Lake eutrophication evaluation method, device and equipment and storage medium

Abstract

The invention provides a lake eutrophication evaluation method, device and equipment and a storage medium, and relates to the technical field of environment evaluation, and the method comprises the steps: obtaining a first hyperspectral satellite image of a first lake; performing water quality parameter inversion on the first hyperspectral satellite image by using a preset total nitrogen inversion model, a preset total phosphorus inversion model and a preset chlorophyll a inversion model to obtain a concentration distribution diagram of each water quality parameter, the water quality parameters including total nitrogen, total phosphorus and chlorophyll a; and generating a eutrophication state distribution diagram of the first lake according to the concentration distribution diagram of each water quality parameter. According to the invention, field sampling is not needed, and lake eutrophication monitoring efficiency can be improved; moreover, water quality parameter inversion is carried out on the first hyperspectral satellite image of the lake by using the inversion model, so that the real water quality condition of each region of the lake can be comprehensively and meticulously obtained, and the lake eutrophication monitoring precision can be further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental assessment, and in particular to a lake eutrophication assessment method, device, equipment and storage medium. Background Art

[0002] Lake eutrophication refers to the phenomenon that excessive accumulation of nutrients such as total nitrogen (TN) and total phosphorus (TP) in lakes leads to abnormal overgrowth of algae and other plankton, a decrease in dissolved oxygen in the water, deterioration of water quality, and mass death of fish and other organisms. Lake eutrophication seriously affects the ecological function and water quality safety of lakes, so it is necessary to conduct lake eutrophication assessment.

[0003] At present, the evaluation of lake eutrophication mainly relies on field sampling. Specifically, the staff needs to set up multiple sampling points in different areas of the lake, collect water samples and bring them back to the laboratory, and use professional chemical analysis methods to measure the concentrations of total nitrogen, total phosphorus and chlorophyll a, and evaluate the degree of lake eutrophication based on the measurement results.

[0004] However, since field sampling requires a lot of manpower, time and energy, the existing technology has the problem of low efficiency in lake eutrophication monitoring. In addition, due to the limitations of manpower, material resources and time in the actual sampling process, the number of sampling points is relatively limited, making it difficult to comprehensively and meticulously reflect the true water quality conditions in various areas of the lake, resulting in the existing technology having the problem of low accuracy in lake eutrophication monitoring. Summary of the invention

[0005] In view of this, a lake eutrophication evaluation method, device, equipment and storage medium are provided to solve the problem of low efficiency and accuracy of lake eutrophication monitoring in the prior art.

[0006] The present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides a lake eutrophication assessment method, comprising:

[0008] Acquire the first hyperspectral satellite image of the first lake;

[0009] Using a preset total nitrogen inversion model, a preset total phosphorus inversion model, and a preset chlorophyll a inversion model, respectively, the first hyperspectral satellite image is used to invert water quality parameters to obtain a concentration distribution map of each water quality parameter, wherein the water quality parameters include total nitrogen, total phosphorus, and chlorophyll a;

[0010] A eutrophication status distribution map of the first lake is generated according to the concentration distribution map of each of the water quality parameters.

[0011] Optionally, the construction process of any one of the preset total nitrogen inversion model, the preset total phosphorus inversion model and the preset chlorophyll a inversion model includes:

[0012] Obtaining the geographic coordinates of a plurality of sampling points of the second lake and the actual concentration of a target water quality parameter at the sampling time, wherein the target water quality parameter is total nitrogen, total phosphorus or chlorophyll a;

[0013] Acquire a second hyperspectral satellite image of the second lake at the sampling time;

[0014] preprocessing the second hyperspectral satellite image;

[0015] Based on the geographic coordinates, extracting the spectral reflectance value of the corresponding position in the preprocessed second hyperspectral satellite image to obtain multi-band spectral information;

[0016] Performing standardization processing on the multi-band spectral information;

[0017] Based on a preset regression algorithm, sensitive bands are screened for the standardized multi-band spectral information to obtain sensitive bands of the target water quality parameters;

[0018] Performing band combination screening on the sensitive bands of the target water quality parameter to obtain the optimal band combination of the target water quality parameter;

[0019] A target inversion model is constructed by using a machine learning algorithm according to the actual concentration of the target water quality parameter and the optimal band combination. The target inversion model is the preset total nitrogen inversion model, the preset total phosphorus inversion model or the preset chlorophyll a inversion model.

[0020] Optionally, preprocessing the second hyperspectral satellite image specifically includes:

[0021] The second hyperspectral satellite image is subjected to radiation calibration processing, atmospheric correction processing and geographic location correction processing.

