Hyperspectral intelligent inversion method and device based on multiple time-space water quality data
By introducing multi-spatial and spatial water quality data and particle swarm optimization algorithm in the hyperspectral water quality inversion method, the shortcomings of the existing methods in monitoring uniform water bodies are solved, efficient and real-time water quality monitoring is achieved, and the accuracy of the model is improved. General use.
Patent Information
- Application Number
- CN202411530879.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-30
- Publication Date
- 2025-02-14
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing hyperspectral water quality inversion method is not applicable when monitoring water bodies with relatively uniform distribution, and there are problems such as long detection cycles, high cost and the inability to obtain water quality data in real time.
The hyperspectral intelligent inversion method based on multi-spatial-time water quality data is adopted, and the model parameters are dynamically updated using particle swarm optimization algorithm to improve the accuracy and general applicability of the model.
It improves the accuracy and general applicability of the water quality inversion model, realizes efficient and real-time monitoring of relatively uniform water bodies, and reduces detection costs and response time.
Smart Images
Figure CN119438100A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure belongs to the technical field of water quality data detection, and more specifically, to a hyperspectral intelligent inversion method and device based on multi-temporal and spatial water quality data. Background Art
[0002] Water quality testing is an important part of environmental protection and water resources management, and is of great significance for monitoring water pollution and assessing the ecological health of water bodies. Traditional water quality testing methods are usually physical and chemical analysis methods, such as laboratory physical and chemical analysis methods or analysis using national standard instruments. Although this method is accurate, it has limitations such as long testing cycles, high costs, and the inability to obtain real-time water quality data.
[0003] With the development of remote sensing technology and spectral analysis technology, hyperspectral remote sensing technology has gradually become an important means in the field of water quality monitoring because it can provide rich spectral information and spatial information. However, most of the existing hyperspectral water quality inversion methods are carried out by simultaneously collecting water samples from several representative points in the study area for laboratory analysis as the "true value" under ideal conditions. However, this type of method is only applicable to water areas with a large range and obvious water quality parameter concentration gradients, and is not applicable to long-term continuous monitoring of water bodies with relatively uniform distribution. Summary of the invention
[0004] The purpose of the present invention is to provide a hyperspectral intelligent inversion method and device based on multi-temporal and spatial water quality data to improve the accuracy and versatility of the inversion model.
[0005] A first aspect of the embodiments of the present disclosure provides a hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data, comprising: Establishing multiple initial hyperspectral data-water quality parameter relationship function models based on normalized hyperspectral data and multi-temporal and spatial water quality data within a target range; the multi-temporal and spatial water quality data includes multiple water quality parameters; updating the state of each particle based on the position of each particle and an evaluation function, wherein the evaluation function corresponds to an evaluation function for each water quality parameter, and each water quality parameter corresponds to a plurality of particles; Update the evaluation function corresponding to the water quality parameter based on the updated state of each particle; According to the comparison result between the output value of the updated evaluation function and the preset evaluation threshold, multiple initial hyperspectral data-water quality parameter relationship function models are iteratively trained to obtain multiple target hyperspectral data-water quality parameter relationship function models.
[0006] A second aspect of the disclosed embodiment provides a hyperspectral intelligent inversion device based on multi-temporal and spatial water quality data, comprising: A data acquisition module, used to establish multiple initial hyperspectral data-water quality parameter relationship function models based on normalized hyperspectral data and multi-temporal and spatial water quality data within a target range; the multi-temporal and spatial water quality data includes multiple water quality parameters; A state updating module, used for updating the state of each particle based on the position of each particle and an evaluation function, wherein the evaluation function is an evaluation function corresponding to each water quality parameter, and each water quality parameter corresponds to a plurality of particles; A data updating module, used for updating the evaluation function corresponding to the water quality parameter based on the updated state of each particle; The module training module is used to iteratively train multiple initial hyperspectral data-water quality parameter relationship function models according to the comparison result between the output value of the updated evaluation function and the preset evaluation threshold to obtain multiple target hyperspectral data-water quality parameter relationship function models.
[0007] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of the above-mentioned hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data are implemented.
[0008] According to a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data are implemented.
