Soil Heavy Metal Pollution Assessment and Early Warning System Based on Hyperspectral Feature Extraction

Abstract

The present invention discloses a soil heavy metal pollution assessment and early warning system based on hyperspectral feature extraction, which relates to the technical field of soil pollution assessment. The system discloses a spectral regular scanning module, a primary pollution early warning module, a simulation model construction module, and a secondary pollution early warning module. The spectral regular scanning module and the primary pollution early warning module are set up to deeply evaluate the heavy metal pollution of the soil by means of hyperspectral feature extraction, and the primary pollution early warning is started in time after the evaluation is abnormal. The simulation model construction module and the secondary pollution early warning module are set up to analyze various parameters of the soil through hyperspectral features, and then construct a soil change simulation model to comprehensively and three-dimensionally analyze the subsequent heavy metal pollution changes of the soil. The secondary pollution early warning is started in time after the analysis is abnormal. Through two-level different pollution early warning mechanisms, the heavy metal pollution of the soil is accurately evaluated from different angles.

Description

Technical Field

[0001] The present invention relates to the technical field of soil pollution assessment, and more specifically, it relates to a soil heavy metal pollution assessment and early warning system based on hyperspectral feature extraction. Background Art

[0002] With the acceleration of industrialization and urbanization, the problem of soil heavy metal pollution has become increasingly serious, posing a serious threat to human health, the ecological environment, and agricultural production. Traditional methods for assessing soil heavy metal pollution mainly rely on laboratory analysis. Although this method is accurate, it is time-consuming, costly, and difficult to achieve large-area and real-time monitoring. Therefore, it is particularly important to develop an efficient, accurate, and real-time soil heavy metal pollution assessment and early warning system.

[0003] In recent years, hyperspectral remote sensing technology has attracted much attention because it can obtain fine spectral information of surface substances. Hyperspectral data contains rich information on material composition and physical and chemical properties. Through specific data processing and analysis methods, rapid and accurate assessment of soil heavy metal pollution can be achieved. However, at present, most of the soil heavy metal pollution assessment methods based on hyperspectral features are still in the preliminary exploration stage, with problems such as low assessment accuracy and imperfect early warning mechanisms.

[0004] In order to overcome the deficiencies of the prior art, the present invention proposes a soil heavy metal pollution assessment and early warning system based on hyperspectral feature extraction. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a soil heavy metal pollution assessment and early warning system based on hyperspectral feature extraction.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] A soil heavy metal pollution assessment and early warning system based on hyperspectral feature extraction, including a spectral regular scanning module, a primary pollution early warning module, a simulation model construction module, and a secondary pollution early warning module;

[0008] The spectral regular scanning module is used to set the spectral scanning period. Whenever the periodic node of the spectral scanning period is reached, a hyperspectral imager is used to scan the target soil area to obtain hyperspectral image data of the target soil area;

[0009] The primary pollution early warning module obtains a primary assessment index of heavy metal pollution based on the hyperspectral image data, and determines whether to initiate a primary pollution early warning based on the comparison result between the primary assessment index of heavy metal pollution and the primary assessment threshold index of heavy metal pollution;

[0010] After starting the primary pollution warning, the simulation model construction module constructs a soil change simulation model;

[0011] The secondary pollution warning module controls the soil change simulation model to simulate the soil changes in a spectral scanning period, and then obtains the secondary evaluation index of heavy metal pollution. Based on the comparison result between the secondary evaluation index of heavy metal pollution and the secondary evaluation threshold index of heavy metal pollution, it determines whether to start the secondary pollution warning.

[0012] Furthermore, the primary evaluation index of heavy metal pollution is obtained based on the hyperspectral image data. Specifically, the hyperspectral image data is subjected to feature extraction to obtain hyperspectral features, the primary evaluation model of heavy metal pollution is obtained, the hyperspectral features are used as the input data of the primary evaluation model of heavy metal pollution, and the primary evaluation index of heavy metal pollution is output by the primary evaluation model of heavy metal pollution.

[0013] Furthermore, after starting the primary pollution warning, a soil change simulation model is constructed. Specifically, a soil parameter analysis model for various soil parameters is obtained, the hyperspectral features are sequentially input into the soil parameter analysis models for various soil parameters, and then various soil parameters are obtained. Based on various soil parameters, a soil change simulation model is constructed.