[0022] Optionally, the preset regression algorithm is a LASSO regression algorithm.

[0023] Optionally, performing band combination screening on the sensitive bands of the target water quality parameter to obtain the optimal band combination of the target water quality parameter specifically includes:

[0024] Based on the ant colony optimization algorithm, the sensitive bands of the target water quality parameters are screened for band combination to obtain the optimal band combination of the target water quality parameters.

[0025] Optionally, the machine learning algorithm includes a linear regression algorithm and a polynomial regression algorithm.

[0026] Optionally, generating a eutrophication state distribution map of the first lake according to the concentration distribution map of each of the water quality parameters specifically includes:

[0027] Calculating the comprehensive nutrient status index of each area of ​​the first lake according to the concentration distribution map of each water quality parameter;

[0028] According to the comprehensive nutrient status index of each area of ​​the first lake, the eutrophication level of each area of ​​the first lake is determined to obtain the eutrophication status distribution map.

[0029] In a second aspect, the present invention provides a lake eutrophication assessment device, comprising:

[0030] An acquisition module, used for acquiring a first hyperspectral satellite image of a first lake;

[0031] an inversion module, for performing water quality parameter inversion on the first hyperspectral satellite image using a preset total nitrogen inversion model, a preset total phosphorus inversion model, and a preset chlorophyll a inversion model, respectively, to obtain a concentration distribution map of each water quality parameter, wherein the water quality parameters include total nitrogen, total phosphorus, and chlorophyll a;

[0032] A generation module is used to generate a eutrophication status distribution map of the first lake according to the concentration distribution map of each water quality parameter.

[0033] In a third aspect, the present invention provides a lake eutrophication assessment device, comprising:

[0034] at least one processor; and,

[0035] a memory communicatively connected to the at least one processor; wherein,

[0036] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can implement each step of the lake eutrophication assessment method as described above.

[0037] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the various steps in the lake eutrophication assessment method as described above are implemented.

[0038] The present invention adopts the above technical scheme, a lake eutrophication evaluation method, including: obtaining a first hyperspectral satellite image of a first lake; using a preset total nitrogen inversion model, a preset total phosphorus inversion model and a preset chlorophyll a inversion model respectively, to invert water quality parameters of the first hyperspectral satellite image, and obtain concentration distribution maps of various water quality parameters, the water quality parameters include total nitrogen, total phosphorus and chlorophyll a; according to the concentration distribution maps of various water quality parameters, a eutrophication state distribution map of the first lake is generated. Based on this, the present invention does not require field sampling, can improve the efficiency of lake eutrophication monitoring, and can be applied to water quality monitoring of large-scale lakes; and, because hyperspectral remote sensing technology has a detailed spectral resolution ability, it can identify the spectral response characteristics of different water quality parameters in different bands. Therefore, the present invention uses the inversion model to invert water quality parameters of the first hyperspectral satellite image of the lake, so that the present invention can comprehensively and meticulously obtain the true water quality status of each area of ​​the lake, thereby improving the accuracy of lake eutrophication monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 It is a flow chart of a lake eutrophication evaluation method provided by an embodiment of the present invention;

[0041] Figure 2 It is a structural schematic diagram of a lake eutrophication evaluation device provided by an embodiment of the present invention;

[0042] Figure 3 It is a structural schematic diagram of a lake eutrophication assessment device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0044] Lake eutrophication refers to the phenomenon that excessive accumulation of nutrients such as total nitrogen and total phosphorus in lakes leads to abnormal overgrowth of algae and other plankton, a decrease in dissolved oxygen in the water, deterioration of water quality, and mass death of fish and other organisms. Lake eutrophication seriously affects the ecological function and water quality safety of lakes, so it is necessary to conduct lake eutrophication assessment.

[0045] At present, the evaluation of lake eutrophication mainly relies on field sampling. Specifically, the staff needs to set up multiple sampling points in different areas of the lake, collect water samples and bring them back to the laboratory, and use professional chemical analysis methods to measure the concentrations of total nitrogen, total phosphorus and chlorophyll a, and evaluate the degree of lake eutrophication based on the measurement results.

[0046] However, since field sampling requires a lot of manpower, time and energy, the existing technology has the problem of low efficiency in lake eutrophication monitoring. In addition, due to the limitations of manpower, material resources and time in the actual sampling process, the number of sampling points is relatively limited, making it difficult to comprehensively and meticulously reflect the true water quality conditions in various areas of the lake, resulting in the existing technology having the problem of low accuracy in lake eutrophication monitoring.

[0047] Based on this, in order to improve the efficiency and accuracy of lake eutrophication monitoring, the present invention provides a lake eutrophication evaluation method, device, equipment and storage medium. The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings.