[0009] The beneficial effects of the hyperspectral intelligent inversion method and device based on multi-temporal and spatial water quality data provided by the embodiments of the present disclosure are: First, the present disclosure collects water quality data at different times and locations within the target range, and improves the accuracy and versatility of the relationship function model by using multi-time and space water quality data calibration. Secondly, the present disclosure introduces a particle swarm optimization algorithm, and the position and state of each particle (i.e., model coefficient) can be dynamically updated according to the evaluation function corresponding to each water quality parameter, thereby continuously optimizing the model performance. As a global optimization algorithm, the particle swarm algorithm directly searches for the optimal solution in the entire band space, is not restricted by the selection of characteristic bands, and can better adapt to different water quality conditions and environmental conditions. Finally, the present disclosure avoids the loss of potential important information caused by selecting only a part of the band for modeling by utilizing all spectral information, thereby improving the accuracy of the inversion results. In summary, the present disclosure uses a particle swarm algorithm for inversion, reduces the workload of parameter adjustment, simplifies the entire water quality inversion process, thereby improving the overall computing efficiency and reducing the response time. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0011] Figure 1 A schematic diagram of a process of a hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data provided by an embodiment of the present disclosure; Figure 2 A specific flow chart of a hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data provided by an embodiment of the present disclosure; Figure 3 A structural block diagram of a hyperspectral intelligent inversion device based on multi-temporal and spatial water quality data provided by an embodiment of the present disclosure; Figure 4 A schematic block diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION
[0012] In the following description, specific details such as specific system structures and technologies are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, it should be clear to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present disclosure with unnecessary details.
[0013] In order to make the purpose, technical solutions and advantages of the present disclosure more clear, specific embodiments will be described below in conjunction with the accompanying drawings.
[0014] Please refer to Figure 1 , Figure 1 A schematic diagram of a process of a hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data provided by an embodiment of the present disclosure, the method comprising: S101: establishing a plurality of initial hyperspectral data-water quality parameter relationship function models based on normalized hyperspectral data and multi-temporal and spatial water quality data within a target range; the multi-temporal and spatial water quality data includes a plurality of water quality parameters.
[0015] In this embodiment, the target range is a predetermined water area, which refers to a specific area corresponding to the hyperspectral data collection and water quality data analysis. The target range can be defined based on geographic location, time, or other relevant factors. For example, a technician can determine the water area to be monitored at a location of interest (a lake, river section, etc.).
[0016] Hyperspectral data is data that describes the reflection or emission characteristics of an object at different wavelengths. The hyperspectral data collected within the target range is normalized to obtain normalized hyperspectral data. Normalization refers to converting hyperspectral data into values relative to a certain standard or reference point to eliminate the influence of external factors such as lighting conditions and instrument differences, so that the data has better consistency and comparability.
[0017] Water quality data refers to various measurements or observations that describe the physical, chemical, and biological properties of water bodies. Water quality data may include, but are not limited to, multiple water quality parameters such as permanganate index, ammonia nitrogen, total phosphorus, total nitrogen, heavy metal content, and chlorophyll a. Multi-temporal and spatiotemporal water quality data refers to water quality data collected at different temporal and spatial scales. These data reflect the dynamic changes of water quality parameters over time and space. The temporal scale can include seasonal changes, daily changes, or fluctuations within shorter time periods (such as hours, minutes, seconds, etc.); the spatial scale can involve differences in different geographical locations within the target range.
[0018] In this embodiment, relevant data needs to be obtained before establishing multiple initial hyperspectral data-water quality parameter relationship function models. The normalized hyperspectral data is obtained by scanning and processing the monitoring area through a hyperspectral scanning imaging system, and the process is as follows: Determine the target range according to the location of interest, determine the installation height and inclination of the hyperspectral scanning imaging system, and place the hyperspectral host, light source, turntable, camera, etc. in the hyperspectral scanning imaging system on a high tower or column, connect the circuit and power on for testing. After the hyperspectral scanning imaging system can work normally, start the hyperspectral scanning imaging system to scan the target range to obtain hyperspectral data. After normalizing the hyperspectral data, normalized hyperspectral data is obtained.
[0019] While collecting hyperspectral data, water samples are collected at different times and locations within the target range, and sent to the laboratory for analysis to obtain multi-temporal and spatial water quality analysis data.
