A surface water pollution feature extraction method and related equipment

By extracting the spectral data and sediment characteristics of surface water, combining the differences in aliasing coefficient and sediment density, confidence correction is made, which solves the complexity and accuracy of surface water pollution components detection in the prior art, and realizes the different polluted components step by step detection and accurate grading.

CN118866180BActive Publication Date: 2024-11-26陕西省环境监测中心站
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411329063.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-24
Publication Date
2024-11-26
Estimated Expiration
2044-09-24

AI Technical Summary

Technical Problem

The existing surface water pollution feature extraction technology is complex in operation, high in cost, and it is difficult to monitor the content distribution changes of polluted components. It is impossible to accurately judge the pollution levels of different polluted components, resulting in complex data processing and inability to achieve step-by-step detection.

Method used

By collecting spectral data of surface water, extracting the spectral characteristic bands of different polluted components, determining the aliasing coefficient and concentration decision interval, combining the sediment density differences of polluted sediments and aliasing interference amount, confidence correction is made to achieve different polluted components Step by step test.

Benefits of technology

It improves the purity and grading accuracy of pollution characteristics of polluted components, and can accurately judge the pollution level among multiple polluted components, so as to achieve step-by-step detection of different polluted components in surface water.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118866180B_ABST
    Figure CN118866180B_ABST
Patent Text Reader

Abstract

The present application provides a method and related equipment for extracting surface water pollution characteristics, which extracts the spectral characteristic bands of polluting components and determines the aliasing coefficient of the aliasing bands in the spectral characteristic bands; determines the concentration decision interval of the polluting components based on the aliasing coefficient and concentration information, and makes decision classification on the spectral characteristic bands of the polluting components through the concentration decision interval to obtain the characteristic decision level; determines the sedimentation density difference of the polluting components through the diffusion characteristics of the polluted sediments and the turbidity information of the water source samples, and then determines the aliasing interference amount of the polluting components through the sedimentation density difference and the aliasing coefficient; makes confidence correction on the characteristic decision level through the aliasing interference amount to obtain the confidence pollution level, and detects the polluting components in the surface water step by step according to the confidence pollution level to obtain the water pollution characteristics. By adopting the scheme of the present application, the pollution levels of different polluting components can be judged in the multivariate polluting components to realize the step-by-step detection of different polluting components in the surface water.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of surface water monitoring technology, and more specifically, to a surface water pollution feature extraction method and related equipment. Background Art

[0002] Pollution feature extraction refers to the process of identifying and extracting characteristic information related to pollutants from complex environmental data. By analyzing spectra, chemical composition or physical parameters, the presence and concentration of pollutants in environmental samples can be separated, identified and quantified. It can effectively distinguish and analyze the characteristics of different pollution components, thereby providing support for environmental monitoring and pollution assessment.

[0003] In surface water pollution monitoring, feature extraction usually relies on physical and chemical analysis, spectral analysis and remote sensing technology. By measuring the concentration or other physical and chemical properties of various pollutants in water samples, combined with characteristic peaks such as absorption, reflection, and fluorescence in spectral analysis technology, pollutants can be identified and quantitatively analyzed. However, the existing surface water pollution feature extraction technology is complex to operate, costly, and difficult to monitor changes in the content distribution of pollutants. The spectral data of the same pollutant component is easily affected by the aliasing interference of other pollutants, resulting in complex data processing and inability to perform environmental adjustments for specific pollutants. How to judge the pollution levels of different pollutants among multiple pollutants to achieve step-by-step detection of different pollutants in surface water has become a difficult problem faced by the industry. Summary of the invention

[0004] The present application provides a surface water pollution feature extraction method and related equipment, which can determine the pollution levels of different pollution components among multiple pollution components to achieve step-by-step detection of different pollution components in surface water.

[0005] In a first aspect, the present application provides a method for extracting surface water pollution features, comprising the following steps:

[0006] Collecting water source samples of surface water in the target area, and then obtaining spectral data of different pollutant components in the water source samples;

[0007] Extracting the bands of the spectral data according to the difference characteristics of different polluting components in the water source sample to obtain the spectral characteristic bands of different polluting components, and then determining the aliasing coefficients of the aliasing bands between different polluting components in all the spectral characteristic bands;

[0008] Determine the concentration decision interval of each pollutant based on the aliasing coefficient and the concentration information of all pollutants in the water source sample, and then make decisions and classify the spectral characteristic bands of different pollutants through each concentration decision interval to obtain the characteristic decision level of each pollutant;

[0009] Obtaining a breakdown spectrum of the polluted sediment in the water source sample, determining the difference in deposition density of different polluting components in the polluted sediment through the diffusion characteristics of the polluted sediment and the turbidity information of the water source sample, and then determining the aliasing interference amount of different polluting components in the breakdown spectrum on each spectral characteristic band through all the deposition density differences and the aliasing coefficient;

[0010] For each pollutant component, the spectral characteristic band of the pollutant component is confidence corrected according to the aliasing interference amount of the pollutant component, and then the confidence pollution level of the characteristic decision level of each pollutant component is obtained. Then, based on all the confidence pollution levels, the different pollutants in the surface water of the target area are detected step by step to obtain the water pollution characteristics corresponding to each confidence pollution level.

[0011] In some embodiments, band extraction is performed on the spectral data based on the difference characteristics of different pollutants in the water source sample to obtain the spectral characteristic bands of different pollutants, specifically including:

[0012] Determine the differential characteristics of different pollution components;

[0013] Mapping the spectral data into a characteristic spectrum graph;

[0014] The characteristic spectrum graph is divided according to various difference features to obtain different characteristic spectrum segments;

[0015] Each characteristic spectral band is subjected to noise reduction processing to obtain the spectral characteristic bands of different pollution components.

[0016] In some embodiments, determining the concentration decision interval of each pollutant based on the aliasing coefficient and the concentration information of all pollutants in the water source sample specifically includes:

[0017] Acquire the concentration information of all pollutants in the water source sample and the aliased bands in each spectral characteristic band;

[0018] Determine the decision combination characteristics of different pollution components through each aliased band and the concentration information;

[0019] Based on the aliasing coefficient, all decision combination features are predicted and decided to obtain decision values ​​of different pollution components;

[0020] Determine the concentration decision interval of each pollutant component based on all decision values.

[0021] In some embodiments, the spectral characteristic bands of different pollution components are classified by decision according to each concentration decision interval, and the characteristic decision level of each pollution component is obtained, which specifically includes:

[0022] Obtain decision values ​​of different pollution components from each concentration decision interval;

[0023] According to each decision value, the spectral characteristic bands of different pollution components are smoothed to obtain smooth characteristic bands of different pollution components;

[0024] The characteristic decision level of each pollution component is determined by all smoothed characteristic bands and all decision values.