[0048] Figure 1 FIG. 1 is a flow chart of a lake eutrophication evaluation method provided by an embodiment of the present invention. Figure 1 As shown, this process includes:

[0049] Step 101: Acquire a first hyperspectral satellite image of a first lake.

[0050] Specifically, the present invention can obtain a first hyperspectral satellite image of a first lake through the Zhuhai-1 satellite.

[0051] Step 102: using a preset total nitrogen inversion model, a preset total phosphorus inversion model, and a preset chlorophyll a inversion model, respectively, water quality parameters are inverted for the first hyperspectral satellite image to obtain a concentration distribution map of each water quality parameter, where the water quality parameters include total nitrogen, total phosphorus, and chlorophyll a.

[0052] It should be noted that the construction process of the preset total nitrogen inversion model, the preset total phosphorus inversion model and the preset chlorophyll a inversion model is the same or similar. Specifically, the construction process of any one of them may include:

[0053] (1021) Obtain the geographic coordinates of multiple sampling points of the second lake and the actual concentration of the target water quality parameter at the sampling time, where the target water quality parameter is total nitrogen, total phosphorus or chlorophyll a.

[0054] In the actual application process, first, the staff set up multiple sampling points in the second lake area, collected water samples at each sampling point, measured the actual concentration of its target water quality parameters, and recorded the geographic coordinates of each sampling point through GPS equipment.

[0055] It should be noted that, when constructing the above three inversion models, in order to improve the construction efficiency, when measuring the target water quality parameters of the water samples collected at each sampling point, the actual concentrations of the three water quality parameters, total nitrogen, total phosphorus or chlorophyll a, can be measured simultaneously, and the actual concentrations of the target water quality parameters can be extracted for use in subsequent processes. Therefore, when constructing other inversion models among the above three inversion models, there is no need to repeatedly collect water samples at each sampling point and measure the target water quality parameters, but directly extract the measured results.

[0056] (1022) Obtain a second hyperspectral satellite image of the second lake at the sampling time.

[0057] Here, the second hyperspectral satellite image of the second lake at the sampling time can also be obtained through the Zhuhai-1 satellite, and the first lake and the second lake can be the same lake.

[0058] (1023) Preprocess the second hyperspectral satellite image.

[0059] When recording electromagnetic wave signals, sensors are subject to various interferences, including noise and nonlinear response. In order to eliminate the influence of these interferences, the second hyperspectral satellite image can be calibrated by radiometric calibration, and the pixel values ​​in the image can be converted into radiometric brightness values ​​with clear physical meaning, so that the remote sensing data can have absolute physical meaning, thereby improving the accuracy of the data.

[0060] Furthermore, the total radiation brightness of the ground target finally measured by the sensor is not a reflection of the true reflectivity of the surface, and it also includes the radiation error caused by atmospheric absorption, especially scattering. Atmospheric correction can eliminate these radiation errors caused by atmospheric influence and invert the true surface reflectivity of the ground object. Therefore, the second hyperspectral satellite image can also be atmospherically corrected to eliminate atmospheric interference in the image and obtain the spectral reflectance information of the real ground object.

[0061] Furthermore, the second hyperspectral satellite image can be geo-corrected to align the image with the recorded geographic coordinates, ensuring that the image data is accurately mapped to the corresponding ground location.

[0062] (1024) Based on the geographic coordinates of multiple sampling points of the second lake, spectral reflectance values ​​of corresponding positions are extracted from the preprocessed second hyperspectral satellite image to obtain multi-band spectral information of each sampling point of the second lake.

[0063] (1025) The following formula is used to standardize the multi-band spectral information to ensure the scale consistency between different bands:

[0064]

[0065] Among them, Z' is the spectral reflectance value after standardization; Z is the original spectral reflectance value; μ is the mean of all original spectral reflectance values; σ is the standard deviation of all original spectral reflectance values.

[0066] (1026) Based on a preset regression algorithm, the multi-band spectral information after standardization is screened for sensitive bands to obtain the sensitive bands of the target water quality parameters.

[0067] In an embodiment of the present invention, the preset regression algorithm may be a least absolute shrinkage and selection operator (LASSO) regression algorithm. The LASSO regression algorithm is a linear regression method with L1 regularization, and its main feature is that an L1 regularization term is added to the loss function, which can automatically reduce some irrelevant or less influential feature coefficients to zero, thereby playing a role in feature selection. This method is particularly suitable for sensitive band screening of high-dimensional data sets (such as hyperspectral band data), which helps to simplify the model and retain sensitive bands that have a greater impact on water quality parameters.