[0014] Furthermore, the soil change simulation model simulates the soil changes in a spectral scanning period, and then obtains the secondary evaluation index of heavy metal pollution. Specifically, c pollution index nodes at the same time intervals are set within the spectral scanning period. During the simulation of the soil change process, whenever a pollution index node is reached, the primary evaluation index of heavy metal pollution and the comprehensive evaluation index of soil parameters at this pollution index node are obtained, and then the heavy metal pollution change index at this pollution index node is obtained. All the heavy metal pollution change indexes are sorted in the order of the pollution index nodes. The previous heavy metal pollution change index and the next heavy metal pollution change index after sorting are compared. When the previous heavy metal pollution change index is less than the next heavy metal pollution change index, the number of pollution aggravation change times is increased by one, and the number of pollution aggravation change times is marked as Xtgh. The two adjacent heavy metal pollution change indexes after sorting are summed to obtain the continuous index of heavy metal pollution change. Set the continuous index of heavy metal pollution change. When the continuous index of heavy metal pollution change is greater than or equal to the continuous index of heavy metal pollution change, the number of continuous pollution change times is increased by one, and the number of continuous pollution change times is marked as Kawt. Using the formula The secondary evaluation index BNk of heavy metal pollution is obtained, where qa is the pollution aggravation change coefficient and qb is the continuous pollution change coefficient.

[0015] Further, the heavy metal pollution change index of the pollution index node is obtained through the following steps: Obtain the primary heavy metal pollution assessment index Mbd and the comprehensive soil parameter assessment index Seg of the pollution index node, and use the formula to obtain the heavy metal pollution change index BTws of the pollution index node, where d1 is the first index coefficient, d2 is the second index coefficient, and d3 is the third index coefficient.

[0016] Further, the comprehensive soil parameter assessment index of the pollution index node is obtained through the following steps: When reaching a pollution index node, obtain various soil parameters of the target soil area entity, obtain the characterization promotion value Bph of various soil parameters of the target soil area entity, p = 1, 2,..., P, p represents the type of soil parameter, and P is the total number of soil parameter types. Set the characterization promotion coefficient as Rh, h = 1, 2, 3,..., h, R1 < R2 < R3 <... < Rh, obtain the characterization promotion gap mean value Fdz, and use the formula to obtain the comprehensive soil parameter assessment index Seg of the pollution index node.

[0017] Further, the characterization promotion gap mean value Fdz is obtained through the following steps: Match all the characterization promotion values in pairs to form a characterization promotion group, calculate the difference between the two characterization promotion values in the characterization promotion group and take the absolute value to obtain the characterization promotion gap value of the characterization promotion group, sum up the characterization promotion gap values of all the characterization promotion groups and take the mean value to obtain the characterization promotion gap mean value Fdz.

[0018] Further, to obtain the characterization promotion value of various soil parameters of the target soil area entity, specifically: Obtain the characterization promotion models corresponding to various soil parameters, input various soil parameters into the corresponding characterization promotion models respectively, and then output the characterization promotion values of various soil parameters.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] Set the spectral regular scanning module and the primary pollution warning module, deeply evaluate the heavy metal pollution of the soil by means of hyperspectral feature extraction, and start the primary pollution warning in time after the evaluation is abnormal. Set the simulation model construction module and the secondary pollution warning module, analyze various parameters of the soil through hyperspectral features, and then construct a soil change simulation model to comprehensively and three-dimensionally analyze the subsequent heavy metal pollution changes of the soil. Start the secondary pollution warning in time after the analysis is abnormal. Through two-level different pollution warning mechanisms, accurately evaluate the heavy metal pollution of the soil from different angles, and provide decision support with a longer time window for soil pollution prevention and control. Description of the Drawings

[0021] Figure 1 It is a system module diagram of a soil heavy metal pollution assessment and early warning system based on hyperspectral feature extraction;

[0022] Figure 2 It is a system operation flow chart of a soil heavy metal pollution assessment and early warning system based on hyperspectral feature extraction. Specific implementation manner

[0023] Refer to Figure 1 - Figure 2 , a soil heavy metal pollution assessment and early warning system based on hyperspectral feature extraction, including a spectral regular scanning module, a primary pollution early warning module, a simulation model construction module, and a secondary pollution early warning module.

[0024] Spectral regular scanning module: Set the spectral scanning period (when the system is running, the spectral scanning period loops infinitely). Whenever the period node of the spectral scanning period is reached, use a hyperspectral imager to scan the target soil area to obtain hyperspectral image data of the target soil area.