[0020] Researchers can explore the potential relationship between spectral features and water quality parameters based on normalized hyperspectral data and multi-temporal and spatial water quality data within the target range. By establishing multiple initial hyperspectral data-water quality parameter relationship function models, water quality parameters can be quickly and accurately inverted, enabling more efficient water quality monitoring.
[0021] S102: updating the state of each particle based on the position of each particle and an evaluation function, wherein the evaluation function is an evaluation function corresponding to each water quality parameter, and each water quality parameter corresponds to a plurality of particles.
[0022] In this embodiment, the particle swarm algorithm is selected for inversion. As a global optimization algorithm, the particle swarm algorithm directly searches for the optimal solution in the entire band space, is not limited by the selection of characteristic bands, and can better adapt to different water quality conditions and environmental conditions. By utilizing all spectral information, the potential loss of important information caused by selecting only a part of the band for modeling is avoided, thereby improving the accuracy of the inversion result.
[0023] In this embodiment, a particle is an individual or solution used in an optimization problem algorithm. They represent a point in the space of the problem solution. All particles corresponding to a water quality parameter are a set of coefficients of the initial hyperspectral data-water quality parameter relationship function model. In the particle swarm algorithm, particles iteratively update their positions and velocities to find the optimal solution.
[0024] The position of each particle refers to a point in the solution space. The evaluation function is used to evaluate the quality or merits of a solution (i.e., the position of the particle) and to guide the direction of the optimization process. Each water quality parameter corresponds to an evaluation function. The state of the particle refers to the current position of the particle during the optimization process. By iteratively updating the state of the particle, the optimal solution can be gradually approached.
[0025] S103: updating the evaluation function corresponding to the water quality parameter based on the updated state of each particle.
[0026] In this embodiment, all particles corresponding to a water quality parameter are a set of coefficients of the initial hyperspectral data-water quality parameter relationship function model. The evaluation function corresponding to the water quality parameter is used to evaluate the performance or accuracy of the model under the current particle state (i.e., the model coefficient). Specifically, the evaluation function is calculated based on the difference between the water quality parameter value predicted by the model and the actual observed value. Based on the state of each particle after the update, the evaluation function corresponding to each water quality parameter needs to be updated, which means that in each iteration, the evaluation function will be recalculated according to the new particle state (i.e., the model coefficient), and the iteration ends when the calculated value of the evaluation function is less than the preset evaluation threshold.
[0027] S104: Iteratively training a plurality of initial hyperspectral data-water quality parameter relationship function models according to a comparison result between the updated evaluation function and a preset evaluation threshold value to obtain a plurality of target hyperspectral data-water quality parameter relationship function models.
[0028] In this embodiment, the initial hyperspectral data-water quality parameter relationship function model is an initial mathematical model established based on the relationship between hyperspectral data and water quality parameters. Each initial hyperspectral data-water quality parameter relationship function model includes the normalized reflectance of different bands of the hyperspectral data, the regression coefficients corresponding to the hyperspectral data of different bands, and the concentrations of different water quality parameters. These coefficients will be adjusted during the iterative training process to optimize the performance of the model. After each iteration, the updated evaluation function value will be compared with the preset evaluation threshold. If the evaluation function value of a model on a certain water quality parameter is less than the preset threshold, then the model can be considered to have completed training and can be used for actual water quality data inversion.
[0029] From the above, it can be concluded that, firstly, the present disclosure collects water quality data at different times and locations within the target range, and improves the accuracy and versatility of the relationship function model by using multi-time and space water quality data calibration. Secondly, the present disclosure introduces the particle swarm optimization algorithm, and the position and state of each particle (i.e., the model coefficient) can be dynamically updated according to the evaluation function corresponding to each water quality parameter, thereby continuously optimizing the model performance. As a global optimization algorithm, the particle swarm algorithm directly searches for the optimal solution in the entire band space, is not restricted by the selection of characteristic bands, and can better adapt to different water quality conditions and environmental conditions. Finally, by utilizing all spectral information, we avoid the loss of potential important information caused by selecting only a part of the band for modeling, thereby improving the accuracy of the inversion results. In summary, the present disclosure uses the particle swarm algorithm for inversion, reduces the workload of parameter adjustment, simplifies the entire water quality inversion process, thereby improving the overall computing efficiency and reducing the response time.