[0068] Based on the LASSO regression algorithm, the process of screening the sensitive bands of the standardized multi-band spectral information and obtaining the sensitive bands of the target water quality parameters may specifically include:

[0069] (10261)Data preparation.

[0070] Based on the standardized multi-band spectral information, the target water quality parameters and hyperspectral band reflectance values ​​of each sampling point of the second lake are organized into a data matrix. Specifically, let X be the band reflectance matrix, where each column represents a band and each row represents a sampling point; Y is the observed value vector of the water quality parameter.

[0071] Assuming there are n sampling points and m hyperspectral bands, the data matrix can be expressed as follows:

[0072]

[0073] Among them, x ijrepresents the spectral reflectance value of the i-th sampling point in the j-th hyperspectral band; y i Represents the actual concentration of the target water quality parameter at the i-th sampling point.

[0074] (10262) LASSO regression model establishment

[0075] In the LASSO regression model, the goal is to minimize the following loss function:

[0076]

[0077] Where n represents the number of samples, that is, the number of sampling points; y i represents the actual concentration of the target water quality parameter at the i-th sampling point; X i represents the spectral reflectance value vector of the i-th sampling point; β represents the regression coefficient vector; α represents the regularization intensity hyperparameter (controls the band screening intensity). Regularization term Used to constrain model complexity, a larger α value will cause more coefficients to be reduced to zero, thereby reducing the number of selected sensitive bands.

[0078] In the actual application process, first, set the value of α. Then, perform variable screening. Specifically, by drawing the coefficient path diagram of LASSO regression, you can intuitively see the changes in the coefficients of each variable under different α values. On this basis, you can use the cross-validation method to select the optimal α value.

[0079] (10263)Filter results

[0080] According to the optimal α value, we can determine which variables' coefficients are compressed to zero, thereby eliminating these variables from the model, and the remaining variables are the selected sensitive bands. In a specific example, assuming that the second hyperspectral satellite image is obtained using Zhuhai-1 satellite, the target water quality parameter is total nitrogen, and the sensitive bands of total nitrogen are:

[0081] B2 band (center wavelength is 466nm), B4 band (center wavelength is 500nm), B9 band (center wavelength is 580nm), B10 band (center wavelength is 596nm), and B12 band (center wavelength is 640nm).

[0082] In another specific example, assuming that the second hyperspectral satellite image is obtained using the Zhuhai-1 satellite, the target water quality parameter is total phosphorus, and the sensitive bands of total phosphorus are:

[0083] B2 band (center wavelength is 466nm), B10 band (center wavelength is 596nm), B15 band (center wavelength is 686nm), B22 band (center wavelength is 780nm), and B31 band (center wavelength is 926nm).

[0084] In another specific example, it is assumed that the second hyperspectral satellite image is obtained using the Zhuhai-1 satellite, the target water quality parameter is chlorophyll a, and the sensitive band of chlorophyll a is:

[0085] B21 band (center wavelength is 776nm), B22 band (center wavelength is 780nm), B24 band (center wavelength is 820nm), B29 band (center wavelength is 896nm), and B32 band (center wavelength is 940nm).

[0086] (1027) Perform band combination screening on the sensitive bands of the target water quality parameters to obtain the optimal band combination of the target water quality parameters.

[0087] In the embodiment of the present invention, the sensitive bands of the target water quality parameters are screened for band combination to obtain the optimal band combination of the target water quality parameters, which may specifically include:

[0088] Based on the Ant Colony Optimization (ACO) algorithm, the sensitive bands of the target water quality parameters are screened for band combination to obtain the optimal band combination of the target water quality parameters.

[0089] The basic idea of ​​the ant colony optimization algorithm is to use ants to leave pheromones on the path and select the path according to the pheromone concentration, so as to find the optimal or nearly optimal solution. Specifically, each ant releases pheromones during movement, and other ants decide their own movement direction based on the pheromone concentration. The higher the pheromone concentration, the more likely the path will be selected. At the same time, pheromones will evaporate over time, which allows new and potentially better paths to be explored.

[0090] In the process of band combination screening, ACO can be used to find the band combination with the strongest correlation with the target water quality parameter (i.e., target variable). The following are the detailed steps for band combination screening using ACO:

[0091] (10271) Initialize ant colony and pheromones.

[0092] Ant colony initialization: Divide the sensitive band combinations into multiple ant individuals, each ant represents a potential band combination scheme. Usually each combination can contain several bands to simulate the path of the ants.

[0093] Pheromone distribution initialization: Assign an initial pheromone concentration τ0 to each band, indicating the initial probability of the band being selected.

[0094] (10272)Ants select band combinations.