[0025] Primary pollution early warning module: Extract features from the hyperspectral image data to obtain hyperspectral features, obtain a primary heavy metal pollution assessment model, use the hyperspectral features as input data for the primary heavy metal pollution assessment model, and the primary heavy metal pollution assessment model outputs a primary heavy metal pollution assessment index. Set a primary heavy metal pollution assessment threshold index (the primary heavy metal pollution assessment threshold index is a preset value of the system). When the primary heavy metal pollution assessment index is greater than or equal to the primary heavy metal pollution assessment threshold index, initiate primary pollution early warning. When the primary heavy metal pollution assessment index is less than the primary heavy metal pollution assessment threshold index, no processing is performed.

[0026] The construction method of the primary heavy metal pollution assessment model is as follows: Collect multiple hyperspectral features, construct a deep learning model, use the hyperspectral features as training data for the deep learning model, assign a primary heavy metal pollution assessment index to each training data. The index range of the primary heavy metal pollution assessment index is (0.1 - 2.0). The closer the primary heavy metal pollution assessment index is to 2.0, the more serious the assessment of heavy metal pollution. The closer the primary heavy metal pollution assessment index is to 0.1, the less serious the assessment of heavy metal pollution. Divide the training data into a training set, a validation set, and a test set according to a set ratio of 3:2:1, and train the training set, the validation set, and the test set. After training is completed, the primary heavy metal pollution assessment model is constructed.

[0027] Simulation model construction module: After initiating primary pollution early warning, obtain a soil parameter analysis model for various soil parameters (soil parameters include soil acid-base parameters, soil composition parameters, heavy metal distribution parameters, etc.), input the hyperspectral features into the soil parameter analysis models for various soil parameters in sequence, and then obtain various soil parameters. Based on various soil parameters, construct a soil change simulation model.

[0028] Each soil parameter corresponds to an analysis method. Therefore, each soil parameter corresponds to a soil parameter analysis model. In the specific implementation manner of the present invention, the construction method of the soil parameter analysis model corresponding to the soil acid-base parameter will be disclosed: collect multiple hyperspectral features, construct a deep learning model, use the hyperspectral features as the training data of the deep learning model, assign a soil acid-base parameter to each training data, divide the training data into a training set, a validation set, and a test set according to the set ratio of 4:1:1, train the training set, the validation set, and the test set, and after the training is completed, construct the soil parameter analysis model of the soil acid-base parameter.

[0029] For the soil parameter analysis model of the soil component parameter, assign a soil component parameter to each training data.

[0030] Secondary pollution warning module: The soil change simulation model simulates the soil change in a spectral scanning period, and then obtains the secondary evaluation index of heavy metal pollution. Set the secondary evaluation threshold index of heavy metal pollution (the secondary evaluation threshold index of heavy metal pollution is a preset value of the system). When the secondary evaluation index of heavy metal pollution is greater than or equal to the secondary evaluation threshold index of heavy metal pollution, start the secondary pollution warning. When the secondary evaluation index of heavy metal pollution is less than the secondary evaluation threshold index of heavy metal pollution, do not make any treatment.

[0031] The soil change simulation model simulates the soil change in a spectral scanning period, and then obtains the secondary evaluation index of heavy metal pollution. Specifically: set c pollution index nodes at the same time intervals in the spectral scanning period. During the simulation of the soil change process, whenever a pollution index node is reached, obtain the primary evaluation index of heavy metal pollution and the comprehensive evaluation index of soil parameters at this pollution index node, and then obtain the heavy metal pollution change index at this pollution index node. Arrange all the heavy metal pollution change indexes in the order of the pollution index nodes. Compare the previous heavy metal pollution change index with the next one after sorting. When the previous heavy metal pollution change index is greater than or equal to the next one, do not make any treatment. When the previous heavy metal pollution change index is less than the next one, increase the pollution intensification change times by one, and mark the pollution intensification change times as Xtgh. Sum the two adjacent heavy metal pollution change indexes after sorting to obtain the heavy metal pollution change continuous index. Set the heavy metal pollution change continuous index (the heavy metal pollution change continuous index is a preset value of the system). When the heavy metal pollution change continuous index is greater than or equal to the heavy metal pollution change continuous index, increase the pollution change continuous times by one. When the heavy metal pollution change continuous index is less than the heavy metal pollution change continuous index, do not make any treatment. Mark the pollution change continuous times as Kawt. Use the formula Obtain the secondary evaluation index BNk of heavy metal pollution, where qa is the pollution intensification change coefficient, qb is the pollution change continuity coefficient, the value of qa is 0.98, and the value of qb is 0.92.