[0025] In some embodiments, determining the difference in deposition density of different pollutant components in the polluted sediments by using the diffusion characteristics of the polluted sediments and the turbidity information of the water source sample specifically includes:

[0026] determining dispersion characteristics of the contaminated sediment;

[0027] Obtaining turbidity information of a water source sample, and then determining relative densities of different pollutant components based on the turbidity information;

[0028] The difference in deposition density of different pollutant components in the polluted sediment is determined based on the respective relative densities and the diffusion characteristics.

[0029] In some embodiments, determining the aliasing interference amount of different pollution components in the breakdown spectrum on each spectral characteristic band based on all deposition density differences and the aliasing coefficient specifically includes:

[0030] Standardizing all the deposition density differences and the aliasing coefficients according to the average value of all the deposition density differences to obtain standard aliasing coefficients and all the standard deposition density differences;

[0031] Determine the characteristic values ​​of the influence of different pollution components on the characteristic bands of the spectrum through the differences of various standard deposition densities and the standard aliasing coefficients;

[0032] Perform interference identification on each spectral characteristic band based on each influencing characteristic value and the breakdown spectrum to obtain each interference band in the interference identification process;

[0033] The fluctuation entropy of the corresponding spectral characteristic band is determined by each interference band;

[0034] The aliasing interference amount of different pollution components on each spectral characteristic band is determined by each fluctuation entropy and the corresponding impact characteristic value.

[0035] In some embodiments, for each pollution component, the confidence correction is performed on the spectral characteristic band of the pollution component according to the aliasing interference amount of the pollution component, and then the confidence pollution level of the characteristic decision level of each pollution component is obtained, which specifically includes:

[0036] Obtain the spectral characteristic band and characteristic decision level of each pollution component;

[0037] For each pollution component, correcting the characteristic peak value in the spectral characteristic band of the pollution component according to the aliasing interference amount of the pollution component, thereby obtaining a plurality of corrected characteristic peak values ​​of each pollution component;

[0038] Determine a characteristic peak matrix of all pollution components according to a plurality of corrected characteristic peaks of each pollution component and all characteristic decision levels;

[0039] The confidence pollution level of the characteristic decision level of each pollution component is determined by the characteristic peak matrix.

[0040] In a second aspect, the present application provides a surface water pollution feature extraction system, comprising:

[0041] An acquisition module is used to collect water source samples of surface water in a target area, and then obtain spectral data of different pollutant components in the water source samples;

[0042] A processing module, used for performing band extraction on the spectral data according to the difference characteristics of different polluting components in the water source sample, obtaining spectral characteristic bands of different polluting components, and then determining aliasing coefficients of aliasing bands between different polluting components in all spectral characteristic bands;

[0043] The processing module is further used to determine the concentration decision interval of each pollutant component based on the aliasing coefficient and the concentration information of all pollutants in the water source sample, and then perform decision classification on the spectral characteristic bands of different pollutants through each concentration decision interval to obtain the characteristic decision level of each pollutant component;

[0044] The processing module is further used to obtain a breakdown spectrum of the polluted sediment in the water source sample, determine the difference in deposition density of different polluting components in the polluted sediment through the diffusion characteristics of the polluted sediment and the turbidity information of the water source sample, and then determine the aliasing interference amount of different polluting components in the breakdown spectrum on each spectral characteristic band through all the deposition density differences and the aliasing coefficient;

[0045] The execution module is used to perform confidence correction on the spectral characteristic band of each pollutant component according to the aliasing interference amount of the pollutant component, and then obtain the confidence pollution level of the characteristic decision level of each pollutant component, and then detect the different pollutants in the surface water of the target area step by step according to all the confidence pollution levels, and obtain the water pollution characteristics corresponding to each confidence pollution level.

[0046] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the above-mentioned surface water pollution feature extraction method.

[0047] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned surface water pollution feature extraction method is implemented.

[0048] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects:

[0049] In the surface water pollution feature extraction method and related equipment provided by the present application, firstly, a water source sample of the surface water in the target area is collected, and then the spectral data of different polluting components in the water source sample are obtained; secondly, the spectral data is band extracted according to the difference characteristics of different polluting components in the water source sample to obtain the spectral characteristic bands of different polluting components, and then the aliasing coefficients of the aliasing bands between different polluting components in all the spectral characteristic bands are determined; then, based on the aliasing coefficient and the concentration information of all polluting components in the water source sample, the concentration decision interval of each polluting component is determined, and then the spectral characteristic bands of different polluting components are decided and graded according to each concentration decision interval to obtain the characteristic decision level of each polluting component. ; Then, the breakdown spectrum of the polluted sediments in the water source sample is obtained, and the sediment density differences of different polluting components in the polluted sediments are determined through the diffusion characteristics of the polluted sediments and the turbidity information of the water source sample, and then the aliasing interference amounts of different polluting components in the breakdown spectrum on each spectral characteristic band are determined through all the sediment density differences and the aliasing coefficient; finally, for each polluting component, the spectral characteristic band of the polluting component is confidence corrected according to the aliasing interference amount of the polluting component, and then the confidence pollution level of the characteristic decision level of each polluting component is obtained, and then the different polluting components in the surface water of the target area are detected step by step according to all the confidence pollution levels, and the water pollution characteristics corresponding to each confidence pollution level are obtained.

[0050] It can be seen that the present application uses the aliasing interference amount to make confidence corrections to the characteristic decision level, obtains the confidence pollution level of each pollutant component, and then detects the different pollutants in the surface water of the target area step by step based on the confidence pollution level, and obtains the water pollution characteristics corresponding to each confidence pollution level; firstly, the spectral characteristic bands are classified by decision through the concentration decision interval, and the characteristic decision level of each pollutant component can be obtained, wherein the characteristic decision level of the pollutant component can reflect the characteristic intensity of each pollutant component within a specific spectral range, which is conducive to accurately analyzing the pollution level of each pollutant component through the characteristic decision level; secondly , the aliasing interference amount is obtained by combining the influencing characteristic value and fluctuation entropy of the pollutant component. The aliasing interference amount can be used to analyze the degree of aliasing interference of different pollutants on the spectrum; finally, the aliasing interference amount is used to perform confidence correction on the feature decision level to obtain the confidence pollution level of each pollutant component, and the confidence pollution level of different pollutants can be obtained, which is beneficial to improve the purity of the pollution characteristics of each pollutant component when extracting surface water pollution characteristics while completing the pollution characteristic classification of multiple pollutants; in summary, the present application scheme can judge the pollution levels of different pollutants in multiple pollutant components to realize the step-by-step detection of different pollutants in surface water. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is an exemplary flow chart of a surface water pollution feature extraction method according to some embodiments of the present application;

[0052] Figure 2 is a schematic diagram of a process for determining a spectral characteristic band according to some embodiments of the present application;

[0053] Figure 3 is a schematic diagram of a process for determining a feature decision level according to some embodiments of the present application;

[0054] Figure 4 is a schematic diagram of exemplary hardware and / or software of a surface water pollution feature extraction system according to some embodiments of the present application;

[0055] Figure 5 It is a structural schematic diagram of a computer device applying a surface water pollution feature extraction method according to some embodiments of the present application. DETAILED DESCRIPTION

[0056] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 , which is an exemplary flow chart of a surface water pollution feature extraction method according to some embodiments of the present application. The surface water pollution feature extraction method 100 mainly includes the following steps:

[0057] In step 101, a water source sample of surface water in a target area is collected, and then spectral data of different pollutant components in the water source sample are obtained.

