding processing to obtain a plurality of grid units also includes: obtaining the vector boundary layer of the target water source by importing the vector data of the target water source into a desktop geographic information system software, using the "Create Grid" tool of the desktop geographic information system software, and setting the grid type, grid range, horizontal interval and vertical interval to create the grid unit.
[0055] In this embodiment, the desktop geographic information system software is QGIS software. The vector data of the target water source can be downloaded from the vector data about the water source released by the environmental protection department or the water resources management department. The grid unit is a square pixel with a resolution of 10m, so the grid type is set to a rectangle, and the grid range is set to ensure that the grid unit can cover the entire target water source, and the horizontal interval and vertical interval are both set to 10. Of course, in other embodiments, the grid type, grid range, horizontal interval and vertical interval can be adaptively adjusted according to the actual needs of creating grid units; other desktop geographic information system software such as ArcGIS can also be used to create the grid unit.
[0056] See also Figure 3 , Figure 3 This is a flow chart of a method for calculating a risk degree coefficient of a water quality monitoring method based on the influence of risk sources in a water source area. The calculation of the risk degree coefficient of each grid unit affected by the risk source also includes the following steps:
[0057] S41: Calculate the distance d between the grid unit and any of the risk sources according to the risk source location information. i,j , where 1≦i≦b, 1≦j≦r, b is the number of grid units, and r is the number of risk sources;
[0058] S42: performing inverse processing on the distance between the grid unit and any of the risk sources to obtain a risk degree coefficient of the risk source to the grid unit;
[0059] S43: According to the risk degree coefficient of the risk source to the grid unit, the risk degree coefficients of the grid unit affected by all the risk sources are summed to obtain the risk degree coefficient of the grid unit. The specific formula can be expressed as: , where 1≦i≦b, 1≦j≦r, b is the number of the grid units, and r is the number of types of the risk sources.
[0060] For step S5: According to the first high-resolution multispectral remote sensing image and the risk degree coefficient of the grid cell, the water quality parameters of the grid cell are obtained according to a preset water quality parameter monitoring model.
[0061] In this embodiment, it further includes: substituting the spectral reflectance of the b1, b2, b3, b4, b5, b6, b7, b8, b8A, b11, and b12 bands of the first high-resolution multispectral remote sensing image and the risk degree coefficient of the grid cell into the water quality parameter monitoring model to obtain the water quality parameters of the grid cell.
[0062] The expression of the water quality parameter monitoring model is:
[0063]
[0064] where s i is the water quality observation point of the i-th target water source, and w k (s i ) represents the weight of the k-th independent variable under the water quality observation point of the i-th target water source; x ik is the value of the k-th independent variable under the water quality observation point of the i-th target water source; is the regression coefficient under the k-th independent variable; the independent variables include: the spectral reflectance of the b1, b2, b3, b4, b5, b6, b7, b8, b8A, b11, and b12 bands of the first high-resolution multispectral remote sensing image, and the risk degree coefficient of the grid cell; p is the number of independent variables; 1 < i < y, where y represents the number of water quality observation points of the target water source; 0 < l < b, where b is the number of grid cells.
[0065] Use the spatial weighted neural network model to calculate the weight w k (s i ) of the k-th independent variable under the water quality observation point of the i-th target water source:
[0066]
[0067] where is the distance from the water quality observation point of the i-th target water source to the m-th grid cell, 1 ≤ m ≤ b, where b is the number of grid cells; SWNN is the spatial weighted neural network model, including a model with n-layer network structure, including: an input layer, an output layer, and n - 2 hidden layers.
[0068] In this embodiment, SWNN is a model of a 4-layer network structure, including an input layer, an output layer and two hidden layers. The input layer is the spatial distance vector between the water quality parameter observation point and all grid points. Before input, the spatial distance vector can be standardized or normalized. Both hidden layers contain a number of neurons, and the neurons of the input layer, output layer and hidden layer are connected in a fully connected manner, that is, each neuron receives the output of all neurons in the previous layer as input. After the neuron obtains the output of all neurons in the previous layer, it performs weighted summation, and the result of the weighted summation is obtained by the activation function to obtain the output of the neuron. In this embodiment, the activation function adopts the PReLU function, which allows negative input values to have a small positive gradient, thereby helping to solve the problem of ReLU gradient disappearance when negative input. And the Dropout technology is used to enhance the generalization ability of the model, and a part of the neurons in the network are randomly "turned off" during the training process, that is, their output is set to 0, which helps to prevent overfitting and improve the performance of the model on unseen data. The output layer is the spatial weight of each of the independent variables.
