Meteorological spectrometer-based multi-band visual meteorological dynamic monitoring method and device

Abstract

The invention relates to the technical field of data processing, in particular to a multi-band visual meteorological dynamic monitoring method and device based on meteorological spectrometers, and the method comprises the steps: synchronously obtaining the multi-band spectral data of a target region through a plurality of meteorological spectrometers; fusing the multi-band spectrum data of the plurality of meteorological spectrometers to generate multi-band spectrum data of the target area; and inverting meteorological data of the target area according to the multi-band spectral data to generate a corresponding visual meteorological dynamic monitoring result. Therefore, the problems of inaccurate meteorological data prediction, poor applicability and the like due to the fact that meteorological data is generated mostly depending on a single data source in the prior art are solved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a multi-band visual meteorological dynamic monitoring method and device based on a meteorological spectrometer. Background Art

[0002] Meteorological monitoring is an important means to understand the earth's climate system, predict weather changes, and respond to natural disasters. With the development of remote sensing technology and spectral analysis technology, the use of meteorological spectrometers for multi-band spectral data collection has become one of the effective ways to obtain high-precision meteorological parameters. However, in practical applications, existing technologies still face some challenges and limitations.

[0003] In related technologies, traditional meteorological monitoring often relies on a single sensor or a combination of limited types of sensors, which limits the spatial coverage and temporal resolution of the data, resulting in inaccurate meteorological data forecasts and poor applicability. Summary of the invention

[0004] The present application provides a multi-band visual meteorological dynamic monitoring method, device, electronic device, storage medium and program based on a meteorological spectrometer to solve the problems that most related technologies rely on a single data source to generate meteorological data, resulting in inaccurate meteorological data prediction and poor applicability.

[0005] The first aspect of the present application provides a multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer, comprising the following steps: using multiple meteorological spectrometers to synchronously acquire multi-band spectral data of a target area; fusing the multi-band spectral data of the multiple meteorological spectrometers to generate multi-band spectral data of the target area; and inverting the meteorological data of the target area according to the multi-band spectral data to generate corresponding visual meteorological dynamic monitoring results.

[0006] Optionally, the fusing of the multi-band spectral data of the multiple meteorological spectrometers to generate the multi-band spectral data of the target area includes: inputting the multi-band spectral data of the multiple meteorological spectrometers into a pre-trained target neural network model, the target neural network model outputs the fused multi-band spectral data of the target area, wherein the target neural network model includes: a convolutional neural network layer, a recurrent neural network layer, a graph neural network layer, an attention mechanism layer, a data fusion layer and an output layer; wherein the convolutional neural network layer is used to process multi-spectral data in space, the recurrent neural network layer is used to process spectral data in time series, the graph neural network layer is used to process spectral data in irregular space, the attention mechanism layer is used to extract key spectral data of multiple neural network layers, and adjust the weights of spectral data corresponding to different bands according to the key spectral data, the data fusion layer is used to fuse the multi-band spectral data of the target area according to the weights of spectral data corresponding to different bands, and the output layer outputs the fused multi-band spectral data of the target area.

[0007] Optionally, adjusting the weights of spectral data corresponding to different bands according to the key spectral data includes: using equal probability sampling to randomly select samples and bands in historical data to construct a PLS model; updating the selection probability of different bands according to the regression coefficient of the PLS model; introducing an exponential decay function to dynamically reduce the selection probability and remove redundant bands; optimizing the PLS model through multiple rounds of sampling iterations, and inputting the spectral data corresponding to the different bands into the PLS model to determine the weight corresponding to each band.

[0008] Optionally, after synchronously acquiring multi-band spectral data of the target area by using a plurality of meteorological spectrometers, the method further comprises: preprocessing the multi-band spectral data, wherein the preprocessing comprises radiation calibration, geometric correction and atmospheric correction.