[0058] In specific implementation, collecting source water samples of surface water in the target area and then obtaining spectral data of different polluting components in the source water samples can be achieved in the following manner, namely: collecting source water samples of surface water in the target area through a transparent container in a surface water pollution detection device; secondly, using a near-infrared spectrometer to test the polluting components in the source water samples, and then using the test data obtained from the test as spectral data of different polluting components in the surface water, wherein the spectral data includes multiple types of polluting component data, and the polluting component data consists of component type, band length, and intensity (concentration), and the spectral data can be used to detect different components of pollutants and the concentrations of each component.

[0059] It should be noted that the surface water pollution detection device in the present application is a portable water quality detector, which is equipped with a transparent container. In the present application, three 5000ml transparent containers are used to simultaneously collect water source samples of surface water in the same target area, wherein the three water source samples are used to extract spectral data, flow pollutant analysis and contaminated sediment analysis, respectively.

[0060] In step 102, band extraction is performed on the spectral data based on the difference characteristics of different polluting components in the water source sample to obtain spectral characteristic bands of different polluting components, and then the aliasing coefficients of aliasing bands between different polluting components in all spectral characteristic bands are determined.

[0061] In some embodiments, reference Figure 2 As shown, this figure is a schematic diagram of the process of determining the spectral characteristic bands shown in some embodiments of the present application. In this embodiment, the spectral data is subjected to band extraction based on the difference characteristics of different pollutant components in the water source sample, and the spectral characteristic bands of different pollutant components can be obtained by the following steps:

[0062] In step 1021, different characteristics of different pollution components are determined;

[0063] In step 1022, the spectral data is mapped into a characteristic spectrum graph;

[0064] In step 1023, the characteristic spectrum graph is divided according to each difference feature to obtain different characteristic spectrum segments;

[0065] In step 1024, each characteristic spectral band is subjected to noise reduction processing to obtain spectral characteristic bands of different pollution components.

[0066] In specific implementation, determining the difference characteristics of different polluting components can be achieved in the following manner, namely: first, obtaining historical pollution data and surface water environmental standards of the target area, and using the historical pollution data and the surface water environmental standards to pre-train the machine learning model to obtain a pre-trained model; secondly, using Raman spectroscopy in the prior art to determine the different polluting components in the surface water of the target area, and using the random forest algorithm to mark each polluting component separately to obtain each marked polluting component; finally, using the pre-trained model to analyze and identify each marked polluting component to obtain the difference characteristics of different polluting components. As a preferred embodiment, a decision tree algorithm can be used when pre-training the machine learning model using the historical pollution data and the surface water environmental standards. The decision tree algorithm can be used to improve the credibility of the pre-trained model in predicting the difference characteristics of different polluting components. In other embodiments, other existing algorithms can also be used to pre-train the machine learning model, which is not limited here.

[0067] It should be noted that the difference characteristics in this application refer to the spectral characteristics that distinguish different pollution components. The difference characteristics include characteristic peaks, absorption band positions, and intensities of the spectrum. The difference characteristics can reflect the optical properties of different pollutants within a specific wavelength range.

[0068] In specific implementation, mapping the spectral data into a characteristic spectrum graph can be achieved in the following manner, namely: the spectral clustering method in the prior art can be used to map the spectral data into a spectral space, and then all linear waves and nonlinear waves in the spectral space are used as characteristic spectrum graphs; the characteristic spectrum graph is divided by various difference features to obtain different characteristic spectrum segments, which can be achieved in the following manner, namely: characteristic spectrum segments refer to spectral bands of pollution components (including other pollution components), and each characteristic spectrum band can be extracted from the characteristic spectrum graph by wavelet transform in the prior art, namely: for each pollution component, the difference characteristics of the pollution component are used as the scale quantity of the wavelet transform, and all characteristic wavelets in the characteristic spectrum graph are extracted by wavelet transform after determining the scale quantity, and then the spectral segment composed of all characteristic wavelets is used as the characteristic spectrum segment of the pollution component, thereby obtaining the characteristic spectrum segment of each pollution component.

[0069] It should be noted that the spectral characteristic bands in the present application refer to the spectral bands of different pollution components, and the spectral characteristic bands include the peak value, position and intensity of the pollution components; in specific implementation, each characteristic spectral band is subjected to noise reduction processing, and the spectral characteristic bands of different pollution components can be obtained by the following method, namely: each characteristic spectral band is subjected to Gaussian smoothing processing respectively to reduce the high-frequency noise in each characteristic spectral band, so that each characteristic spectral band after smoothing processing is used as the spectral characteristic band of the corresponding pollution component, wherein the spectral characteristic bands of different pollution components can be used to identify and distinguish different pollution components, which is conducive to the subsequent qualitative analysis of different pollution components, and each spectral characteristic band also contains spectral data of other pollution components.

[0070] In some embodiments, determining the aliasing coefficients of aliasing bands between different pollution components in all spectral feature bands may be implemented by the following steps:

[0071] Selecting a spectral characteristic band as a selected spectral characteristic band;

[0072] Determine all characteristic peaks in the selected spectral characteristic band;

[0073] Determine the aliased bands in the selected spectral characteristic bands according to each characteristic peak value;

[0074] Determine the aliasing degree in the selected spectral characteristic band by the correlation coefficient between each spectral data in the aliasing band;

[0075] Continue to determine the aliasing degree of the remaining spectral characteristic bands;

[0076] The aliasing coefficients of the aliased bands between different pollution components are determined based on all aliasing degrees.

[0077] In specific implementation, determining all characteristic peaks in the selected spectral feature band can be achieved in the following manner, namely: detecting the selected spectral feature band through the Fourier transform algorithm in the prior art to obtain all characteristic peaks in the selected spectral feature band, wherein the use of the Fourier transform algorithm can effectively reduce the impact of noise in the selected spectral feature on the detection process, and the characteristic peak detected refers to the peak in the spectral feature band that is more capable of representing the corresponding pollution component.

[0078] In addition, in a specific implementation, the following steps can be used to determine the aliasing band in the selected spectral characteristic band according to each characteristic peak value, namely: intercepting the middle 80% of the partial band of the selected spectral characteristic band, and taking the average value of all characteristic peak values ​​in the partial band as the average peak value, for all two adjacent characteristic peaks in the selected spectral characteristic band, obtaining the wavelength between the two characteristic peaks, and calculating the sum of all peak values ​​between the two characteristic peaks, and then taking the ratio of the sum to the wavelength as the judgment value of the spectral characteristic band between the two adjacent characteristic peaks, and then obtaining the judgment value between all two adjacent characteristic peaks, and then extracting each adjacent characteristic peak value between the two adjacent characteristic peaks whose judgment value is greater than the average peak value, and then taking the spectral characteristic band between each adjacent characteristic peak value extracted as the aliasing band.