[0069] Of course, in other embodiments, the number of layers of the SWNN network structure may be adaptively adjusted, and other activation functions such as Sigmoid may be used.
[0070] See also Figure 4 , Figure 4 This is a flow chart of a method for constructing a water quality parameter monitoring model based on a water quality monitoring method for water source risk sources in this application. Constructing the water quality parameter monitoring model also includes the following steps:
[0071] S51: Obtaining water quality monitoring data of the target water source from the water quality observation point of the target water source within any time range;
[0072] S52: Acquire a second multispectral remote sensing image of the target water source within the time range;
[0073] S53: performing spatial downscaling on the spectral bands in the second multispectral remote sensing image to obtain a second high-resolution multispectral remote sensing image;
[0074] S54: Calculating the water quality monitoring data of each of the grid cells according to the water quality monitoring data of the target water source;
[0075] S55: According to the water quality monitoring data under the grid unit, the second high-resolution multispectral remote sensing image and the risk degree coefficient of the grid unit, the regression coefficient is obtained by the least square method according to the expression of the water quality parameter monitoring model. .
[0076] Wherein, for step S51: obtaining water quality monitoring data of the target water source from the water quality observation point of the target water source within any time range.
[0077] See also Figure 5 , Figure 5 This is a flow chart of a method for preprocessing water quality monitoring data of a water quality monitoring method based on the influence of water source risk sources in this application. In this embodiment, the staff can obtain the water quality monitoring data of the water source released in the past year or five years on the official public data platform of the environmental protection department, and the water quality monitoring data includes at least: the concentration of total nitrogen and total phosphorus. Taking the 12-month water quality monitoring data of the Yellow River Basin from multiple water quality observation points in the past year as an example, the preprocessing of the water quality monitoring data also includes the following steps:
[0078] S511: Calculate the average value of the water quality monitoring data according to the following formula:
[0079]
[0080] in, is the average value of the water quality monitoring data at the above several time points, t is the number of time points, Total i is the water quality monitoring data at the i-th time point. In this embodiment, t=12, Total i is the water quality monitoring data for the ith month;
[0081] S512: Calculate the standard deviation of the water quality monitoring data according to the following formula based on the average value of the water quality monitoring data
[0082]
[0083] Among them, SD Total is the standard deviation of the water quality monitoring data;
[0084] S513: According to the average value of the water quality monitoring data and the standard deviation of the water quality monitoring data, determine the upper limit value of the water quality monitoring data and the lower limit value of the water quality monitoring data according to the following formula, and eliminate the water quality monitoring data that exceeds the upper limit value of the water quality monitoring data and the lower limit value of the water quality monitoring data
[0085]
[0086] Among them, UpperLimit Total is the upper limit of the water quality monitoring data, LowerLimit Total It is the lower limit value of the water quality monitoring data.
[0087] Of course, in other embodiments, water quality monitoring data of other specific time periods can be obtained from other platforms such as scientific research institutions and project databases, and the establishment of time points can be adaptively adjusted according to actual needs; the water quality monitoring data can also include other water quality parameters such as lead and mercury concentrations.
[0088] For step S52 and step S53, the principles of preprocessing the second multispectral remote sensing image and spatially downscaling the band information of the second multispectral remote sensing image are the same as those of preprocessing the first multispectral remote sensing image and spatially downscaling the band information of the first multispectral remote sensing image, and will not be repeated here.
[0089] Example 2
[0090] See also Figure 6 , Figure 6 This is a schematic diagram of a water quality monitoring system based on the impact of water source risk sources in this application. This application also provides a water quality monitoring system based on the impact of water source risk sources, including:
[0091] Multispectral remote sensing image acquisition module: used to acquire a first multispectral remote sensing image of a target water source;
[0092] A multispectral remote sensing image processing module: used for performing spatial downscaling processing on the spectral bands in the first multispectral remote sensing image to obtain a first high-resolution multispectral remote sensing image;
[0093] Risk source acquisition module: used to acquire geographic space information of the target water source and its surrounding areas, and acquire some risk source information according to the first high-resolution multi-spectral remote sensing image and the geographic space information;
[0094] Risk degree coefficient acquisition module: used to obtain the vector boundary layer of the target water source, and perform grid processing to obtain a number of grid units, and calculate the risk degree coefficient of each grid unit affected by the risk source according to the risk source information;
[0095] Water quality parameter distribution result acquisition module: used to obtain the water quality parameters of the grid unit according to the first high-resolution multispectral remote sensing image and the risk degree coefficient of the grid unit and according to a preset water quality parameter monitoring model.