[0095] Each ant selects the next band based on the pheromone and heuristic information on the current band. Here, the heuristic information is the correlation between the band and the target water quality parameter. The probability that the Kth ant selects the Ith band is is defined as follows:

[0096]

[0097] Among them, [τ i ] γ represents the pheromone concentration in the i-th band; [η i ] δ represents the heuristic information on the i-th band, for example, the correlation between band i and the target water quality parameter; γ and δ are used to control the influence weights of pheromone and heuristic information; Σ j It means to perform a summation operation on all possible bands j.

[0098] In the actual execution process, each ant establishes a band combination through multiple selections until the preset number of bands or other conditions are met.

[0099] (10273)Fitness evaluation.

[0100] The band combination of each ant is evaluated according to the fitness function. The present invention can adopt any of the following fitness functions:

[0101]

[0102] Among them, R 2 It represents the coefficient of determination, which is between 0 and 1. 2 The closer the value is to 1, the better the model fits the data. 2 The closer the value is to 0, the worse the model fit is. i represents the actual concentration of the target water quality parameter at the i-th sampling point; Represents y i The predicted value of Represents all y i The average value of .

[0103]

[0104] Here, RMSE stands for root mean square error.

[0105] (10274)Pheromone update.

[0106] After each iteration, all ants will update the pheromone on the selected band path according to their fitness values. The formula for pheromone update is as follows:

[0107] τ i=(1-ρ)·τ i +Δτ i ...... (8)

[0108] Among them, (1-ρ) represents the pheromone volatility coefficient, which is used to control the decay rate of pheromone; Δτ i represents the pheromone increment of the i-th band, which is usually proportional to the corresponding fitness value and can be expressed as:

[0109]

[0110] Where Q is a constant, f k represents the fitness value of the Kth ant with respect to the current band combination (R 2 or RMSE).

[0111] (10275) Iterative optimization.

[0112] Repeat steps (10272) to (10274) to increase the pheromone concentration on the band combination path with a high fitness value through multiple iterations, and gradually find the optimal band combination.

[0113] (10276) Output the optimal band combination.

[0114] After multiple rounds of iterations, the path with the highest pheromone concentration represents the optimal band combination, which performs best in terms of correlation or prediction accuracy. The ACO algorithm can effectively screen sensitive bands and band combinations to meet the needs of hyperspectral inversion for band combination selection.

[0115] Continuing with the above example, assuming that the second hyperspectral satellite image is obtained using Zhuhai-1 satellite, when the target water quality parameter is total nitrogen, the optimal band combination is: [B21 band + B22 band]; when the target water quality parameter is total phosphorus, the optimal band combination is: B22 band + B31 band; when the target water quality parameter is chlorophyll a, the optimal band combination is:

[0116] (1028) A target inversion model is constructed by using a machine learning algorithm according to the actual concentration of the target water quality parameter and the optimal band combination. The target inversion model is a preset total nitrogen inversion model, a preset total phosphorus inversion model, or a preset chlorophyll a inversion model.

[0117] Specifically, first, a machine learning algorithm is selected and an inversion equation is constructed. The present invention can select a linear regression algorithm or a polynomial regression algorithm. In the linear regression algorithm, it is assumed that there is a linear relationship between the band combination (independent variable) and the target water quality parameter (dependent variable); when there is no linear relationship between the band combination and the target water quality parameter, a polynomial relationship between the two is considered.

[0118] Then, the data (relevant data of each sampling point) is divided into a training set and a test set, and the selected machine learning algorithm is trained using the training set to obtain the optimal inversion model of the target water quality parameter. It should be noted that the specific training method is an existing technology and will not be described in detail here.

[0119] In a specific example, assuming that the polynomial regression algorithm is selected and applied to the present invention, when the target water quality parameter is total nitrogen, the optimal inversion model obtained is: y = 85.7960x 0.4841 ; When the target water quality parameter is total phosphorus, the optimal inversion model is: y = -2E-07x 2 +0.0002x+0.0621; when the target water quality parameter is chlorophyll a, the optimal inversion model is: y=-20.843x 2 +57.085x-16.634. Among them, y represents the concentration of the target water quality parameter, and x represents the spectral reflectance value under the band or band combination.

[0120] Next, the test set is used to evaluate the model performance. Specifically, the sampling point data in the test set is used to compare the model prediction value with the measured value, and the determination coefficient R is used to 2 The prediction performance of the model is evaluated by using the root mean square error (RMSE) and mean absolute error (MAE) formulas, and the model parameters are optimized according to the evaluation results to obtain the inversion model (the preset total nitrogen inversion model, the preset total phosphorus inversion model or the preset chlorophyll a inversion model) that is finally applied to this application.

