River water quality parameter inversion method and system, storage medium and electronic equipment

Abstract

The invention provides a river water quality parameter inversion method and system, a storage medium and electronic equipment. The method comprises the following steps: acquiring a remote sensing image and a water quality monitoring parameter of a to-be-monitored river; acquiring a reflectivity image of the river area to be monitored based on the remote sensing image; acquiring a characteristic wave band based on the reflectivity image and the water quality monitoring parameters; and based on a water quality parameter inversion model, according to the reflectivity image corresponding to the characteristic wave band, obtaining the content of total phosphorus and total nitrogen in the water body of the to-be-monitored river. According to the river water quality parameter inversion method and system, the storage medium and the electronic equipment, remote sensing inversion of the river water quality parameters is realized based on a multi-objective optimization modeling strategy, and rapid and large-range monitoring of the total phosphorus and total nitrogen concentration of the river is facilitated.

Description

Technical Field

[0001] The present invention belongs to the technical field of water environment remote sensing monitoring, and in particular relates to a river water quality parameter inversion method, system, storage medium and electronic equipment. Background Art

[0002] Rivers are one of the important freshwater resources on the earth. They have multiple functions such as irrigation, navigation, regulating local microclimate and maintaining regional ecosystem balance. The health of their water environment directly affects human survival and social development. Excessive nitrogen and phosphorus in rivers will lead to eutrophication of water bodies, causing water quality deterioration and producing large-scale impacts. Therefore, accurately estimating the total phosphorus (TP) and total nitrogen (TN) content in rivers is of great significance for monitoring and managing the river ecological environment and ensuring the healthy and stable quality of water bodies. The "Jilin-1" Spectrum 01 and 02 satellites can obtain remote sensing data with a resolution of 5 meters in 7 spectral bands, which brings new opportunities for water quality inversion. However, the optical properties of river water bodies are complex and diverse, and there are complex interactions between the various components. Accurately inverting the TP and TN content in rivers based on the "Jilin-1" Spectrum 01 and 02 satellite data has also become an open problem that needs to be solved urgently.

[0003] The existing methods for inverting river water quality parameters based on remote sensing can be divided into the following four categories:

[0004] (1) Empirical method

[0005] The empirical method for water quality parameter inversion is to perform mathematical and statistical analysis based on spectral reflectance (single band or multi-band combination) and the corresponding water quality parameter concentration, and establish a regression equation based on the principle of minimum error. The empirical model is a completely or semi-data-driven method, which does not consider the water quality radiation transmission mechanism, is mainly based on experience and statistical relationships, and lacks generalization ability.

[0006] (2) Analytical method

[0007] The analytical model method is based on the radiation transfer theory, analyzes the absorption and scattering characteristics of each water quality index for a specific spectral band, and establishes an inversion model for water quality parameters. This type of model is highly versatile and does not require too many water quality samples and remote sensing data, but the establishment of the model has high requirements for spectral measurement data and requires the support of specific spectral measurement instruments and measurement environments. There are many limitations in practical applications.

[0008] (3) Semi-analytical method

[0009] The semi-analytical model is also based on the radiation transfer theory and the optical properties of water quality indicators, and uses statistical methods to obtain some parameters in the water quality inversion model to reduce the workload of modeling. The construction of this model is relatively simple and does not require complex biochemical analysis of water quality indicators, but some parameters in the model are dependent on objective factors such as region and climate.

[0010] (4) Machine Learning Methods

[0011] The optical properties of river water are relatively complex, and traditional linear models cannot accurately monitor water quality parameters in the water environment. The machine learning model is a typical nonlinear model that can better solve complex nonlinear problems and has significant effects on improving the accuracy of water quality parameter inversion models in the field of remote sensing water quality monitoring.

[0012] In the field of remote sensing river water quality index inversion, machine learning methods have higher fitting accuracy and universality. However, the algorithm still has the following problems during use:

[0013] (1) The parameters of existing methods, such as support vector machine regression, are mainly based on heuristics. The best parameter combinations are selected through human experience and experiments. This is time-consuming and the selected parameters are not representative.

[0014] (2) The mapping relationship of traditional machine learning algorithms is that multiple input variables correspond to one output parameter. When multiple target variables are required, models of corresponding parameters need to be established separately for fitting, ignoring the potential relationship between the target parameters, resulting in poor accuracy.