[0030] In one embodiment of the present disclosure, the hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data further includes: normalizing the hyperspectral data collected within the target range to obtain normalized hyperspectral data.
[0031] In this embodiment, in order to ensure the consistency of the hyperspectral intelligent inversion analysis process based on multi-temporal and spatial water quality data, the spectral data related to the multi-temporal and spatial water quality data are normalized through an electronic whiteboard. In the data acquisition stage, a standard reference whiteboard with a reflectivity of 100% is first used to collect sunlight data to establish a stable and reproducible illumination benchmark so that the effects caused by changes in the incident spectrum of sunlight can be accurately corrected in subsequent data processing. Secondly, a black camera lens cap is covered to obtain a completely black calibration image, and the reference whiteboard is scanned to obtain a completely white calibration image to eliminate the effects of changes in the incident spectrum of sunlight at different times.
[0032] From the above, it can be concluded that this embodiment uses an electronic whiteboard to normalize spectral data, and uses a standard reference whiteboard and calibration images for data acquisition and correction, which can ensure the consistency of the multi-temporal and spatial hyperspectral water quality inversion analysis process, improve the accuracy and reliability of the data, and lay a solid foundation for subsequent analysis and modeling.
[0033] In one embodiment of the present disclosure, multiple initial hyperspectral data-water quality parameter relationship function models are established based on normalized hyperspectral data and multi-temporal and spatial water quality data within a target range, including: The initial hyperspectral data-water quality parameter relationship function model is established according to the first formula. The first formula is:
[0034] in, is the concentration of water quality parameters, are all constants; is the normalized reflectivity of the ith band at time t, and n is a constant; is the reflectivity of the lens corrected for the i-th band at the geographic coordinates (x, y) at time t, is the reflectance data of the all-black calibration image of the ith band at the geographic coordinates (x, y) at time t. It is the whiteboard reflectance data of the i-th band at the geographic coordinates (x, y) at time t.
[0035] In this embodiment, water quality parameters include permanganate index, ammonia nitrogen, total phosphorus, total nitrogen, heavy metal content, chlorophyll a, etc. Each water quality parameter can establish an initial hyperspectral data-water quality parameter relationship function model through the first formula.
[0036] From the above, it can be concluded that this embodiment uses an electronic whiteboard to normalize spectral data, and uses a standard reference whiteboard and calibration images for data acquisition and correction, which can ensure the consistency of the multi-temporal and spatial hyperspectral water quality inversion analysis process and improve the accuracy and reliability of the data.
[0037] In one embodiment of the present disclosure, the evaluation function corresponding to each water quality parameter is calculated according to the second formula; The second formula is:
[0038] Among them, F is the evaluation function, The concentration of water quality parameters obtained by analyzing water quality samples within the target range, for example, can be obtained by performing physical and chemical analysis of water quality samples collected in multiple time and space; is the water quality parameter concentration inverted by the initial hyperspectral data-water quality parameter relationship function model, is a constant, n is a constant, is the normalized reflectivity of the ith band at the coordinate (x, y) at time t, The inversion results are fitted by establishing a regression relationship between normalized reflectance and water quality parameter data. For the remainder.
[0039] In this embodiment, the second formula is simplified as follows:
[0040] in, is the reflectivity of the lens corrected for the i-th band at the geographic coordinates (x, y) at time t, is the reflectance data of the all-black calibration image of the ith band at the geographic coordinates (x, y) at time t. It is the whiteboard reflectance data of the i-th band under the geographic coordinates (x, y) at time t.
[0041] In one embodiment of the present disclosure, updating the state of each particle based on the position of each particle and the evaluation function includes: Determine the individual optimal position of each particle based on the position of each particle and the evaluation function; Determine the global optimal position of the particle corresponding to each water quality parameter based on the individual optimal position of each particle; The state of each particle is updated based on the individual optimal position and the global optimal position.
[0042] In this embodiment, all particles corresponding to a water quality parameter represent a set of coefficients When flying in n+1 dimensional space, the position of each particle can be initialized using historical experience values.
[0043] In the process of updating the particle state, the individual optimal position of each particle is first determined based on the position of each particle and the evaluation function corresponding to the water quality parameter. Each water quality parameter corresponds to n+1 particles. The position with the smallest evaluation function value among all positions of a single particle during its flight is taken as the individual optimal position of the single particle. The position with the smallest evaluation function value among the individual optimal positions of all n+1 particles is taken as the global optimal position of the n+1 particles corresponding to each water quality parameter.