[0032] The heavy metal pollution change index of the pollution index node is obtained through the following steps: Obtain the primary evaluation index Mbd of heavy metal pollution and the comprehensive evaluation index Seg of soil parameters of this pollution index node, and use the formula To obtain the heavy metal pollution change index BTws of this pollution index node, where d1 is the first index coefficient, d2 is the second index coefficient, d3 is the third index coefficient, the value of d1 is 0.81, the value of d2 is 1.39, and the value of d3 is 0.68.

[0033] The comprehensive evaluation index of soil parameters of the pollution index node is obtained through the following steps: When reaching a pollution index node, obtain various soil parameters of the target soil area entity, obtain the characterization promotion value Bph of various soil parameters of the target soil area entity, p = 1, 2,..., P, p represents the type of soil parameter, and P is the total number of soil parameter types. Set the characterization promotion coefficient as Rh, h = 1, 2, 3,..., h, R1 < R2 < R3 <... < Rh, and each characterization promotion coefficient Rh corresponds to a range of characterization promotion values Bph. The range of the characterization promotion value Bph includes (0, Bp1], (Bp1, Bp2],..., (Bph - 1, Bph]. When Bph ∈ (0, Bp1], the characterization promotion coefficient is R1. Match all the characterization promotion values in pairs to form a characterization promotion group. Calculate the difference between the two characterization promotion values in the characterization promotion group and take the absolute value to obtain the characterization promotion gap value of this characterization promotion group. Sum up the characterization promotion gap values of all the characterization promotion groups and take the average value to obtain the characterization promotion gap average value Fdz, and use the formula To obtain the comprehensive evaluation index Seg of soil parameters of this pollution index node.

[0034] Obtain the characterization promotion values of various soil parameters of the target soil area entity, specifically: Obtain the characterization promotion models corresponding to various soil parameters, and input various soil parameters into the corresponding characterization promotion models respectively, and then output the characterization promotion values of various soil parameters.

[0035] Each soil parameter corresponds to a characterization promotion model. The difference between the parsing models of different soil parameters lies only in the different training data. In the specific implementation manner of the present invention, the construction method of the characterization promotion model corresponding to the soil acid-base parameter will be disclosed: collect multiple soil acid-base parameters, construct a deep learning model, use the soil acid-base parameters as the training data of the deep learning model, assign a characterization promotion value to each training data, and the value range of the characterization promotion value is (1.1~3.0). The closer the characterization promotion value is to 3.0, the more the soil acid-base parameter promotes the development of heavy metal pollution. The closer the characterization promotion value is to 1.1, the more the soil acid-base parameter inhibits the development of heavy metal pollution. Divide the training data into a training set, a validation set, and a test set according to the set ratio of 5:2:1, train the training set, the validation set, and the test set. After the training is completed, the characterization promotion model of the soil acid-base parameter is constructed.

[0036] If it is to construct the characterization promotion model of the soil component parameter, then collect multiple soil component parameters. The closer the characterization promotion value is to 3.0, the more the soil component parameter promotes the development of heavy metal pollution. The closer the characterization promotion value is to 1.1, the more the soil component parameter inhibits the development of heavy metal pollution.

[0037] Based on various soil parameters, construct a soil change simulation model. Specifically: select a dedicated soil simulation software, create a target soil area entity in the simulation software, and add corresponding parameters to the target soil area entity based on various soil parameters, such as soil acid-base parameters, soil component parameters, heavy metal distribution parameters, etc. The soil change simulation model can simulate the subsequent changes of the target soil area.

[0038] Set a spectral regular scanning module and a primary pollution warning module to deeply evaluate the heavy metal pollution of the soil by means of hyperspectral feature extraction, and start the primary pollution warning in time after the evaluation is abnormal. Set a simulation model construction module and a secondary pollution warning module to analyze various parameters of the soil through hyperspectral features, and then construct a soil change simulation model to comprehensively and three-dimensionally analyze the subsequent heavy metal pollution changes of the soil. Start the secondary pollution warning in time after the analysis is abnormal. Through two-level different pollution warning mechanisms, accurately evaluate the heavy metal pollution of the soil from different angles, and provide decision support with a longer time window for soil pollution prevention and control.