[0015] Therefore, how to provide a remote sensing-based river water quality parameter inversion method to solve the problems that the existing technology is difficult to apply to the joint estimation of multiple water quality parameters and the inversion model has poor adaptability and low accuracy, has become a technical problem that needs to be urgently solved by technical personnel in this field. Summary of the invention

[0016] In view of the shortcomings of the prior art mentioned above, the purpose of the present invention is to provide a river water quality parameter inversion method, system, storage medium and electronic equipment, which realize remote sensing inversion of river water quality parameters based on a multi-objective optimization modeling strategy and realize the monitoring of total phosphorus and total nitrogen concentrations in river water bodies.

[0017] In a first aspect, the present invention provides a method for inverting river water quality parameters, the method comprising the following steps: obtaining remote sensing images and water quality monitoring parameters of the river to be monitored; obtaining a reflectivity image of the river area to be monitored based on the remote sensing image; obtaining characteristic bands based on the reflectivity image and the water quality monitoring parameters; and obtaining the total phosphorus and total nitrogen contents of the river to be monitored according to the reflectivity image corresponding to the characteristic band based on a water quality parameter inversion model.

[0018] In an implementation of the first aspect, acquiring the reflectivity image of the river to be monitored based on the remote sensing image includes the following steps:

[0019] Performing geometric correction on the remote sensing image;

[0020] Carry out radiometric calibration on the remote sensing images after geometric correction;

[0021] Perform atmospheric correction on remote sensing images after radiometric calibration;

[0022] Water body extraction is performed on the remote sensing image after atmospheric correction to obtain a reflectivity image of the river area to be monitored.

[0023] In an implementation of the first aspect, acquiring a characteristic band based on the reflectivity image and the water quality monitoring parameter comprises the following steps:

[0024] Performing band expansion on the reflectivity image to obtain the reflectivity of the expanded band;

[0025] A spectral index independent variable data set of river water quality is formed based on the band reflectance of the reflectance image and the reflectance of the extended band;

[0026] A characteristic band is obtained based on the spectral index independent variable data set and the water quality monitoring parameter.

[0027] In an implementation of the first aspect, the reflectivity image is band expanded, and the reflectivity of the expanded band is obtained using the following formula:

[0028]

[0029] Among them, x i and x j Represent the reflectivity of the i-th and j-th bands respectively, x i+j 、x i-j 、x i×j 、x i / j and x i，j Indicates the reflectivity of the extended band.

[0030] In an implementation of the first aspect, generating a characteristic band based on the spectral index independent variable data set and the water quality monitoring parameter comprises the following steps:

[0031] Calculate the Pearson correlation coefficient between the variables in the spectral index independent variable data set and the total phosphorus concentration and total nitrogen concentration in the water quality monitoring parameters Where R P and R N represents the Pearson correlation coefficient between the ith variable and the total phosphorus concentration and total nitrogen concentration in the water quality monitoring parameters, respectively, i represents the value of the i-th variable, Represents x i The mean of y iP and iN represent the i-th total phosphorus concentration and total nitrogen concentration in the water quality monitoring parameters respectively, and Respectively represent y iP and iN The mean of , n represents the number of variables;

[0032] The bands corresponding to the variables whose Pearson correlation coefficients are all greater than a preset threshold are selected as feature bands.

[0033] In an implementation of the first aspect, based on the water quality parameter inversion model, obtaining the total phosphorus and total nitrogen content of the river to be monitored according to the reflectivity image corresponding to the characteristic band includes the following steps:

[0034] Constructing a water quality parameter inversion model; the water quality parameter inversion model is obtained by optimizing a multi-objective support vector regression model based on a genetic algorithm;

[0035] The reflectivity image corresponding to the characteristic band is input into the water quality parameter inversion model to obtain the total phosphorus and total nitrogen contents of the river to be monitored output by the water quality parameter inversion model.

[0036] In an implementation of the first aspect, constructing a water quality parameter inversion model comprises the following steps:

[0037] Construct a multi-objective support vector regression model;

[0038] Calculate the penalty factor of the multi-objective support vector regression model and the real value of the kernel parameter in the radial basis kernel function based on the genetic algorithm;

[0039] Substitute the real value into the multi-objective support vector regression model to obtain the water quality parameter inversion model.