[0096] It should be noted that the water quality monitoring system based on the influence of water source risk sources provided by the present application stores the data obtained when implementing a water quality monitoring method based on the influence of water source risk sources in a one-to-one correspondence in the storage of the system. When relevant calculations are required, the data required for the calculation can be directly obtained from the storage.
[0097] It should also be noted that the water quality monitoring system based on the influence of water source risk sources provided in the above embodiment only uses the division of the above-mentioned functional modules as an example when implementing a water quality monitoring method based on the influence of water source risk sources. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the equipment is divided into different functional modules to complete all or part of the functions described above. In addition, the water quality monitoring system based on the influence of water source risk sources provided in the above embodiment and the water quality monitoring method based on the influence of water source risk sources in Example 1 belong to the same concept. The implementation process is detailed in the method embodiment, which will not be repeated here.
[0098] The present application is not limited to the above-mentioned implementation modes. If various changes or modifications to the present application do not depart from the spirit and scope of the present application, and if these changes and modifications fall within the claims of the present application and the scope of equivalent technologies, the present application is also intended to include these changes and modifications.
Claims
1. A water quality monitoring method based on the impact of water source risk sources, characterized in that: Includes the following steps: Acquire a first multispectral remote sensing image of a target water source area; Performing spatial downscaling processing on the spectral bands in the first multispectral remote sensing image to obtain a first high-resolution multispectral remote sensing image; Obtaining geospatial information of the target water source and its surrounding areas, and obtaining information on a number of risk sources based on the first high-resolution multispectral remote sensing image and the geospatial information; Wherein, the risk source information includes: risk source location information; The vector boundary layer of the target water source is obtained, and grid processing is performed to obtain a plurality of grid units. According to the risk source information, the risk degree coefficient of each grid unit affected by the risk source is calculated, and the following steps are also included: Calculating the distance between the grid unit and any of the risk sources according to the risk source location information; Performing inverse processing on the distance between the grid unit and any of the risk sources to obtain a risk degree coefficient of the risk source to the grid unit; According to the risk degree coefficient of the risk source to the grid unit, the risk degree coefficients of the grid unit affected by all the risk sources are summed to obtain the risk degree coefficient of the grid unit; According to the first high-resolution multispectral remote sensing image and the risk degree coefficient of the grid unit, the water quality parameters of the grid unit are obtained according to a preset water quality parameter monitoring model, including: substituting the spectral reflectance of the b1, b2, b3, b4, b5, b6, b7, b8, b8A, b11 and b12 bands of the first high-resolution multispectral remote sensing image and the risk degree coefficient of the grid unit into the water quality parameter monitoring model to obtain the water quality parameters of the grid unit, and the expression of the water quality parameter monitoring model is: where s i is the water quality observation point of the i-th target water source, and w k (s i ) represents the weight of the k-th independent variable under the water quality observation point of the i-th target water source; x ik is the value of the k-th independent variable under the water quality observation point of the i-th target water source; is the regression coefficient under the k-th independent variable; the independent variables include: the spectral reflectance of the b1, b2, b3, b4, b5, b6, b7, b8, b8A, b11, and b12 bands of the first high-resolution multispectral remote sensing image, and the risk degree coefficient of the grid cell; p is the number of independent variables; 1 < i < y, where y represents the number of water quality observation points of the target water source; 0 < l < b, where b is the number of grid cells; among them, the weight of the k-th independent variable under the water quality observation point of the i-th target water source is calculated using a spatial weighted neural network model: in, is the distance from the water quality observation point of the i-th target water source to the m-th grid unit, 1≦m≦b, b is the number of the grid units; SWNN is a spatial weighted neural network model.