[0044] At each iteration, each particle updates its state according to its individual optimal and global optimal positions as follows:
[0045] Where w is the inertia weight, and are different accelerating particles, representing the speeds at which the particles are controlled to approach the individual optimal position and the global optimal position respectively; and is a random number uniformly distributed in the interval [0, 1]. represents the position of the nth particle at the Tth iteration, represents the velocity of the nth particle at the Tth iteration, represents the historical individual optimal position recorded by the nth particle at the Tth iteration, satisfying ; represents the global optimal position recorded by n+1 particles, satisfying .
[0046] In one embodiment of the present disclosure, after the state of each particle is updated, the method further includes: The individual optimal position of each particle and the global optimal position of the particle corresponding to each water quality parameter are updated according to the state of each particle.
[0047] In this embodiment, the individual optimal position of each particle and the global optimal position of the particle corresponding to each water quality parameter are updated according to Formula 5. The global optimal position of the particle corresponding to each water quality parameter is the global optimal position of all particles (n+1 particles) corresponding to each water quality parameter.
[0048] The fifth formula is as follows: like
[0049] but
[0050] otherwise
[0051]
[0052] refer to Figure 2 , Figure 2The specific flow chart of the hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data is shown in the figure. First, a hyperspectral scanning imaging system is deployed within the target range to obtain multi-temporal and spatial hyperspectral data and normalize the hyperspectral data. At the same time, water samples at different times and locations within the target range are collected and analyzed to obtain multi-temporal and spatial water quality data. The processing methods include but are not limited to laboratory physical and chemical analysis methods and methods of analysis using instruments and equipment. Secondly, multiple hyperspectral data-water quality parameter relationship function models of different bands are established based on the normalized hyperspectral data and multi-temporal and spatial water quality data (water quality data obtained by analyzing water quality samples collected at different times and locations within the target range). Each particle represents a set of coefficients of the relationship function. The particle position can be initialized with an empirical value, and then the particle state is updated according to the individual optimal and global optimal positions. The comprehensive deviation between the true value (data obtained by collection) and the measured value (data obtained by model inversion) of multi-temporal and spatial water quality data is used as the evaluation function, and the evaluation function is calculated using multi-temporal and spatial water quality data. If the evaluation function is less than the preset evaluation threshold, a set of optimal relationship function coefficients is output; if the output of the evaluation function is greater than or equal to the preset evaluation threshold, the particle state is updated according to the individual optimal and global optimal positions, and the model iterative optimization continues until the output of the evaluation function is less than the preset evaluation threshold, and a set of optimal relationship function coefficients is output to obtain the relationship function of hyperspectral intelligent inversion (i.e., the target hyperspectral data-water quality parameter relationship function model).
[0053] Corresponding to the hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data in the above embodiment, Figure 3 This is a block diagram of a hyperspectral intelligent inversion device based on multi-temporal and spatial water quality data provided by an embodiment of the present disclosure. For ease of explanation, only the parts related to the embodiment of the present disclosure are shown. Figure 3 The hyperspectral intelligent inversion device 20 based on multi-temporal and spatial water quality data includes: a data acquisition module 21, a state update module 22, a data update module 23 and a module training module 24.
[0054] The data acquisition module 21 is used to establish multiple initial hyperspectral data-water quality parameter relationship function models based on the normalized hyperspectral data and multi-temporal and spatial water quality data within the target range; the multi-temporal and spatial water quality data includes multiple water quality parameters; A state updating module 22, for updating the state of each particle based on the position of each particle and an evaluation function, wherein the evaluation function is an evaluation function corresponding to each water quality parameter, and each water quality parameter corresponds to a plurality of particles; A data updating module 23, used for updating the evaluation function corresponding to the water quality parameter based on the updated state of each particle; The module training module 24 is used to iteratively train multiple initial hyperspectral data-water quality parameter relationship function models according to the comparison result between the output value of the updated evaluation function and the preset evaluation threshold to obtain multiple target hyperspectral data-water quality parameter relationship function models.
[0055] In one embodiment of the present disclosure, the data acquisition module 21 is further used to: The hyperspectral data collected within the target range is normalized to obtain normalized hyperspectral data.