[0040] In a second aspect, the present invention provides a river water quality parameter inversion system, the system comprising a first acquisition module, a second acquisition module, a third acquisition module and an inversion module;

[0041] The first acquisition module is used to acquire remote sensing images and water quality monitoring parameters of the river to be monitored;

[0042] The second acquisition module is used to acquire the reflectivity image of the river area to be monitored based on the remote sensing image;

[0043] The third acquisition module is used to acquire characteristic bands based on the reflectivity image and the water quality monitoring parameters;

[0044] The inversion module is used to obtain the total phosphorus and total nitrogen content of the water body of the river to be monitored based on the water quality parameter inversion model and the reflectivity image corresponding to the characteristic band.

[0045] In a third aspect, the present invention provides an electronic device, the electronic device comprising: a processor and a memory;

[0046] The memory is used to store computer programs;

[0047] The processor is used to execute the computer program stored in the memory so that the electronic device executes the above-mentioned river water quality parameter inversion method.

[0048] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by an electronic device, implements the above-mentioned river water quality parameter inversion method.

[0049] As described above, the river water quality parameter inversion method, system, storage medium and electronic device of the present invention have the following beneficial effects:

[0050] (1) Remote sensing inversion of river water quality parameters was achieved based on a multi-objective optimization modeling strategy, and real-time soft measurement of total phosphorus and total nitrogen concentrations in river water was realized;

[0051] (2) The complex relationship between water quality parameters is fully considered, and the multi-target water quality parameters of total phosphorus and total nitrogen in the river are jointly estimated to improve the inversion accuracy;

[0052] (3) It solves the problems that existing methods are difficult to apply to the joint estimation of multiple water quality parameters and the inversion models have poor adaptability and low accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 Shown is a flow chart of a river water quality parameter inversion method in one embodiment of the present invention;

[0054] Figure 2 Shown is a flow chart of an embodiment of the present invention in which a reflectivity image of a river area to be monitored is obtained based on a remote sensing image;

[0055] Figure 3 Shown is a flow chart of an embodiment of the present invention for obtaining characteristic bands based on reflectivity images and water quality monitoring parameters;

[0056] Figure 4 Shown is a schematic structural diagram of a river water quality parameter inversion system according to an embodiment of the present invention;

[0057] Figure 5 It is a schematic structural diagram of an electronic device of the present invention in one embodiment. DETAILED DESCRIPTION

[0058] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0059] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0060] The following embodiments of the present invention provide a river water quality parameter inversion method, which is applied to electronic devices. The electronic devices described in the present invention may include mobile phones, tablet computers, laptop computers, wearable devices, vehicle-mounted devices, augmented reality (AR) / virtual reality (VR) devices, ultra-mobile personal computers (UMPC), netbooks, personal digital assistants (PDA), etc., which have wireless charging functions. The embodiments of the present invention do not impose any restrictions on the specific types of electronic devices.

[0061] For example, the electronic device may be a station (STAION, ST) in a WLAN with a wireless charging function, a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA) device, a handheld device with a wireless charging function, a computing device or other processing device, a computer, a laptop computer, a handheld communication device, a handheld computing device, and / or other devices for communicating on a wireless system and a next generation communication system, such as a mobile terminal in a 5G network, a mobile terminal in a future evolved Public Land Mobile Network (PLMN), or a mobile terminal in a future evolved Non-terrestrial Network (NTN).

[0062] For example, the electronic device can communicate with the network and other devices through wireless communication. The above wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobilecommunication (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), email, Short Messaging Service (SMS), BT, GNSS, WLAN, NFC, FM, and / or IR technology. The GNSS can include Global Positioning System (GPS), Global Navigation Satellite System (GLONASS), BeiDou navigation Satellite System (BDS), Quasi-Zenith Satellite System (QZSS) and / or Satellite Based Augmentation Systems (SBAS).

[0063] The technical solutions in the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0064] like Figure 1 As shown, in one embodiment, the river water quality parameter inversion method of the present invention includes steps S1 to S4.

[0065] Step S1, obtaining remote sensing images and water quality monitoring parameters of the river to be monitored.

[0066] Specifically, a remote sensing image of the river to be monitored is obtained based on a remote sensing image acquisition device, wherein the remote sensing image multispectral remote sensing data includes 7 bands (B1-B7) with a spatial resolution of 5m and a band range between 403nm-899.5nm.