[0009] Optionally, the generating of visualized dynamic meteorological monitoring results by inverting meteorological data of the target area according to the multi-band spectral data includes: extracting target data related to meteorology in the multi-band spectral data, wherein the target data includes reflectivity, transmittance and absorptivity of different bands; inputting a target inversion model according to the target data, and the target inversion model outputs meteorological data corresponding to the target area; and generating visualized dynamic meteorological monitoring results according to the meteorological data corresponding to the target area.

[0010] Optionally, the calculation formula of the target inversion model is: ;in, Expressed as the attenuation coefficient corresponding to the i-th band, Represented as the i-th band.

[0011] The second aspect of the present application provides a multi-band visual meteorological dynamic monitoring device based on a meteorological spectrometer, characterized in that it includes: an acquisition module for synchronously acquiring multi-band spectral data of a target area using multiple meteorological spectrometers; a fusion module for fusing the multi-band spectral data of the multiple meteorological spectrometers to generate multi-band spectral data of the target area; and an inversion module for inverting the meteorological data of the target area according to the multi-band spectral data to generate corresponding visual meteorological dynamic monitoring results.

[0012] The third aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to perform the multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer as described in the above embodiment.

[0013] The fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to perform the multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer as described in the above embodiment.

[0014] The fifth aspect of the present application provides a computer program product, including a computer program or instructions, characterized in that when the computer program or instructions are executed, a multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer as described in the above embodiment is implemented.

[0015] Therefore, this application has at least the following beneficial effects: The embodiments of the present application can utilize multiple meteorological spectrometers to synchronously acquire multi-band spectral data of a target area; fuse the multi-band spectral data of multiple meteorological spectrometers to generate multi-band spectral data of a target area; and through the synergistic complementarity of multi-source data, break through the technical limitations of a single sensor or a single frequency band, significantly improve the accuracy, timeliness and breadth of application of meteorological monitoring; generate corresponding visualized meteorological dynamic monitoring results based on the inversion of meteorological data of the target area based on the multi-band spectral data, thereby improving the accuracy of meteorological detection results.

[0016] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which: Figure 1 A flowchart of a multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer provided according to an embodiment of the present application; Figure 2This is an example diagram of a multi-band visual meteorological dynamic monitoring device based on a meteorological spectrometer provided according to an embodiment of the present application; Figure 3 It is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0018] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0019] The following describes the multi-band visual meteorological dynamic monitoring method, device, electronic device, storage medium and program based on a meteorological spectrometer in an embodiment of the present application with reference to the accompanying drawings.

[0020] Specifically, Figure 1 A schematic flow chart of a multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer provided in an embodiment of the present application.

[0021] like Figure 1 As shown, the multi-band visual meteorological dynamic monitoring method based on meteorological spectrometer includes the following steps: In step S101, multiple meteorological spectrometers are used to synchronously acquire multi-band spectral data of a target area.

[0022] It is understandable that the embodiment of the present application can use multiple meteorological spectrometers to synchronously acquire multi-band spectral data of the target area to facilitate the subsequent generation of corresponding meteorological data.

[0023] It should be noted that meteorological spectrometers include visible light and infrared imagers, hyperspectral imagers, Fourier transform infrared spectrometers, ultraviolet spectrometers and millimeter wave humidity detectors. Meteorological spectrometers are not limited to space applications. Ground stations also use similar equipment to monitor local or regional meteorological conditions.

[0024] In an embodiment of the present application, after using multiple meteorological spectrometers to synchronously acquire multi-band spectral data of a target area, the method includes: preprocessing the multi-band spectral data, wherein the preprocessing includes radiation calibration, geometric correction and atmospheric correction.

[0025] It should be noted that since the original spectral data is usually the digitized values ​​recorded by the sensor, these values ​​are affected by the sensor characteristics (such as gain, bias, noise) and cannot be directly used for quantitative analysis. In order to eliminate the influence of the sensor, this application needs to pre-process the multi-band spectral data. Specifically, through radiation calibration, the digitized values ​​recorded by the sensor are converted into radiation brightness, reflecting the radiation energy of the target object in a specific band, eliminating the influence of the sensor hardware characteristics on the data, and making the data obtained by different sensors or at different times comparable. The formula is as follows: , where L is the radiant brightness, DN is the original digital value, B is the bias value, and G is the gain coefficient.