[0056] In one embodiment of the present disclosure, the data acquisition module 21 is specifically used for: The initial hyperspectral data-water quality parameter relationship function model is established according to the first formula. The first formula is:
[0057] in, is the concentration of water quality parameters, are constants, express Any value in ; is the normalized reflectivity of the ith band at time t, and n is a constant; is the reflectivity of the lens corrected for the i-th band at the geographic coordinates (x, y) at time t, is the reflectance data of the all-black calibration image of the ith band at the geographic coordinates (x, y) at time t. It is the whiteboard reflectance data of the i-th band at the geographic coordinates (x, y) at time t.
[0058] In one embodiment of the present disclosure, the status update module 22 is specifically used to: The evaluation function corresponding to each water quality parameter is calculated according to the second formula; The second formula is:
[0059] Among them, F is the evaluation function, The concentration of water quality parameters obtained by analyzing water quality samples within the target range. is the water quality parameter concentration inverted by the initial hyperspectral data-water quality parameter relationship function model, is a constant, n is a constant, is the normalized reflectivity of the ith band at the coordinate (x, y) at time t, The inversion results are fitted by establishing a regression relationship between normalized reflectance and water quality parameter data. For the remainder.
[0060] In one embodiment of the present disclosure, the status update module 22 is specifically used to: Determine the individual optimal position of each particle based on the position of each particle and the evaluation function; Determine the global optimal position of multiple particles corresponding to each water quality parameter based on the individual optimal position of each particle; The state of each particle is updated based on the individual optimal position and the global optimal position.
[0061] In one embodiment of the present disclosure, the status update module 22 is specifically used to: Update the state of each particle according to the third and fourth formulas. The third formula is:
[0062] The fourth formula is:
[0063] Where w is the inertia weight, and For different accelerated particles, and is a random number uniformly distributed in the interval [0, 1]. represents the position of the nth particle at the Tth iteration, represents the velocity of the nth particle at the Tth iteration, represents the historical individual optimal position recorded by the nth particle at the Tth iteration, Represents the global optimal position recorded by n+1 particles.
[0064] In one embodiment of the present disclosure, the status update module 22 is further configured to: The individual optimal position of each particle and the global optimal position of multiple particles corresponding to each water quality parameter are updated according to the state of each particle.
[0065] See also Figure 4 , Figure 4 A schematic block diagram of an electronic device provided by an embodiment of the present disclosure. Figure 4 The electronic device 300 in the embodiment shown may include: one or more processors 301, one or more input devices 302, one or more output devices 303 and one or more memories 304. The processors 301, input devices 302, output devices 303 and memories 304 communicate with each other via a communication bus 305. The memory 304 is used to store computer programs, which include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to call the program instructions to execute the functions of each module / unit in the above-mentioned device embodiments, such as Figure 3 The functions of modules 21 to 24 are shown.
[0066] It should be understood that in the embodiment of the present disclosure, the processor 301 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0067] The input device 302 may include a touch panel, a fingerprint collection sensor (for collecting the user's fingerprint information and fingerprint direction information), a microphone, etc., and the output device 303 may include a display (LCD, etc.), a speaker, etc.
[0068] The memory 304 may include a read-only memory and a random access memory, and provide instructions and data to the processor 301. A portion of the memory 304 may also include a non-volatile random access memory. For example, the memory 304 may also store information about the device type.
[0069] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiments of the present disclosure can execute the implementation methods described in the first and second embodiments of the hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data provided in the embodiments of the present disclosure, and can also execute the implementation methods of the electronic device described in the embodiments of the present disclosure, which will not be repeated here.
[0070] In another embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by the processor, all or part of the processes in the above-mentioned embodiment method are implemented, and the computer program can also be completed by instructing the relevant hardware through the computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.
[0071] The computer-readable storage medium may be an internal storage unit of the electronic device of any of the aforementioned embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium may also be an external storage device of the electronic device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Furthermore, the computer-readable storage medium may also include both an internal storage unit of the electronic device and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.
[0072] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.
[0073] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the electronic devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0074] In the several embodiments provided in the present application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or it can be an electrical, mechanical or other form of connection.
[0075] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present disclosure.
[0076] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0077] The above are only specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present disclosure, and these modifications or replacements should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.