[0067] Water quality monitoring parameters are collected based on water quality monitoring equipment, wherein the water quality monitoring parameters mainly include total phosphorus concentration and total nitrogen concentration.

[0068] Step S2: acquiring a reflectivity image of the river area to be monitored based on the remote sensing image.

[0069] Specifically, Figure 2 As shown, obtaining the reflectivity image of the river to be monitored based on the remote sensing image includes the following steps:

[0070] 21) Performing geometric correction on the remote sensing image.

[0071] Among them, geometric correction is based on the standard topographic map, and selects rivers, roads, buildings and other distinctive features as ground control points (GCP). The quadratic polynomial method is used to correct the image to eliminate the image geometric distortion caused by the satellite sensor attitude, atmospheric refraction and the earth's rotation.

[0072] 22) Carry out radiometric calibration on the remote sensing images after geometric correction.

[0073] Among them, radiation calibration is based on the quantitative relationship between the DN (Digital Number) value and the radiation brightness above the atmosphere, converting the dimensionless DN value of the remote sensing image into a reflectivity with physical meaning, eliminating the error caused by the performance limitations of the satellite sensor itself. It should be noted that the DN value is the brightness value of the remote sensing image pixel, the grayscale value of the recorded object, without unit, and is an integer value. The value size is related to the radiation resolution of the sensor, the emissivity of the object, the atmospheric transmittance and scattering rate, etc.

[0074] 23) Perform atmospheric correction on the remote sensing images after radiation calibration.

[0075] Among them, atmospheric correction refers to the process of converting the apparent reflectivity of the atmosphere into the true reflectivity of ground targets, eliminating the reflectivity deviation of ground objects caused by atmospheric interference.

[0076] 24) Extracting water bodies from the remote sensing image after atmospheric correction to obtain a reflectivity image of the river area to be monitored.

[0077] The water body extraction is performed by masking the remote sensing image after atmospheric correction using the river vector file to obtain the reflectivity image of the river to be monitored.

[0078] Step S3: acquiring characteristic bands based on the reflectivity image and the water quality monitoring parameters.

[0079] Specifically, Figure 3 As shown, obtaining characteristic bands based on the reflectivity image and the water quality monitoring parameters includes the following steps:

[0080] 31) Performing band expansion on the reflectivity image to obtain the reflectivity of the expanded band.

[0081] Among them, the original images of the 01 and 02 satellites of the "Jilin-1" spectrum only have 7 bands with a resolution of 5 meters available, which is difficult to meet the needs of quantitative remote sensing inversion of water quality. Therefore, the present invention adopts a band combination method to expand the number of characteristic bands, and at the same time can effectively reduce the interference of environmental noise signals (light changes, atmospheric images, etc.) on the spectral information of remote sensing images, and improve the quality of remote sensing data. Based on the principle of the reflectivity equivalence relationship between remote sensing data and measured spectral data, the present invention expands the bands of reflectivity images through the following 5 remote sensing data band combination methods, and the formula is as follows:

[0082]

[0083] Among them, x i and x j Represent the reflectivity of the i-th and j-th bands respectively, x i+j 、x i-j 、x i×j 、x i / j and x i，j Indicates the reflectivity of the extended band.

[0084] 32) A spectral index independent variable dataset of river water quality is constructed based on the band reflectance of the reflectance image and the reflectance of the extended band.

[0085] Among them, the reflectance of 7 bands of the original images of the "Jilin-1" spectrum 01 and 02 satellites and the expanded band reflectance were selected to form the spectral index independent variable data set of total phosphorus and total nitrogen concentrations in river water bodies.

[0086] 33) Acquire characteristic bands based on the spectral index independent variable data set and the water quality monitoring parameters.

[0087] Among them, for the variables in the spectral index independent variable data set, namely each reflectance, the Pearson correlation coefficient between the variables in the spectral index independent variable data set and the total phosphorus concentration and total nitrogen concentration in the water quality monitoring parameters is calculated. Where R P and R N represents the Pearson correlation coefficient between the ith variable and the total phosphorus concentration and total nitrogen concentration in the water quality monitoring parameters, respectively, i represents the value of the i-th variable, Represents x i The mean of y iP and iN represent the i-th total phosphorus concentration and total nitrogen concentration in the water quality monitoring parameters respectively, and Respectively represent y iP and iNThe mean of n represents the number of variables. Then, the bands corresponding to the variables whose Pearson correlation coefficients are all greater than the preset threshold are selected as characteristic bands. The spectral index independent variable data set is updated based on the characteristic bands. The updated spectral index independent variable data set is integrated with the water quality total phosphorus and total nitrogen index data corresponding to the same point to obtain a sample data set.