[0079] In a specific implementation, determining the aliasing degree in the selected spectral feature band by the correlation coefficient between each spectral data in the aliasing band can be achieved in the following manner, namely: obtaining all spectral data in the aliasing band, determining the average value of all spectral data, calculating the covariance between two adjacent spectral data, substituting the average value and each covariance into the calculation formula of the correlation coefficient, and obtaining the correlation coefficient of all adjacent spectral data in the aliasing band, and then listing the spectral data with a correlation coefficient greater than a preset threshold as approximately irrelevant spectral data, and then using the information entropy between all approximately irrelevant spectral data as the selected spectral feature band. The aliasing degree, wherein the preset threshold value can be obtained by analyzing the correlation coefficients of different pollution components according to historical experimental data, that is, the correlation coefficients of different pollution components can be calculated by the autoregressive analysis model in the prior art; the aliasing coefficients of the aliasing bands between different pollution components based on all aliasing degrees can be determined in the following way, that is, the aliasing degree is an indicator to measure the degree of aliasing of pollution components in the spectral characteristic bands, that is, the more aliasing bands there are in the spectral characteristic bands with larger aliasing degrees, and the fewer aliasing bands there are in the spectral characteristic bands with smaller aliasing degrees, and the standard deviation of all aliasing degrees can be used as the aliasing coefficients of the aliasing bands between different pollution components.

[0080] It should be noted that the aliasing coefficient in the present application is an indicator to measure the degree of confusion of the pollution components in the spectral characteristic band. The aliasing coefficient can reflect the degree of overlap and interference between the spectral signals of different pollution components in each spectral characteristic band. The aliasing coefficient is determined by calculating the correlation coefficient between the spectral data of each pollution component in the aliasing band. The aliasing coefficient can represent the aliasing degree of the pollution components in each spectral characteristic band, that is, the higher the aliasing coefficient, the more aliasing bands are contained in the spectral data between different pollution components, and the smaller the aliasing coefficient, the fewer aliasing bands are contained in the spectral data between different pollution components.

[0081] In step 103, the concentration decision interval of each pollutant is determined based on the aliasing coefficient and the concentration information of all pollutants in the water source sample, and then the spectral characteristic bands of different pollutants are graded according to each concentration decision interval to obtain the characteristic decision level of each pollutant.

[0082] In some embodiments, determining the concentration decision interval of each pollutant based on the aliasing coefficient and the concentration information of all pollutants in the water source sample can be implemented by the following steps:

[0083] Acquire the concentration information of all pollutants in the water source sample and the aliased bands in each spectral characteristic band;

[0084] Determine the decision combination characteristics of different pollution components through each aliased band and the concentration information;

[0085] Based on the aliasing coefficient, all decision combination features are predicted and decided to obtain decision values ​​of different pollution components;

[0086] Determine the concentration decision interval of each pollutant component based on all decision values.

[0087] In specific implementation, the concentration information of all polluting components in the water source sample can be obtained in the following manner, namely: using the spectrophotometric method in the prior art to respectively detect the concentrations of different polluting components in the water source samples in the target area, and sorting the concentration data of each polluting component from large to small, thereby obtaining a concentration hierarchy sequence of different polluting components, performing principal component analysis on the concentration hierarchy sequence, extracting each principal component value and the average principal component value in the principal component analysis process, and hierarchically dividing all concentrations in the concentration hierarchy sequence according to each principal component value and the average principal component value, for example: taking the concentration corresponding to a principal component value higher than the average principal component value as a high-level concentration, and taking the concentration corresponding to a principal component value lower than the average principal component value as a low-level concentration, thereby taking all concentrations after hierarchical division as the concentration information of all polluting components in the water source sample.

[0088] It should be noted that the concentration hierarchy sequence in the present application refers to a sequence formed by sorting the concentration data of pollutants in water source samples, and the concentration information refers to the distribution information of the concentrations of different pollutants in water source samples. The concentration information can evaluate the impact of different pollutants on the ecological environment, thereby determining the pollutants that need to be treated first.

[0089] In specific implementation, the decision combination features of different pollution components can be determined by each aliasing band and the concentration information in the following manner, namely: obtaining a concentration level sequence corresponding to the concentration information, and mapping the concentration level sequence to a spectral image to obtain a concentration level spectrum, obtaining aliasing bands in each spectral feature band, and mapping the aliasing bands in each spectral feature band to the concentration level spectrum, thereby using the overlapping bands in each spectral feature band and the concentration level spectrum as the decision combination features of the corresponding pollution components.

[0090] In addition, in the specific implementation, predictive decisions are made on all decision combination features based on the aliasing coefficient, and the decision values ​​of different pollution components can be obtained in the following way, namely: all the bands in the concentration level spectrum are used as training data to initialize a prediction model, and the aliasing coefficient is used as a constraint parameter of the prediction model, and then the prediction model analyzes each decision combination feature separately to obtain the decision value of each pollution component, wherein using all the bands in the concentration level spectrum to train the prediction model is beneficial to improving the credibility of the prediction model, and using the aliasing coefficient as the constraint parameter of the prediction model. The constraint parameters can filter out the interfering bands of other polluting components in the decision combination characteristics; the concentration decision interval of each polluting component can be determined according to all decision values ​​in the following manner, namely: the decision value refers to the edge values ​​for distinguishing different polluting components. In the concentration level spectrum, the decision value is at the edge of each band with obvious differences, which plays a decision-making role in distinguishing different polluting components. Therefore, for each polluting component, when the difference between two adjacent decision values ​​of the polluting component is the smallest, the two adjacent decision values ​​are used as the upper and lower limits of the concentration decision interval of the polluting component, respectively, to obtain the concentration decision interval of each polluting component.

[0091] It should be noted that the concentration decision interval in the present application refers to the range interval for distinguishing the concentrations of different pollutants in water source samples. The concentration decision interval is used to determine the corresponding bands of each pollutant in the spectrum. The concentration decision interval can be used to reduce the noise generated by the presence of other pollutants in each spectral characteristic band, thereby reducing the probability of band aliasing.

[0092] In some embodiments, reference Figure 3 As shown, this figure is a schematic diagram of the process of determining the characteristic decision level shown in some embodiments of the present application. In this embodiment, the spectral characteristic bands of different pollution components are classified by each concentration decision interval, and the characteristic decision level of each pollution component can be obtained by the following steps:

[0093] In step 1031, decision values ​​of different pollution components are obtained from each concentration decision interval;

[0094] In step 1032, the spectral characteristic bands of different pollution components are smoothed according to each decision value to obtain smoothed characteristic bands of different pollution components;

[0095] In step 1033, the feature decision level of each pollution component is determined by all smooth feature bands and all decision values.