[0121] Among them, the mean absolute error MAE formula is defined as follows:

[0122]

[0123] Coefficient of determination R 2 It is used to measure the fit of the model. It shows the model's ability to explain the data. The closer the coefficient of determination is to 1, the better the model fits the data, that is, the model can explain most of the variation in the data. The root mean square error (RMSE) is used to measure the degree of error between the predicted value and the true value. The smaller the root mean square error, the smaller the deviation between the model's predicted value and the true value, and the higher the model's prediction accuracy. The mean absolute error (MAE) is used to measure the accuracy of the model's prediction results. It is the average of the absolute errors between the predicted value and the true value. It can intuitively reflect the average deviation of the predicted value from the true value. The smaller the mean absolute error, the more accurate the model's prediction results.

[0124] Step 103: Generate a eutrophication status distribution map of the first lake according to the concentration distribution maps of each water quality parameter.

[0125] In the embodiment of the present invention, generating a eutrophication state distribution map of the first lake according to the concentration distribution map of each water quality parameter may specifically include:

[0126] According to the concentration distribution map of each water quality parameter, the comprehensive nutrient status index of each area of ​​the first lake is calculated, and then the eutrophication level of each area of ​​the first lake is determined according to the comprehensive nutrient status index of each area of ​​the first lake to obtain a eutrophication status distribution map. Among them, the eutrophication levels include: oligotrophy, mesotrophy, mild eutrophy, moderate eutrophy and severe eutrophy.

[0127] Specifically, the concentration distribution map includes the concentration of the corresponding water quality parameter for each pixel of the first hyperspectral satellite image. For example, the concentration distribution map of total nitrogen includes the concentration of total nitrogen for each pixel of the first hyperspectral satellite image. Each pixel corresponds to an area of ​​the first lake. Therefore, the concentrations of the three water quality parameters in each area of ​​the first lake can be determined based on the concentration distribution maps of the three water quality parameters.

[0128] In the process of calculating the comprehensive nutrient status index of each area of ​​the first lake according to the concentration distribution map of each water quality parameter, first, for each area of ​​the first lake, the concentrations of the three water quality parameters in the area are determined according to the concentration distribution map of the three water quality parameters, and then, according to the total nitrogen concentration in the area, the nutrient status index of the total nitrogen concentration in the area is calculated (the calculation method is the existing technology), and similarly, the nutrient status index of the total phosphorus concentration and the nutrient status index of chlorophyll a in the area are calculated. Then, the nutrient status index of the three water quality parameters is substituted into the following formula to obtain the comprehensive nutrient status index TLI (Σ) of the area:

[0129]

[0130] Wherein, M represents the number of types of water quality parameters. In the present invention, M=3; W j represents the weight of the j-th water quality parameter. In a specific example, the weights of total nitrogen, total phosphorus and chlorophyll a can be 0.329, 0.278 and 0.393 respectively; TLI(j) represents the nutrient status index of the j-th water quality parameter; r ij It represents the correlation coefficient between the i-th water quality parameter and the j-th water quality parameter. j , TLI(j) are both dimensionless.

[0131] In the process of determining the eutrophication level of each area of ​​the first lake according to the comprehensive nutrient status index of each area of ​​the first lake, the generated comprehensive nutrient status index can be graded for eutrophication using a complete remote sensing image processing platform (The Environment for Visualizing Images, ENVI) or a reclassification plug-in in ArcGIS (ArcGIS Geographic Information System).

[0132] The present invention can pre-set the comprehensive nutrient status index range corresponding to different eutrophication levels. Subsequently, the eutrophication level of the area can be determined by analyzing the comprehensive nutrient status index range to which the comprehensive nutrient status index of any area of ​​the first lake belongs. In a specific example, as shown in Table 1, for each area of ​​the first lake, if the comprehensive nutrient status index of the area is less than 30, the eutrophication level of the area is determined to be oligotrophic; if the comprehensive nutrient status index of the area is greater than or equal to 30 and less than 50, the eutrophication level of the area is determined to be mesotrophic; if the comprehensive nutrient status index of the area is greater than or equal to 50 and less than 60, the eutrophication level of the area is determined to be slightly eutrophic; if the comprehensive nutrient status index of the area is greater than or equal to 60 and less than 70, the eutrophication level of the area is determined to be moderately eutrophic; if the comprehensive nutrient status index of the area is greater than or equal to 70, the eutrophication level of the area is determined to be severely eutrophic.