[0088] Step S4: Based on the water quality parameter inversion model, the total phosphorus and total nitrogen content of the water body of the river to be monitored is obtained according to the reflectivity image corresponding to the characteristic band.

[0089] Specifically, based on the water quality parameter inversion model, obtaining the total phosphorus and total nitrogen content of the water body of the river to be monitored according to the reflectivity image corresponding to the characteristic band includes the following steps:

[0090] 41) Constructing a water quality parameter inversion model; the water quality parameter inversion model is obtained by optimizing a multi-objective support vector regression model based on a genetic algorithm.

[0091] Among them, the genetic algorithm (GA) is used to optimize the multi-output support vector machine regression model (MSVR) to obtain the water quality parameter inversion model GA-MSVR. The traditional support vector machine regression is to establish a regression function based on the input data set to predict continuous unknown values. For one-dimensional regression problems, support vector machine regression can effectively fit the nonlinear relationship between input samples and dependent variables. However, when multiple observations need to be output simultaneously, the inherent connection between the predicted variables is ignored. MSVR can effectively take into account the dependency between variables. Sample data set P = {x i ,y i}, where xi is the input feature vector, xi∈Rd, yi is the target vector, y i ∈R Q Each target variable has a corresponding regression coefficient W j and b j (j=1,2,…,Q). Therefore, based on the one-dimensional support vector machine regression problem, the multi-dimensional output objective function formula is expanded as follows: in, C is the penalty factor; W = [w 1 ,…w Q ],b=[b 1 , …b Q ]; L ε (u) represents the multidimensional Vapnik ε-insensitive loss function: Among them, ε is an insensitive coefficient used to control the simulation accuracy. When ε = 0, the problem is converted into a kernel ridge regression prediction problem for each component; when ε ≠ 0, the quasi-Newton method is used to solve each iterative calculation.

[0092] The multi-dimensional target variable problem is solved by the iterative reweighted least-squares (IRWLS) process. First, based on L ε (u) The first-order Taylor expansion of L p (W, b) is approximated by the following formula:

[0093]

[0094] In the formula, k is the number of iterations; L' p (W k , b k ) is L p (W k , b k ). p (W, b) is estimated twice, and the formula is as follows:

[0095]

[0096] Where τ is the sum of all constant terms; Formula L" p The value and gradient change of (W, b) and L p (W, b) when W = W k , b=b k The time is consistent.

[0097] To obtain the optimal solution of the original function, the calculation idea is as follows: First, k, W k , b k The value is set to 0, and the calculation formula L" p a in (W, b) i and Next, calculate the formula L” p The solution of (W, b) is W s and b s . Define the matrix P of the descent search direction of the original objective function k , each column is a descending search direction for each regression coefficient, and the formula is Then, the solution for the next step is calculated as follows: Use the backtracking algorithm to calculate the step length η k , set the initial value to 1 and judge the condition. If it is not satisfied, set η kMultiply it with a positive number less than 1, and perform calculations and judgments until the conditions are met. Finally, calculate and a i Let k = k + 1 and calculate L again. p The solution of (W, b) is continued until convergence.

[0098] In the above MSVR fitting process, the selection of penalty factor C and kernel parameter σ in radial basis kernel function will have a direct impact on the fitting effect and accuracy, so genetic algorithm is used to optimize these two parameters. Genetic algorithm imitates the principle of Darwin's theory of biological evolution and reproduces the better individuals selected in the previous generation to generate a group of potential solutions. The new generation has better characteristics than the old generation. After inheritance and mutation, according to the evolution law of "survival of the fittest", the better characteristics will be spread throughout the group through integration and exchange, so as to obtain the best individual in space or the global optimal solution. Based on GA method, the simultaneous optimization of the two parameters of MSVR model, penalty factor C and kernel parameter σ, is realized as follows.