[0039] The above formulas are all dimensionless and take their numerical values for calculation. The formula is a formula obtained by collecting a large amount of data and performing software simulation to get the closest to the real situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0040] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable 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 contains a set of one or more available media. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0041] It should be understood that in various embodiments of the present application, the sequence numbers of the above processes do not imply the order of execution, and the order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0042] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0043] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0044] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections between each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0045] If the above functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the essence of the technical solution of the present application, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs, and other various media that can store program codes.

[0046] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. Soil heavy metal pollution assessment and early warning system based on hyperspectral feature extraction, characterized in that: It includes spectrum periodic scanning module, primary pollution warning module, simulation model building module and secondary pollution warning module; The spectral periodic scanning module is used to set the spectral scanning cycle, and whenever the period node of the spectral scanning cycle is reached, the target soil area is scanned by a hyperspectral imager to obtain the hyperspectral image data of the target soil area; The primary pollution warning module obtains a primary assessment index of heavy metal pollution based on the hyperspectral image data, and determines whether to initiate a primary pollution warning based on a comparison result of the primary assessment index of heavy metal pollution with a primary assessment threshold index of heavy metal pollution; The simulation model building module builds a soil change simulation model after the primary pollution warning is initiated; The secondary pollution warning module controls the soil change simulation model to simulate the soil change of a spectral scanning cycle, thereby obtaining a secondary assessment index of heavy metal pollution, and determines whether to start a secondary pollution warning based on the comparison result of the secondary assessment index of heavy metal pollution and the secondary assessment threshold index of heavy metal pollution; The soil change simulation model simulates the soil change of a spectral scanning cycle, and then obtains the secondary assessment index of heavy metal pollution. Specifically, c pollution index nodes with the same time interval are set in the spectral scanning cycle. In the process of simulating soil changes, whenever a pollution index node is reached, the primary assessment index of heavy metal pollution and the comprehensive assessment index of soil parameters of the pollution index node are obtained, and then the heavy metal pollution change index of the pollution index node is obtained. All heavy metal pollution change indexes are sorted in the order of the pollution index nodes, and the previous heavy metal pollution change index adjacent to the sorting is compared with the next heavy metal pollution change index. When the previous heavy metal pollution change index is less than the next heavy metal pollution change index, the number of pollution aggravation changes is increased by one, and the number of pollution aggravation changes is marked as Xtgh. The two adjacent heavy metal pollution change indexes after sorting are summed to obtain the continuous index of heavy metal pollution change, and the continuous index of heavy metal pollution change is set. When the continuous index of heavy metal pollution change is greater than or equal to the continuous index of heavy metal pollution change, the continuous number of pollution changes is increased by one, and the continuous number of pollution changes is marked as Kawt. The formula is used. The secondary assessment index of heavy metal pollution BNk is obtained, where qa is the coefficient of pollution aggravation change and qb is the coefficient of pollution change continuity; The heavy metal pollution change index of the pollution index node is obtained by the following steps: obtain the heavy metal pollution primary assessment index Mbd and the soil parameter comprehensive assessment index Seg of the pollution index node, and use the formula The heavy metal pollution change index BTws of the pollution index node is obtained, where d1 is the first exponential coefficient, d2 is the second exponential coefficient, and d3 is the third exponential coefficient.

2. The soil heavy metal pollution assessment and early warning system based on hyperspectral feature extraction according to claim 1 is characterized in that: A primary assessment index of heavy metal pollution is obtained based on hyperspectral image data, specifically: feature extraction is performed on the hyperspectral image data to obtain hyperspectral features, a primary assessment model for heavy metal pollution is obtained, the hyperspectral features are used as input data of the primary assessment model for heavy metal pollution, and the primary assessment model for heavy metal pollution outputs the primary assessment index for heavy metal pollution.

3. The soil heavy metal pollution assessment and early warning system based on hyperspectral feature extraction according to claim 1 is characterized in that: After the primary pollution warning is initiated, a soil change simulation model is constructed, specifically: a soil parameter analysis model of various soil parameters is obtained, the hyperspectral features are input into the soil parameter analysis model of various soil parameters in turn, and then various soil parameters are obtained, and a soil change simulation model is constructed based on various soil parameters.

4. The soil heavy metal pollution assessment and early warning system based on hyperspectral feature extraction according to claim 1 is characterized in that: The characterization promotion values ​​of various soil parameters of the target soil area entity are obtained, specifically: the characterization promotion models corresponding to various soil parameters are obtained, various soil parameters are input into the corresponding characterization promotion models respectively, and then the characterization promotion values ​​of various soil parameters are outputted.

Images