[0133] Table 1

[0134]

[0135]

[0136] The embodiment of the present invention adopts the above technical solution, a lake eutrophication evaluation method, including: obtaining a first hyperspectral satellite image of a first lake; using a preset total nitrogen inversion model, a preset total phosphorus inversion model and a preset chlorophyll a inversion model respectively, to invert water quality parameters of the first hyperspectral satellite image, and obtain concentration distribution maps of various water quality parameters, the water quality parameters include total nitrogen, total phosphorus and chlorophyll a; according to the concentration distribution map of each water quality parameter, a eutrophication state distribution map of the first lake is generated. Based on this, the present invention does not require field sampling, can improve the efficiency of lake eutrophication monitoring, and can be applied to water quality monitoring of large-scale lakes; and, because hyperspectral remote sensing technology has a detailed spectral resolution ability, it can identify the spectral response characteristics of different water quality parameters in different bands. Therefore, the present invention uses the inversion model to invert water quality parameters of the first hyperspectral satellite image of the lake, so that the present invention can comprehensively and carefully obtain the true water quality status of each area of ​​the lake, thereby improving the accuracy of lake eutrophication monitoring.

[0137] In addition, the comprehensive nutrient status index of each area of ​​the first lake is calculated according to the concentration distribution map of each water quality parameter, and then the eutrophication level of each area of ​​the first lake is determined according to the comprehensive nutrient status index of each area of ​​the first lake to obtain a eutrophication status distribution map, so that the staff can intuitively understand the eutrophication level of each area of ​​the first lake according to the eutrophication status distribution map, which is convenient for users to evaluate the eutrophication status of the lake.

[0138] Based on a general inventive concept, the present invention also provides a lake eutrophication evaluation device. Figure 2 Schematic diagram of the structure of a lake eutrophication assessment device provided by an embodiment of the present invention. Figure 1 As shown, the device comprises:

[0139] The acquisition module 21 is used to acquire a first hyperspectral satellite image of a first lake.

[0140] The inversion module 22 is used to invert water quality parameters of the first hyperspectral satellite image using a preset total nitrogen inversion model, a preset total phosphorus inversion model and a preset chlorophyll a inversion model to obtain a concentration distribution map of each water quality parameter, wherein the water quality parameters include total nitrogen, total phosphorus and chlorophyll a.

[0141] The generating module 23 is used to generate a eutrophication state distribution map of the first lake according to the concentration distribution map of each water quality parameter.

[0142] Optionally, the lake eutrophication assessment device of the present invention may further include a model building module, and the model building module specifically includes:

[0143] The first acquisition unit is used to obtain the geographical coordinates of multiple sampling points of the second lake and the actual concentration of the target water quality parameter at the sampling time, where the target water quality parameter is total nitrogen, total phosphorus or chlorophyll a.

[0144] The second acquisition unit is used to acquire a second hyperspectral satellite image of the second lake at the sampling time.

[0145] The preprocessing unit is used to preprocess the second hyperspectral satellite image.

[0146] The extraction unit is used to extract the spectral reflectance value of the corresponding position in the preprocessed second hyperspectral satellite image based on the geographic coordinates to obtain multi-band spectral information.

[0147] The standardization processing unit is used to perform standardization processing on multi-band spectral information.

[0148] The first screening unit is used to screen the sensitive bands of the standardized multi-band spectral information based on a preset regression algorithm to obtain the sensitive bands of the target water quality parameters.

[0149] The second screening unit is used to perform band combination screening on the sensitive bands of the target water quality parameters to obtain the optimal band combination of the target water quality parameters.

[0150] A construction unit is used to construct a target inversion model by using a machine learning algorithm according to the actual concentration of the target water quality parameter and the optimal band combination. The target inversion model is a preset total nitrogen inversion model, a preset total phosphorus inversion model or a preset chlorophyll a inversion model.

[0151] Optionally, the pre-processing unit is specifically used to perform radiation calibration processing, atmospheric correction processing and geographic location correction processing on the second hyperspectral satellite image.

[0152] Optionally, the preset regression algorithm is a LASSO regression algorithm.

[0153] Optionally, the second screening unit is specifically used to perform band combination screening on sensitive bands of the target water quality parameter based on an ant colony optimization algorithm to obtain an optimal band combination of the target water quality parameter.

[0154] Optionally, the machine learning algorithm includes a linear regression algorithm and a polynomial regression algorithm.

[0155] Optionally, the generating module 23 may be specifically used for:

[0156] (1) Based on the concentration distribution diagram of each water quality parameter, calculate the comprehensive nutrient status index of each area of ​​the first lake.

[0157] (2) According to the comprehensive nutrient status index of each area of ​​the first lake, the eutrophication level of each area of ​​the first lake is determined to obtain a eutrophication status distribution map.