[0099] 1) First, the fitness function needs to be determined. The example of the present invention selects the mean absolute percentage error (MAPE) as the objective function. The smaller the value, the better the corresponding result. The formula is as follows:

[0100]

[0101] Where M TP and M TN Respectively represent the true total phosphorus concentration and total nitrogen concentration of the kth point in the test set, M' TP and M' TN They refer to the predicted total phosphorus concentration and total nitrogen concentration at the corresponding points, respectively.

[0102] 2) Next, establish an initial population, set the initial population size Popsize = 300, each individual has a chromosome consisting of two gene segments of parameters C and σ, and initialize GA parameters, set the maximum number of iterations Maxgen = 100, crossover probability Crossoverpercent = 0.6, mutation probability Mutationpercent = 0.01, selection probability Selectionprobability = 0.08, and fitness threshold is 10%. Binary encoding of the parameter penalty factor C and the kernel parameter σ is obtained and Form a set of individual solutions. and Represents the genotype of the corresponding chromosome of C and σ, and the number of digits is n C and n σThe binary string corresponds to a set of real-valued parameters C and σ. At the same time, the reflectivity corresponding to each pixel in the remote sensing image needs to be encoded and solved, and so on. When calculating the real values ​​of the parameters C and σ, the genotype needs to be converted to the phenotype, that is, the binary code is interpreted as a real value. The conversion function formula of the parameters C and σ is In the formula, p corresponds to the real value of parameter P, which is the phenotype of the chromosome with l-bit gene. Given the range of C and σ and the chromosome length d, the binary value of the chromosome genotype can be converted into a decimal real value. P and max P They respectively represent the minimum and maximum real-valued values ​​of the parameter P (in the example of the present invention, the value range of parameter C is [0.001, 300], and the value range of σ is [0.001, 10]), d is the decimal integer corresponding to the binary string of the chromosome segment, and the required p value is the actual decimal real value.

[0103] 3) Secondly, for each individual in the previous step, the penalty factor C and the real value of the kernel parameter σ are decoded based on the conversion function, the fitness value is calculated, and the individuals are sorted from high to low according to the size of the fitness value. A certain number of individuals are selected from the best to the worst according to the selection probability, and the chromosomes with low mean absolute percentage error enter the next generation.

[0104] 4) Secondly, determine whether the fitness of the optimal individual reaches the set threshold. If it has, stop the evolution and return the optimal solution represented by the optimal individual. Otherwise, proceed to the next step to perform evolution operations on the population to generate the next generation of individuals.

[0105] 5) Then, genetic operations are performed on the population, and individual genes are crossed and mutated according to the set crossover probability and mutation probability to produce genes with better fitness and obtain chromosomes with better fitness.

[0106] 6) Finally, determine whether to terminate population evolution based on the maximum number of iterations and the fitness threshold. When the fitness of the best individual in a certain generation of population reaches the set evolutionary generation, or the fitness function value of the best individual reaches the set threshold, stop the evolution of the population and obtain the optimal chromosome. The optimal chromosome is converted into the corresponding real value of the penalty factor C and the nuclear parameter σ. Otherwise, enter the next generation of evolution.

[0107] Therefore, after constructing the multi-objective support vector regression model, the penalty factor of the multi-objective support vector regression model and the real value of the kernel parameter in the radial basis kernel function are calculated based on the genetic algorithm, and the value of the MSVR model penalty factor C and the kernel parameter σ is set to the real value of the two parameters obtained by GA optimization, so as to obtain the water quality parameter inversion model. In addition, 70% of the data in the total sample data set are randomly selected as training samples to form a training sample set, which is input into the water quality parameter inversion model for training to obtain a trained water quality parameter inversion model GA-MSVR. Among them, the MSVR method can reduce the calculation load and greatly improve the operation efficiency. The water quality inversion model GA-MSVR has the advantages of fast speed and high accuracy in the field of water quality parameter inversion.

[0108] 42) Inputting the reflectance image corresponding to the characteristic band into the water quality parameter inversion model, and obtaining the total phosphorus and total nitrogen content of the water body of the river to be monitored output by the water quality parameter inversion model.

[0109] The reflectance images corresponding to the characteristic bands of the river are traversed to obtain the estimated results of the total phosphorus and total nitrogen concentrations of the river.

[0110] The protection scope of the river water quality parameter inversion method described in the embodiment of the present invention is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing or replacing steps in the prior art based on the principles of the present invention are included in the protection scope of the present invention.