2. A water quality monitoring method based on the impact of water source risk sources according to claim 1, characterized in that: Constructing the water quality parameter monitoring model also includes the following steps: Obtain water quality monitoring data of the target water source from the water quality observation point of the target water source within any time range; Acquire a second multispectral remote sensing image of the target water source within the time range; Performing spatial downscaling on the spectral bands in the second multispectral remote sensing image to obtain a second high-resolution multispectral remote sensing image; Calculating the water quality monitoring data of each of the grid cells according to the water quality monitoring data of the target water source; The regression coefficient is obtained by the least squares method according to the water quality monitoring data under the grid unit, the second high-resolution multispectral remote sensing image and the risk degree coefficient of the grid unit and in accordance with the expression of the water quality parameter monitoring model.
3. A water quality monitoring method based on the impact of water source risk sources according to claim 2, characterized in that: The risk source information also includes: risk source type; wherein the risk source type includes: direct risk source, fixed risk source and agricultural non-point source pollution.
4. A water quality monitoring method based on the impact of water source risk sources according to claim 2, characterized in that: The water quality monitoring data at least include: the concentration of total nitrogen and total phosphorus; The step of preprocessing the water quality monitoring data includes: Calculating the average value of the water quality monitoring data; Calculate the standard deviation of the water quality monitoring data according to the average value of the water quality monitoring data; According to the average value of the water quality monitoring data and the standard deviation of the water quality monitoring data, the upper limit value of the water quality monitoring data and the lower limit value of the water quality monitoring data are determined, and the water quality monitoring data exceeding the upper limit value of the water quality monitoring data and the lower limit value of the water quality monitoring data are eliminated.
5. The water quality monitoring method based on the impact of water source risk sources according to claim 1 is characterized in that: The step of preprocessing the first multispectral remote sensing image includes: performing atmospheric correction and radiation correction on the first multispectral remote sensing image.
6. A water quality monitoring system based on the impact of water source risk sources, characterized in that: include: Multispectral remote sensing image acquisition module: used to acquire a first multispectral remote sensing image of a target water source; A multispectral remote sensing image processing module: used for performing spatial downscaling processing on the spectral bands in the first multispectral remote sensing image to obtain a first high-resolution multispectral remote sensing image; Risk source acquisition module: used to acquire geographic space information of the target water source and its surrounding areas, and acquire some risk source information according to the first high-resolution multi-spectral remote sensing image and the geographic space information; Wherein, the risk source information includes: risk source location information; The risk degree coefficient acquisition module is used to obtain the vector boundary layer of the target water source, and perform grid processing to obtain a plurality of grid units, and calculate the risk degree coefficient of each grid unit affected by the risk source according to the risk source information, and also includes the following steps: Calculating the distance between the grid unit and any of the risk sources according to the risk source location information; Performing inverse processing on the distance between the grid unit and any of the risk sources to obtain a risk degree coefficient of the risk source to the grid unit; According to the risk degree coefficient of the risk source to the grid unit, the risk degree coefficients of the grid unit affected by all the risk sources are summed to obtain the risk degree coefficient of the grid unit; The water quality parameter distribution result acquisition module is used to obtain the water quality parameters of the grid unit according to the first high-resolution multispectral remote sensing image and the risk degree coefficient of the grid unit according to the preset water quality parameter monitoring model, including: substituting the spectral reflectance of the b1, b2, b3, b4, b5, b6, b7, b8, b8A, b11 and b12 bands of the first high-resolution multispectral remote sensing image and the risk degree coefficient of the grid unit into the water quality parameter monitoring model to obtain the water quality parameters of the grid unit. The expression of the water quality parameter monitoring model is: where s i is the water quality observation point of the i-th target water source, and w k (s i ) represents the weight of the k-th independent variable under the water quality observation point of the i-th target water source; x ik is the value of the k-th independent variable under the water quality observation point of the i-th target water source; is the regression coefficient under the k-th independent variable; the independent variables include: the spectral reflectance of the b1, b2, b3, b4, b5, b6, b7, b8, b8A, b11, and b12 bands of the first high-resolution multispectral remote sensing image, and the risk degree coefficient of the grid cell; p is the number of independent variables; 1 < i < y, where y represents the number of water quality observation points of the target water source; 0 < l < b, where b is the number of grid cells; among them, the weight of the k-th independent variable under the water quality observation point of the i-th target water source is calculated using a spatial weighted neural network model: in, is the distance from the water quality observation point of the i-th target water source to the m-th grid unit, 1≦m≦b, b is the number of the grid units; SWNN is a spatial weighted neural network model.