[0158] Based on a general inventive concept, the present invention also provides a lake eutrophication assessment device. Figure 3 Schematic diagram of the structure of a lake eutrophication assessment device provided by an embodiment of the present invention. Figure 3 As shown, the device 300 includes:

[0159] at least one processor 310; and,

[0160] A memory 330 is communicatively connected to at least one processor 310; wherein,

[0161] The memory 330 stores instructions 320 that can be executed by at least one processor 310. The instructions 320 are executed by the at least one processor 310 so that the at least one processor 310 can implement each step of the lake eutrophication assessment method as described above.

[0162] Based on a general inventive concept, the present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, each step in the lake eutrophication assessment method as described above is implemented.

[0163] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0164] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0165] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present invention belong.

[0166] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0167] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0168] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0169] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0170] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0171] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A lake eutrophication assessment method, characterized in that: include: Acquire the first hyperspectral satellite image of the first lake; Using a preset total nitrogen inversion model, a preset total phosphorus inversion model, and a preset chlorophyll a inversion model, respectively, the first hyperspectral satellite image is used to invert water quality parameters to obtain a concentration distribution map of each water quality parameter, wherein the water quality parameters include total nitrogen, total phosphorus, and chlorophyll a; A eutrophication status distribution map of the first lake is generated according to the concentration distribution map of each of the water quality parameters.

2. The lake eutrophication evaluation method according to claim 1, characterized in that: The construction process of any one of the preset total nitrogen inversion model, the preset total phosphorus inversion model and the preset chlorophyll a inversion model includes: Obtaining the geographic coordinates of a plurality of sampling points of the second lake and the actual concentration of a target water quality parameter at the sampling time, wherein the target water quality parameter is total nitrogen, total phosphorus or chlorophyll a; Acquire a second hyperspectral satellite image of the second lake at the sampling time; preprocessing the second hyperspectral satellite image; Based on the geographic coordinates, extracting the spectral reflectance value of the corresponding position in the preprocessed second hyperspectral satellite image to obtain multi-band spectral information; Performing standardization processing on the multi-band spectral information; Based on a preset regression algorithm, sensitive bands are screened for the standardized multi-band spectral information to obtain sensitive bands of the target water quality parameters; Performing band combination screening on the sensitive bands of the target water quality parameter to obtain the optimal band combination of the target water quality parameter; A target inversion model is constructed by using a machine learning algorithm according to the actual concentration of the target water quality parameter and the optimal band combination. The target inversion model is the preset total nitrogen inversion model, the preset total phosphorus inversion model or the preset chlorophyll a inversion model.

3. The lake eutrophication evaluation method according to claim 2, characterized in that: Preprocessing the second hyperspectral satellite image specifically includes: The second hyperspectral satellite image is subjected to radiation calibration processing, atmospheric correction processing and geographic location correction processing.

4. The lake eutrophication evaluation method according to claim 2, characterized in that: The preset regression algorithm is the LASSO regression algorithm.

5. The lake eutrophication evaluation method according to claim 2, characterized in that: The sensitive bands of the target water quality parameters are screened by band combination to obtain the optimal band combination of the target water quality parameters, specifically including: Based on the ant colony optimization algorithm, the sensitive bands of the target water quality parameters are screened for band combination to obtain the optimal band combination of the target water quality parameters.

6. The lake eutrophication evaluation method according to claim 2, characterized in that: The machine learning algorithms include linear regression algorithms and polynomial regression algorithms.

7. The lake eutrophication evaluation method according to claim 1, characterized in that: Generating a eutrophication state distribution map of the first lake according to the concentration distribution map of each of the water quality parameters specifically includes: Calculating the comprehensive nutrient status index of each area of ​​the first lake according to the concentration distribution map of each water quality parameter; According to the comprehensive nutrient status index of each area of ​​the first lake, the eutrophication level of each area of ​​the first lake is determined to obtain the eutrophication status distribution map.

8. A lake eutrophication assessment device, characterized in that: include: An acquisition module, used for acquiring a first hyperspectral satellite image of a first lake; an inversion module, for performing water quality parameter inversion on the first hyperspectral satellite image using a preset total nitrogen inversion model, a preset total phosphorus inversion model, and a preset chlorophyll a inversion model, respectively, to obtain a concentration distribution map of each water quality parameter, wherein the water quality parameters include total nitrogen, total phosphorus, and chlorophyll a; A generation module is used to generate a eutrophication status distribution map of the first lake according to the concentration distribution map of each water quality parameter.

9. A lake eutrophication assessment device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can implement each step in the lake eutrophication assessment method as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, each step of the lake eutrophication assessment method according to any one of claims 1 to 7 is implemented.

Images