[0096] In specific implementation, the spectral characteristic bands of different polluting components are smoothed according to each decision value, and the smoothed characteristic bands of different polluting components can be obtained in the following manner, namely: for each spectral characteristic band, the moving average method in the prior art can be used to smooth the spectral data in the spectral characteristic band that is equal to the corresponding decision value, and the smoothed spectral characteristic band is used as the smoothed characteristic band to obtain the smoothed characteristic bands of different polluting components, wherein the fluctuation of the spectral data corresponding to the decision value in the spectral characteristic band can be reduced by smoothing, thereby reducing the noise generated by the spectral data of other polluting components in the spectral characteristic band, which is conducive to distinguishing and extracting different polluting components in water source samples.

[0097] In addition, in the specific implementation, the characteristic decision level of each pollution component can be determined by all the smooth characteristic bands and all the decision values. The following method is adopted, namely: all the smooth characteristic bands are combined into a correlation matrix, namely: each smooth characteristic band is used as a row of the matrix, and the composed matrix is ​​used as the correlation matrix, and the mean of all decision values ​​is used as the average decision value, the correlation matrix is ​​normalized by the average decision value, and all the normalized smooth characteristic bands are sorted in the order of the spectral intensity mean from small to large to obtain a hierarchical correlation matrix, so that each row major sequence in the hierarchical correlation matrix is ​​used as the characteristic decision level of the corresponding pollution component, wherein the row major sequence starts from 1, and each row of the hierarchical correlation matrix corresponds to a smooth characteristic band, and each smooth characteristic band corresponds to a pollution component.

[0098] It should be noted that the feature decision level in the present application is a decision indicator for measuring the pollution levels of different pollution components, that is, the pollution components with higher feature decision levels can be judged as higher pollution levels, and the pollution components with lower feature decision levels can be judged as lower pollution levels. Each feature decision level corresponds to a pollution component, and the pollution level of each pollution component in complex spectral data can be accurately identified through the feature decision level.

[0099] In step 104, a breakdown spectrum of the contaminated sediments in the water source sample is obtained, and the deposition density differences of different pollutant components in the contaminated sediments are determined through the diffusion characteristics of the contaminated sediments and the turbidity information of the water source sample. Then, the aliasing interference amount of different pollutant components in the breakdown spectrum on each spectral characteristic band is determined based on all the deposition density differences and the aliasing coefficient.

[0100] In a specific implementation, obtaining the breakdown spectrum of the contaminated sediment in the water source sample can be achieved in the following manner, namely: first, obtaining a water source sample of surface water in the target area, subjecting the water source sample to a static treatment, and extracting a contaminated sediment mixture (containing a small amount of water) from a transparent container after the static treatment, and drying the contaminated sediment mixture to obtain contaminated sediment; secondly, ablating the contaminated sediment using a high-energy pulse laser in the prior art, and transmitting specific light radiation emitted by the plasma during the ablation process to a spectrometer; finally, extracting the breakdown spectrum of the contaminated sediment from the spectrometer; wherein the contaminated sediment mixture refers to contaminated sediment formed by the contaminated components in the water source sample after deposition, and the contaminated sediment refers to a solid sample formed after the contaminated sediment is dried.

[0101] It should be noted that, in this embodiment, the standing time for the standing treatment of the water source sample can be set between 24 and 48 hours, and the standing environment can be kept lightless and constant temperature to avoid external factors interfering with the sedimentation process. The temperature during low-temperature drying needs to be controlled between 40 and 60°C to avoid damaging the organic components in the sediment. During the standing treatment process, a laser particle size analyzer is used to measure and record the particle size data and sedimentation rate of the contaminated sediment to facilitate subsequent analysis of the particles and distribution of the deposited pollutants.

[0102] In some embodiments, determining the difference in deposition density of different pollutant components in the polluted sediments by using the diffusion characteristics of the polluted sediments and the turbidity information of the water source sample can be achieved by the following steps:

[0103] determining dispersion characteristics of the contaminated sediment;

[0104] Obtaining turbidity information of a water source sample, and then determining relative densities of different pollutant components based on the turbidity information;

[0105] The difference in deposition density of different pollutant components in the polluted sediment is determined based on the respective relative densities and the diffusion characteristics.

[0106] In specific implementation, the diffusion characteristics of the polluted sediment can be determined in the following manner, namely: obtaining particle size data of the polluted sediment, the particle size data refers to the particle radius of the deposited pollutants, using Matplotlib in the prior art to convert the particle size data into a particle size distribution diagram, obtaining the deposition velocity of the polluted sediment, and using Matplotlib to write the deposition velocity into the particle size distribution diagram, and then using the numerical value in the fused particle size distribution diagram as the diffusion characteristics of the polluted sediment, wherein the diffusion characteristics are characteristic values ​​that describe the diffusion of the polluted sediment, and the diffusion characteristics include the deposition velocity and particle size data of different polluting components in the polluted sediment, and the diffusion characteristics are helpful in analyzing the influence of the polluted sediment on the corresponding spectral characteristic bands of each polluting component.

[0107] In addition, in a specific implementation, the turbidity information of the water source sample can be obtained in the following manner, namely: a turbidity detector (or turbidity sensor) can be used to shake the water source sample for detection, and the turbidity value of the water source sample after shaking is recorded, so as to use the turbidity value obtained by the detection as the turbidity information of the water source sample, wherein the turbidity value is a floating point data, and two decimal places are retained during detection, and the unit of the turbidity value is NTU. Obtaining the turbidity information of the water source sample can analyze the degree of obstruction caused by the water source sample to the passage of light, determine the scattering of light by suspended matter and the absorption of light by solute molecules in the water source sample, and is conducive to judging the influence of polluted sediments on spectral data, thereby improving the accuracy of feature extraction of different pollution components.

[0108] In specific implementation, the relative density of different polluting components determined according to the turbidity information can be achieved in the following manner, namely: the linear regression model in the prior art can be used to perform regression analysis on the turbidity information to obtain the concentrations of different polluting components, and the concentrations of each polluting component are respectively used as the relative density of the corresponding polluting component, wherein the regression analysis model can be established through historical turbidity detection experimental data; the difference in deposition density of different polluting components in the polluted sediment based on each relative density and the diffusion characteristics can be achieved in the following manner, namely: the deposition velocity and all particle size data in the diffusion characteristics are extracted, for each polluting component, the Stokes' law in the prior art is used, and the deposition velocity and particle size corresponding to the polluting component are substituted into the Stokes' law formula, so that the calculated result is used as the deposition density of the polluting component, and then the deposition density of each polluting component is obtained, and then the difference between the deposition density of each polluting component and the mean of the deposition density of all polluting components is used as the deposition density difference of each polluting component.