Claims
1. A hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data, characterized in that: include: Based on the normalized hyperspectral data and multi-temporal and spatial water quality data within the target range, multiple initial hyperspectral data-water quality parameter relationship function models are established; The multi-temporal and spatial water quality data includes a plurality of water quality parameters; updating the state of each particle based on the position of each particle and an evaluation function, wherein the evaluation function corresponds to an evaluation function for each water quality parameter, and each water quality parameter corresponds to a plurality of particles; Update the evaluation function corresponding to the water quality parameter based on the updated state of each particle; According to the comparison result between the output value of the updated evaluation function and the preset evaluation threshold, multiple initial hyperspectral data-water quality parameter relationship function models are iteratively trained to obtain multiple target hyperspectral data-water quality parameter relationship function models.
2. The hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data according to claim 1, characterized in that: Also includes: The hyperspectral data collected within the target range is normalized to obtain normalized hyperspectral data.
3. The hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data according to claim 1, characterized in that: The method of establishing multiple initial hyperspectral data-water quality parameter relationship function models based on the normalized hyperspectral data and multi-temporal and spatial water quality data within the target range includes: The initial hyperspectral data-water quality parameter relationship function model is established according to the first formula, and the first formula is: in, is the concentration of water quality parameters, are constants, express Any value in ; is the normalized reflectivity of the ith band at time t, and n is a constant; is the reflectivity of the lens corrected for the i-th band at the geographic coordinates (x, y) at time t, is the reflectance data of the all-black calibration image of the ith band at the geographic coordinates (x, y) at time t. It is the whiteboard reflectance data of the i-th band at the geographic coordinates (x, y) at time t.
4. The hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data according to claim 1, characterized in that: The evaluation function corresponding to each water quality parameter is calculated according to the second formula; The second formula is: Among them, F is the evaluation function, The concentration of water quality parameters obtained by analyzing water quality samples within the target range. is the water quality parameter concentration inverted by the initial hyperspectral data-water quality parameter relationship function model, is a constant, n is a constant, is the normalized reflectivity of the ith band at the coordinate (x, y) at time t, The inversion results are fitted by establishing a regression relationship between normalized reflectance and water quality parameter data. For the remainder.
5. The hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data according to claim 1, characterized in that: Update the state of each particle based on its position and evaluation function, including: Determine the individual optimal position of each particle based on the position of each particle and the evaluation function; Determine the global optimal position of the particle corresponding to each water quality parameter based on the individual optimal position of each particle; The state of each particle is updated based on the individual optimal position and the global optimal position.
6. The hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data according to claim 5 is characterized in that: The updating of the state of each particle based on the individual optimal position and the global optimal position comprises: The state of each particle is updated according to the third formula and the fourth formula, wherein the third formula is: The fourth formula is: Where w is the inertia weight, and For different accelerated particles, and is a random number uniformly distributed in the interval [0, 1]. represents the position of the nth particle at the Tth iteration, represents the velocity of the nth particle at the Tth iteration, represents the historical individual optimal position recorded by the nth particle at the Tth iteration, Represents the global optimal position recorded by n+1 particles.
7. The hyperspectral intelligent inversion method based on multi-temporal and spatial water quality data according to claim 1, characterized in that: After the state of each particle is updated, the method further includes: The individual optimal position of each particle and the global optimal position of the particle corresponding to each water quality parameter are updated according to the state of each particle.
8. A hyperspectral intelligent inversion device based on multi-temporal and spatial water quality data, characterized in that: include: A data acquisition module is used to establish multiple initial hyperspectral data-water quality parameter relationship function models based on normalized hyperspectral data and multi-temporal and spatial water quality data within the target range; The multi-temporal and spatial water quality data includes a plurality of water quality parameters; A state updating module, used for updating the state of each particle based on the position of each particle and an evaluation function, wherein the evaluation function is an evaluation function corresponding to each water quality parameter, and each water quality parameter corresponds to a plurality of particles; A data updating module, used for updating the evaluation function corresponding to the water quality parameter based on the updated state of each particle; The module training module is used to iteratively train multiple initial hyperspectral data-water quality parameter relationship function models according to the comparison result between the output value of the updated evaluation function and the preset evaluation threshold to obtain multiple target hyperspectral data-water quality parameter relationship function models.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.