[0111] An embodiment of the present invention also provides a river water quality parameter inversion system, which can implement the river water quality parameter inversion method described in the present invention. However, the implementation device of the river water quality parameter inversion system described in the present invention includes but is not limited to the structure of the river water quality parameter inversion system listed in this embodiment. All structural deformations and replacements of the prior art made according to the principles of the present invention are included in the protection scope of the present invention.

[0112] like Figure 4 As shown, in one embodiment, the river water quality parameter inversion system of the present invention includes a first acquisition module 41 , a second acquisition module 42 , a third acquisition module 43 and an inversion module 44 .

[0113] The first acquisition module 41 is used to acquire remote sensing images and water quality monitoring parameters of the river to be monitored.

[0114] The second acquisition module 42 is connected to the first acquisition module 41, and is used to acquire the reflectivity image of the river area to be monitored based on the remote sensing image.

[0115] The third acquisition module 43 is connected to the first acquisition module 41 and the second acquisition module 42, and is used to acquire characteristic bands based on the reflectivity image and the water quality monitoring parameters.

[0116] The inversion module 44 is connected to the third acquisition module 43, and is used to obtain the total phosphorus and total nitrogen content of the water body of the river to be monitored based on the water quality parameter inversion model and the reflectivity image corresponding to the characteristic band.

[0117] Among them, the structures and principles of the first acquisition module 41, the second acquisition module 42, the third acquisition module 43 and the inversion module 44 correspond one to one with the steps in the above-mentioned river water quality parameter inversion method, so they are not repeated here.

[0118] In the several embodiments provided by the present invention, it should be understood that the disclosed system, device or method can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules / units is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules or units, which can be electrical, mechanical or other forms.

[0119] The modules / units described as separate components may or may not be physically separated, and the components displayed as modules / units may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules / units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention. For example, the functional modules / units in the various embodiments of the present invention may be integrated into one processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into one module / unit.

[0120] Those of ordinary skill in the art should further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0121] The embodiment of the present invention also provides a computer-readable storage medium. A person of ordinary skill in the art can understand that all or part of the steps in the method for implementing the above embodiment can be completed by instructing the processor through a program, and the program can be stored in a computer-readable storage medium, and the storage medium is a non-transitory medium, such as a random access memory, a read-only memory, a flash memory, a hard disk, a solid-state hard disk, a magnetic tape, a floppy disk, an optical disc, and any combination thereof. The above storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid-state disk (SSD)), etc.

[0122] An embodiment of the present invention further provides an electronic device, which includes a processor and a memory.

[0123] The memory is used to store computer programs.

[0124] The memory includes: ROM, RAM, disk, USB flash drive, memory card or CD and other media that can store program codes.

[0125] The processor is connected to the memory and is used to execute the computer program stored in the memory so that the electronic device executes the above-mentioned river water quality parameter inversion method.

[0126] Preferably, the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0127] like Figure 5As shown, the electronic device of the present invention is in the form of a general computing device. The components of the electronic device may include but are not limited to: one or more processors or processing units 51, a memory 52, and a bus 53 connecting different system components (including the memory 52 and the processing unit 51).

[0128] Bus 53 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnect (PCI) bus.

[0129] Electronic devices typically include a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device, including volatile and non-volatile media, removable and non-removable media.

[0130] The memory 52 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 521 and / or cache memory 522. The electronic device may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 523 may be used to read and write non-removable, non-volatile magnetic media ( Figure 5 not shown, usually called a "hard drive"). Although Figure 5 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 53 via one or more data medium interfaces. The memory 52 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0131] A program / utility 524 having a set (at least one) of program modules 5241 may be stored, for example, in the memory 52, such program modules 5241 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination thereof may include an implementation of a network environment. The program modules 5241 generally perform the functions and / or methods of the embodiments described herein.

[0132] The electronic device may also communicate with one or more external devices (e.g., keyboards, pointing devices, displays, etc.), one or more devices that enable a user to interact with the electronic device, and / or any device that enables the electronic device to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via input / output (I / O) interface 54. Furthermore, the electronic device may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via network adapter 55. Figure 5 As shown, the network adapter 55 communicates with other modules of the electronic device via the bus 53. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0133] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A method for inverting river water quality parameters, characterized in that: The method comprises the following steps: Obtain remote sensing images and water quality monitoring parameters of the rivers to be monitored; Acquire a reflectivity image of the river area to be monitored based on the remote sensing image; Obtaining characteristic bands based on the reflectivity image and the water quality monitoring parameters; Based on the water quality parameter inversion model, the total phosphorus and total nitrogen content of the water body of the river to be monitored is obtained according to the reflectivity image corresponding to the characteristic band.