[0109] It should be noted that the sedimentation density difference in the present application refers to the difference in density between different polluting components in the water source sample when they are deposited. The sedimentation density difference is caused by the different particle sizes and chemical properties of different polluting components. By calculating the sedimentation density difference of each polluting component, it is helpful to monitor the concentration and distribution of different polluting components in the water body and facilitate the analysis of deposited pollutants.

[0110] In some embodiments, determining the aliasing interference amount of different pollution components in the breakdown spectrum on each spectral characteristic band based on all deposition density differences and the aliasing coefficient can be achieved by using the following steps:

[0111] Standardizing all the deposition density differences and the aliasing coefficients according to the average value of all the deposition density differences to obtain standard aliasing coefficients and all the standard deposition density differences;

[0112] Determine the characteristic values ​​of the influence of different pollution components on the characteristic bands of the spectrum through the differences of various standard deposition densities and the standard aliasing coefficients;

[0113] Perform interference identification on each spectral characteristic band based on each influencing characteristic value and the breakdown spectrum to obtain each interference band in the interference identification process;

[0114] The fluctuation entropy of the corresponding spectral characteristic band is determined by each interference band;

[0115] The aliasing interference amount of different pollution components on each spectral characteristic band is determined by each fluctuation entropy and the corresponding impact characteristic value.

[0116] In specific implementation, all deposition density differences and the aliasing coefficient are standardized according to the difference average of all deposition density differences, and the standard aliasing coefficient and all standard deposition density differences can be obtained by the following method, namely: taking the mean of all deposition density differences as the difference average, initializing a standardized model, taking the difference average as the constraint parameter of the standardized model, and standardizing all deposition density differences and the aliasing coefficient through the standardized model with the constraint parameter set, so as to obtain the standard aliasing coefficient and all standard deposition density differences.

[0117] In addition, in the specific implementation, the influence characteristic values ​​of different pollution components on the spectral characteristic bands are determined by the differences in each standard deposition density and the standard aliasing coefficient, which can be implemented in the following manner, namely: for each pollution component, the product of the standard deposition density difference corresponding to the pollution component and the standard aliasing coefficient can be used as the influence characteristic value corresponding to the pollution component, thereby obtaining the influence characteristic value of each pollution component on the spectral characteristic band, wherein the influence characteristic value refers to the degree of interference of the pollution component on the spectral characteristic band; interference identification is performed on each spectral characteristic band based on each influence characteristic value and the breakdown spectrum, and each interference band in the interference identification process can be implemented in the following manner, namely: for each pollution component, the breakdown spectrum peak value with the same influence characteristic value of the pollution component is extracted from the breakdown spectrum, and then the breakdown spectrum band composed of all the extracted breakdown spectrum peaks is used as the influence band, and the influence band of the pollution component is mapped to the corresponding spectral characteristic band to obtain the interference mapping band, the different bands between the interference mapping band and the spectral characteristic band are extracted, and the combined band of all the extracted bands is used as the interference band of the pollution component, thereby obtaining the interference band of each pollution component.

[0118] In specific implementation, the fluctuation entropy of the corresponding spectral characteristic band determined by each interference band can be achieved in the following manner, namely: the fluctuation entropy refers to the uncertainty measure of the fluctuation of the spectral data in the interference band under the interference of the influencing characteristic value, and the peak values ​​in each interference band can be described by entropy characteristics through the information entropy in the prior art to obtain the fluctuation entropy of each interference band; the aliasing interference amount of different pollution components on each spectral characteristic band determined by each fluctuation entropy and the corresponding influencing characteristic value can be achieved in the following manner, namely: the fluctuation entropy can be used to quantify the interference complexity of different pollution components in the spectral characteristic band, which is conducive to identifying and distinguishing pollution components. The influencing characteristic value includes the concentration and particle size of the pollution component, which can be used to analyze the influence of the pollution component in the spectral characteristic band. Therefore, the product of the influencing characteristic value of each pollution component and the fluctuation entropy can be used as the aliasing interference amount of the corresponding pollution component on each spectral characteristic band. The aliasing interference amount is conducive to analyzing the relationship between the influencing characteristic value and the fluctuation entropy of each pollution component.

[0119] It should be noted that the aliasing interference amount in the present application is an indicator to measure the degree of interference caused by different polluting components on the spectral characteristic bands, that is, the larger the aliasing interference amount, the greater the interference degree of the corresponding polluting component on the spectral characteristic bands, and the more aliasing of other polluting components exists in the spectral characteristic bands; the smaller the aliasing interference amount, the smaller the interference degree of the corresponding polluting component on the spectral characteristic bands, and the less aliasing of other polluting components exists in the spectral characteristic bands. The aliasing interference amount combines the influence characteristic value and fluctuation entropy of the polluting component. Therefore, the interference effect of the polluting component can be fully reflected through the aliasing interference amount, thereby improving the reliability of spectral analysis.

[0120] In step 105, for each pollutant component, the spectral characteristic band of the pollutant component is confidence corrected according to the aliasing interference amount of the pollutant component, and then the confidence pollution level of the characteristic decision level of each pollutant component is obtained. Then, based on all the confidence pollution levels, different pollutants in the surface water of the target area are detected step by step to obtain the water pollution characteristics corresponding to each confidence pollution level.

[0121] In some embodiments, for each pollution component, confidence correction is performed on the spectral characteristic band of the pollution component according to the aliasing interference amount of the pollution component, and then the confidence pollution level of the characteristic decision level of each pollution component is obtained, which can be achieved by the following steps:

[0122] Obtain the spectral characteristic band and characteristic decision level of each pollution component;

[0123] For each pollution component, correcting the characteristic peak value in the spectral characteristic band of the pollution component according to the aliasing interference amount of the pollution component, thereby obtaining a plurality of corrected characteristic peak values ​​of each pollution component;

[0124] Determine a characteristic peak matrix of all pollution components according to a plurality of corrected characteristic peaks of each pollution component and all characteristic decision levels;

[0125] The confidence pollution level of the characteristic decision level of each pollution component is determined by the characteristic peak matrix.

[0126] In specific implementation, the characteristic peak in the spectral characteristic band of the pollutant component is corrected according to the aliasing interference amount of the pollutant component, and multiple corrected characteristic peaks of the pollutant component are obtained in the following manner, namely: all characteristic peaks and all peaks in the spectral characteristic band of the pollutant component are obtained, the average value of all peaks is taken as the peak mean, and then the product of the aliasing interference amount of the pollutant component and the peak mean is taken as the correction amount, and the difference between each characteristic peak and the correction amount is taken as the multiple corrected characteristic peaks of the pollutant component, thereby obtaining each corrected characteristic peak in each spectral characteristic band.