[0026] When electromagnetic waves pass through the atmosphere, they are affected by atmospheric scattering, absorption and reflection, resulting in the inconsistency between the radiation brightness recorded by the sensor and the true radiation characteristics of the surface or atmosphere. Through atmospheric correction, the radiation brightness is converted into surface reflectivity (in percentage) to reflect the target object's ability to reflect solar radiation, so as to eliminate the influence of atmospheric scattering and absorption and obtain the true reflection characteristics of the surface, which is convenient for subsequent ground object classification and environmental monitoring. The formula is: , where ρ is the reflectivity; L is the radiance; d is the distance between the sun and the earth; E0 is the solar irradiance outside the atmosphere; and θ is the solar zenith angle.

[0027] In step S102 , multi-band spectral data of a plurality of meteorological spectrometers are fused to generate multi-band spectral data of a target area.

[0028] It can be understood that the embodiments of the present application can fuse the multi-band spectral data of multiple meteorological spectrometers to generate multi-band spectral data of the target area. Through the synergistic complementarity of multi-source data, it can break through the technical limitations of a single sensor or a single frequency band and significantly improve the accuracy, timeliness and application breadth of meteorological monitoring.

[0029] In an embodiment of the present application, multi-band spectral data of a plurality of meteorological spectrometers are fused to generate multi-band spectral data of a target area, including: inputting the multi-band spectral data of a plurality of meteorological spectrometers into a pre-trained target neural network model, the target neural network model outputting the fused multi-band spectral data of the target area, wherein the target neural network model includes: a convolutional neural network layer, a recurrent neural network layer, a graph neural network layer, an attention mechanism layer, a data fusion layer and an output layer; wherein the convolutional neural network layer is used to process multi-spectral data in space, the recurrent neural network layer is used to process spectral data in time series, the graph neural network layer is used to process spectral data in irregular space, the attention mechanism layer is used to extract key spectral data of a plurality of neural network layers, and adjust the weights of spectral data corresponding to different bands according to the key spectral data, the data fusion layer is used to fuse the multi-band spectral data of the target area according to the weights of spectral data corresponding to different bands, and the output layer outputs the fused multi-band spectral data of the target area.

[0030] It can be understood that the embodiment of the present application can input the multi-band spectral data of multiple meteorological spectrometers into a pre-trained target neural network model, and the target neural network model outputs the fused multi-band spectral data of the target area. The target neural network model integrates multiple architectures such as CNN, RNN, GNN, etc., and can process complex and changeable spectral data, enhance the generalization ability of the model, and enable it to perform well under different climate and terrain conditions. By fusing data from multiple meteorological spectrometers, the system can significantly improve the spatial resolution, temporal continuity and spatial coverage of spectral data, and provide more comprehensive and accurate data support for meteorological forecasting and environmental monitoring.

[0031] Specifically, the convolutional neural network layer is mainly responsible for processing multi-spectral data in space. Through convolution operations, it can extract local features in the image, such as edges, textures, etc. These features are crucial for understanding the spatial distribution of spectral data. Multi-layer convolution and pooling operations are used to gradually abstract higher-level spatial features; the recurrent neural network layer is used to process spectral data in time series, and can capture the changing trends and periodic characteristics of spectral data over time. Variants such as long short-term memory networks or gated recurrent units are used to alleviate the long-term dependency problem of traditional RNNs; the graph neural network layer models spectral data in irregular spaces, such as irregular distribution of objects in remote sensing images, and represents data through graph structures. GNN can capture the complex relationships between nodes, construct a graph representation of spectral data, and extract features using algorithms such as graph convolutional networks or graph attention networks. The attention mechanism layer is used to extract key spectral data from multiple neural network layers and dynamically adjust the weights of spectral data in different bands based on these key data. This helps the model focus on the data that has the greatest impact on the final result. It uses a multi-head attention mechanism to calculate the correlation between spectral data in different bands and assign weights accordingly. The data fusion layer uses weighted averaging or weighted summation based on the weights assigned by the attention mechanism layer, and combines the attention weights to fuse spectral data from different bands to generate multi-band spectral data of the target area. The output layer is responsible for outputting the fused multi-band spectral data of the target area, which can be used for subsequent meteorological analysis, environmental monitoring and other applications.