2. The method for inverting river water quality parameters according to claim 1, characterized in that: Acquiring the reflectivity image of the river to be monitored based on the remote sensing image comprises the following steps: Performing geometric correction on the remote sensing image; Carry out radiometric calibration on the remote sensing images after geometric correction; Perform atmospheric correction on remote sensing images after radiometric calibration; Water body extraction is performed on the remote sensing image after atmospheric correction to obtain a reflectivity image of the river area to be monitored.

3. The method for inverting river water quality parameters according to claim 1, characterized in that: Acquiring characteristic bands based on the reflectivity image and the water quality monitoring parameters comprises the following steps: Performing band expansion on the reflectivity image to obtain the reflectivity of the expanded band; A spectral index independent variable data set of river water quality is formed based on the band reflectance of the reflectance image and the reflectance of the extended band; A characteristic band is obtained based on the spectral index independent variable data set and the water quality monitoring parameter.

4. The method for inverting river water quality parameters according to claim 1, characterized in that: The reflectivity image is band expanded, and the reflectivity of the expanded band is obtained using the following formula: Among them, x i and x j Represent the reflectivity of the i-th and j-th bands respectively, x i+j 、x i-j 、x i×j 、x i / j and x i，j Indicates the reflectivity of the extended band.

5. The method for inverting river water quality parameters according to claim 1, characterized in that: Generating characteristic bands based on the spectral index independent variable data set and the water quality monitoring parameters comprises the following steps: Calculate the Pearson correlation coefficient between the variables in the spectral index independent variable data set and the total phosphorus concentration and total nitrogen concentration in the water quality monitoring parameters Where R P and R N represents the Pearson correlation coefficient between the ith variable and the total phosphorus concentration and total nitrogen concentration in the water quality monitoring parameters, respectively, i represents the value of the i-th variable, Represents x i The mean of y iP and iN represent the i-th total phosphorus concentration and total nitrogen concentration in the water quality monitoring parameters respectively, and Respectively represent y iP and iN The mean of , n represents the number of variables; The bands corresponding to the variables whose Pearson correlation coefficients are all greater than a preset threshold are selected as feature bands.

6. The method for inverting river water quality parameters according to claim 1, characterized in that: Based on the water quality parameter inversion model, obtaining the total phosphorus and total nitrogen content of the water body of the river to be monitored according to the reflectivity image corresponding to the characteristic band includes the following steps: Constructing a water quality parameter inversion model; the water quality parameter inversion model is obtained by optimizing a multi-objective support vector regression model based on a genetic algorithm; The reflectivity image corresponding to the characteristic band is input into the water quality parameter inversion model to obtain the total phosphorus and total nitrogen content of the water body of the river to be monitored output by the water quality parameter inversion model.

7. The method for inverting river water quality parameters according to claim 6, characterized in that: The construction of water quality parameter inversion model includes the following steps: Construct a multi-objective support vector regression model; Calculate the penalty factor of the multi-objective support vector regression model and the real value of the kernel parameter in the radial basis kernel function based on the genetic algorithm; Substitute the real value into the multi-objective support vector regression model to obtain the water quality parameter inversion model.

8. A river water quality parameter inversion system, characterized in that: The system comprises a first acquisition module, a second acquisition module, a third acquisition module and an inversion module; The first acquisition module is used to acquire remote sensing images and water quality monitoring parameters of the river to be monitored; The second acquisition module is used to acquire the reflectivity image of the river area to be monitored based on the remote sensing image; The third acquisition module is used to acquire characteristic bands based on the reflectivity image and the water quality monitoring parameters; The inversion module is used to obtain the total phosphorus and total nitrogen content of the water body of the river to be monitored based on the water quality parameter inversion model and the reflectivity image corresponding to the characteristic band.

9. An electronic device, characterized in that: The electronic device comprises: a processor and a memory; The memory is used to store computer programs; The processor is used to execute the computer program stored in the memory so that the electronic device executes the river water quality parameter inversion method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by an electronic device, the river water quality parameter inversion method described in any one of claims 1 to 7 is implemented.

Images