[0127] In addition, in the specific implementation, the characteristic peak matrix of all pollution components can be determined according to the multiple corrected characteristic peaks of each pollution component and all the characteristic decision levels. It can be achieved in the following way, namely: for each pollution component, use the unique hot encoding in the prior art to convert all the corrected characteristic peaks of the pollution component into a characteristic peak vector, and use the corresponding characteristic decision level of the pollution component and the proportion of all characteristic decision levels as the weight value of the pollution component, and then use the product of the weight value of the pollution component and the characteristic peak vector as the weighted characteristic peak vector of the pollution component, and then obtain the weighted characteristic peak vector of each pollution component, and arrange all the weighted characteristic peak vectors in order from large to small according to the absolute value of the vector, and use the arranged matrix as the characteristic peak matrix of all pollution components, that is, the weighted characteristic peak vector with the largest vector absolute value is arranged in the top row of the matrix, and the weighted characteristic peak vector with the second largest vector absolute value is added. The weighted characteristic peak vector is arranged in the second row of the matrix, and other vectors are arranged in turn, and finally the weighted characteristic peak vector with the smallest vector absolute value is arranged in the bottom row of the matrix; the confidence pollution level of the characteristic decision level of each pollution component determined by the characteristic peak matrix can be implemented in the following way, namely: for each pollution component, the row major sequence corresponding to the pollution component in the characteristic peak matrix is ​​used as the confidence pollution level of the characteristic decision level of the pollution component, and then the confidence pollution level of the characteristic decision level of each pollution component is obtained, wherein the row major sequence starts from 1; it should be noted that the confidence pollution level in the present application is an indicator for measuring the pollution level of the pollution component, for example: the pollution component with a confidence pollution level of "1" has the highest pollution level, wherein the higher the pollution level, the greater the influence of the aliasing band of the spectral data of the pollution component, and different pollution components can be extracted step by step based on the confidence pollution level of each pollution component.

[0128] It should be noted that in the present application, different polluting components in the surface water of the target area are detected step by step according to all the confidence pollution levels, and the water pollution characteristics corresponding to each confidence pollution level are obtained, which means that the purification characteristics of each polluting component are detected step by step based on the confidence pollution level, and then each purification characteristic is used as the water pollution characteristic corresponding to the confidence pollution level, wherein the purification characteristic refers to the spectral characteristic after filtering out other polluting components from the spectral characteristic band of the polluting component; in specific implementation, the corrected characteristic peak can be used to replace the characteristic peak in the corresponding spectral characteristic band, and then the spectral data in the spectral characteristic band after replacing the characteristic peak can be used as the purification characteristic of the corresponding polluting component, and the purification characteristics of the corresponding polluting component are extracted in sequence from small to large according to the confidence pollution level of each polluting component, and then each purification characteristic is used as the water pollution characteristic corresponding to the confidence pollution level, which can reduce the influence of other polluting components on the spectral data, so as to reduce the aliasing data of other redundant polluting components in the spectral data of the polluting components in the surface water, which is conducive to analyzing the pollution degree of different polluting components on the surface water of the target area, so as to carry out targeted improvement and control of the water pollution of the surface water in the target area according to the pollution degree of different polluting components.

[0129] On the other hand, in some embodiments, the present application provides a surface water pollution feature extraction system, referring to Figure 4 , which is a schematic diagram of exemplary hardware and / or software of a surface water pollution feature extraction system according to some embodiments of the present application. The surface water pollution feature extraction system 400 includes: an acquisition module 401, a processing module 402 and an execution module 403, which are described as follows:

[0130] Acquisition module 401, in this application, acquisition module 401 is mainly used to collect water source samples of surface water in the target area, and then obtain spectral data of different pollutant components in the water source samples;

[0131] Processing module 402, in the present application, the processing module 402 is used to extract the spectral data by using the difference characteristics of different polluting components in the water source sample to obtain the spectral characteristic bands of different polluting components, and then determine the aliasing coefficients of the aliasing bands between different polluting components in all spectral characteristic bands;

[0132] The processing module 402 in the present application is also used to determine the concentration decision interval of each pollutant component based on the aliasing coefficient and the concentration information of all pollutants in the water source sample, and then perform decision classification on the spectral characteristic bands of different pollutants through each concentration decision interval to obtain the characteristic decision level of each pollutant component;

[0133] The processing module 402 in the present application is also used to obtain the breakdown spectrum of the polluted sediment in the water source sample, determine the sediment density difference of different pollutants in the polluted sediment through the diffusion characteristics of the polluted sediment and the turbidity information of the water source sample, and then determine the aliasing interference amount of different pollutants in the breakdown spectrum on each spectral characteristic band through all the sediment density differences and the aliasing coefficient;

[0134] Execution module 403. In the present application, execution module 403 is mainly used to perform confidence correction on the spectral characteristic band of each pollutant component according to the aliasing interference amount of the pollutant component, and then obtain the confidence pollution level of the characteristic decision level of each pollutant component, and then detect the different pollutants in the surface water of the target area step by step according to all the confidence pollution levels, and obtain the water pollution characteristics corresponding to each confidence pollution level.

[0135] In addition, the present application also provides a computer device, which includes a memory and a processor, the memory stores codes, and the processor is configured to obtain the codes and execute the above-mentioned surface water pollution feature extraction method.

[0136] In some embodiments, reference Figure 5 , which is a schematic diagram of the structure of a computer device using a surface water pollution feature extraction method according to some embodiments of the present application. The surface water pollution feature extraction method in the above embodiment can be Figure 5 The computer device 500 shown in the figure is implemented, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503 and at least one communication interface 504.

[0137] The processor 501 may be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0138] The communication bus 502 may be used to transmit information between the above-mentioned components.

[0139] The memory 503 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, an optical disc storage (including a compressed optical disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a magnetic disk or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory 503 may exist independently and be connected to the processor 501 via the communication bus 502. The memory 503 may also be integrated with the processor 501.

[0140] The memory 503 is used to store the program code for executing the solution of the present application, and the execution is controlled by the processor 501. The processor 501 is used to execute the program code stored in the memory 503. The program code may include one or more software modules. The surface water pollution feature extraction method in the above embodiment can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0141] The communication interface 504 uses any transceiver or other device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0142] In a specific implementation, as an embodiment, a computer device may include multiple processors, each of which may be a single-CPU processor or a multi-CPU processor. The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0143] The above-mentioned computer device may be a general-purpose computer device or a special-purpose computer device. In a specific implementation, the computer device may be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device or an embedded device. The embodiment of the present application does not limit the type of computer device.

[0144] In addition, the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned surface water pollution feature extraction method is implemented.

[0145] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0146] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for extracting surface water pollution characteristics, characterized in that: The steps include: Collecting water source samples of surface water in the target area, and then obtaining spectral data of different pollutant components in the water source samples; The spectral data is subjected to band extraction based on the difference characteristics of different polluting components in the water source sample to obtain spectral characteristic bands of different polluting components, and then the aliasing coefficients of aliasing bands between different polluting components in all spectral characteristic bands are determined, wherein the aliasing coefficient is an indicator to measure the degree of confusion of polluting components in spectral characteristic bands, and the aliasing coefficient can reflect the degree of mutual overlap and interference of spectral signals of different polluting components in each spectral characteristic band; Determine the concentration decision interval of each pollutant based on the aliasing coefficient and the concentration information of all pollutants in the water source sample, wherein the concentration decision interval refers to the range interval for distinguishing the concentrations of different pollutants in the water source sample, and then perform decision classification on the spectral characteristic bands of different pollutants through each concentration decision interval to obtain the characteristic decision level of each pollutant; Obtaining a breakdown spectrum of the polluted sediment in the water source sample, determining the difference in deposition density of different polluting components in the polluted sediment through the diffusion characteristics of the polluted sediment and the turbidity information of the water source sample, and then determining the aliasing interference amount of different polluting components in the breakdown spectrum on each spectral characteristic band through all the deposition density differences and the aliasing coefficient; For each pollutant component, the spectral characteristic band of the pollutant component is confidence-corrected by the aliasing interference of the pollutant component, and then the confidence pollution level of the characteristic decision level of each pollutant component is obtained. Then, according to all the confidence pollution levels, different pollutants in the surface water of the target area are detected step by step to obtain the water pollution characteristics corresponding to each confidence pollution level; The aliasing interference amount of different pollution components in the breakdown spectrum on each spectral characteristic band is determined by all the deposition density differences and the aliasing coefficient, specifically including: Standardizing all the deposition density differences and the aliasing coefficients according to the average value of all the deposition density differences to obtain standard aliasing coefficients and all the standard deposition density differences; Determine the characteristic values ​​of the influence of different pollution components on the characteristic bands of the spectrum through the differences of various standard deposition densities and the standard aliasing coefficients; Perform interference identification on each spectral characteristic band based on each influencing characteristic value and the breakdown spectrum to obtain each interference band in the interference identification process; The fluctuation entropy of the corresponding spectral characteristic band is determined by each interference band; The aliasing interference amount of different pollution components on each spectral characteristic band is determined by each fluctuation entropy and the corresponding impact characteristic value.

2. The method according to claim 1, characterized in that The spectral data is subjected to band extraction based on the difference characteristics of different polluting components in the water source sample to obtain the spectral characteristic bands of different polluting components, specifically including: Determine the differential characteristics of different pollution components; Mapping the spectral data into a characteristic spectrum graph; The characteristic spectrum graph is divided according to various difference features to obtain different characteristic spectrum segments; Each characteristic spectral band is subjected to noise reduction processing to obtain the spectral characteristic bands of different pollution components.

3. The method according to claim 1, characterized in that Determining the concentration decision interval of each pollutant component based on the aliasing coefficient and the concentration information of all pollutants in the water source sample specifically includes: Acquire the concentration information of all pollutants in the water source sample and the aliased bands in each spectral characteristic band; Determine the decision combination characteristics of different pollution components through each aliased band and the concentration information; Based on the aliasing coefficient, all decision combination features are predicted and decided to obtain decision values ​​of different pollution components; Determine the concentration decision interval of each pollutant component based on all decision values; Among them, the decision combination characteristics of different pollution components are determined by each aliasing band and the concentration information in the following manner, namely: obtaining the concentration level sequence corresponding to the concentration information, and mapping the concentration level sequence to the spectral image to obtain the concentration level spectrum, obtaining the aliasing bands in each spectral feature band, and mapping the aliasing bands in each spectral feature band to the concentration level spectrum, so as to use the overlapping bands in each spectral feature band and the concentration level spectrum as the decision combination characteristics of the corresponding pollution components.

4. The method according to claim 1, characterized in that The spectral characteristic bands of different pollution components are graded through each concentration decision interval, and the characteristic decision level of each pollution component is obtained, including: Obtain decision values ​​of different pollution components from each concentration decision interval; According to each decision value, the spectral characteristic bands of different pollution components are smoothed to obtain smooth characteristic bands of different pollution components; The characteristic decision level of each pollution component is determined by all smoothed characteristic bands and all decision values.

5. The method according to claim 1, characterized in that Determining the difference in deposition density of different pollutant components in the polluted sediments through the diffusion characteristics of the polluted sediments and the turbidity information of the water source sample specifically includes: determining dispersion characteristics of the contaminated sediment; Obtaining turbidity information of a water source sample, and then determining relative densities of different pollutant components based on the turbidity information; The difference in deposition density of different pollutant components in the polluted sediment is determined based on the respective relative densities and the diffusion characteristics.

6. The method according to claim 1, characterized in that For each pollution component, the spectral characteristic band of the pollution component is confidence-corrected by the aliasing interference amount of the pollution component, and then the confidence pollution level of the characteristic decision level of each pollution component is obtained, which specifically includes: Obtain the spectral characteristic band and characteristic decision level of each pollution component; For each pollution component, correcting the characteristic peak value in the spectral characteristic band of the pollution component according to the aliasing interference amount of the pollution component, thereby obtaining a plurality of corrected characteristic peak values ​​of each pollution component; Determine a characteristic peak matrix of all pollution components according to a plurality of corrected characteristic peaks of each pollution component and all characteristic decision levels; The confidence pollution level of the characteristic decision level of each pollution component is determined by the characteristic peak matrix.

7. A surface water pollution feature extraction system, which uses the method according to any one of claims 1 to 6 to extract surface water pollution features, characterized in that: The surface water pollution feature extraction system includes: An acquisition module is used to collect water source samples of surface water in a target area, and then obtain spectral data of different pollutant components in the water source samples; A processing module, used for performing band extraction on the spectral data according to the difference characteristics of different polluting components in the water source sample, obtaining spectral characteristic bands of different polluting components, and then determining aliasing coefficients of aliasing bands between different polluting components in all spectral characteristic bands; The processing module is further used to determine the concentration decision interval of each pollutant component based on the aliasing coefficient and the concentration information of all pollutants in the water source sample, and then perform decision classification on the spectral characteristic bands of different pollutants through each concentration decision interval to obtain the characteristic decision level of each pollutant component; The processing module is further used to obtain a breakdown spectrum of the polluted sediment in the water source sample, determine the difference in deposition density of different polluting components in the polluted sediment through the diffusion characteristics of the polluted sediment and the turbidity information of the water source sample, and then determine the aliasing interference amount of different polluting components in the breakdown spectrum on each spectral characteristic band through all the deposition density differences and the aliasing coefficient; The execution module is used to perform confidence correction on the spectral characteristic band of each pollutant component according to the aliasing interference amount of the pollutant component, and then obtain the confidence pollution level of the characteristic decision level of each pollutant component, and then detect the different pollutants in the surface water of the target area step by step according to all the confidence pollution levels, and obtain the water pollution characteristics corresponding to each confidence pollution level.

8. A computer device, comprising a memory and a processor, wherein the memory stores a code, characterized in that: The processor is configured to obtain the code and execute the surface water pollution feature extraction method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the surface water pollution feature extraction method according to any one of claims 1 to 6 is implemented.