[0032] In an embodiment of the present application, the weights of spectral data corresponding to different bands are adjusted according to key spectral data, including: using equal probability sampling to randomly select samples and bands in historical data to construct a PLS model; updating the selection probability of different bands according to the regression coefficient of the PLS model; introducing an exponential decay function to dynamically reduce the selection probability and remove redundant bands; multiple rounds of sampling iteratively optimize the PLS model, and inputting the spectral data corresponding to different bands into the PLS model to determine the weight corresponding to each band.

[0033] It should be noted that a certain number of samples and bands are randomly selected from the historical data set. The selection is based on equal probability, that is, the probability of each sample and band being selected is the same. These selected samples and bands are used to build a partial least squares model, namely, a PLS model.

[0034] In the PLS model, the regression coefficient reflects the contribution of each band to the model prediction results. According to these regression coefficients, the selection probability of different bands can be updated. Specifically, the band with a larger absolute value of the regression coefficient has a higher selection probability.

[0035] In the process of multiple rounds of sampling, in order to prevent the model from being too dependent on certain specific bands, an exponential decay function is introduced to dynamically reduce the probability of being selected. The exponential decay function can gradually reduce the probability of being selected according to factors such as the number of times and time the band is selected. When the probability of a band being selected is reduced to a certain level, it can be regarded as a redundant band and eliminated.

[0036] In the process of multiple rounds of sampling, the PLS model is continuously iterated and optimized. Each round of sampling will update the selection probability of the band according to the results of the previous round, input the spectral data corresponding to different bands into the optimized PLS model, and determine the weight corresponding to each band by calculating the regression coefficient or related indicators. Through multiple rounds of sampling iterative optimization, the global optimal solution can be gradually approached, and the prediction accuracy of the model can be improved.

[0037] In step S103, the meteorological data of the target area is inverted according to the multi-band spectral data to generate corresponding visualized meteorological dynamic monitoring results.

[0038] It can be understood that the embodiment of the present application generates corresponding visualized meteorological dynamic monitoring results based on the meteorological data of the target area inverted by multi-band spectral data, thereby improving the accuracy of the meteorological data monitoring results.

[0039] In an embodiment of the present application, a visualized dynamic meteorological monitoring result is generated by inverting the meteorological data of a target area based on multi-band spectral data, including: extracting target data related to meteorology from the multi-band spectral data, wherein the target data includes reflectivity, transmittance and absorptivity of different bands; inputting a target inversion model based on the target data, and the target inversion model outputs meteorological data corresponding to the target area; and generating a visualized dynamic meteorological monitoring result based on the meteorological data corresponding to the target area.

[0040] Among them, the calculation formula of the target inversion model is: in, Expressed as the attenuation coefficient corresponding to the i-th band, Represented as the i-th band.

[0041] Specifically, based on the correlation between meteorological elements and spectral data, specific bands are selected for extraction. For example, some bands may be more sensitive to meteorological elements such as cloud cover and precipitation. A specific calculation formula is used to take the reflectivity, transmittance and absorptivity of different bands as input to output the meteorological data corresponding to the target area.

[0042] It should be noted that in the model, Represents the attenuation coefficient corresponding to the i-th band. These coefficients need to be set and adjusted according to the relationship between actual meteorological elements and spectral data. The model is trained and verified using known meteorological data and corresponding spectral data to ensure the accuracy and reliability of the model.

[0043] According to the multi-band visualized meteorological dynamic monitoring method based on meteorological spectrometer proposed in the embodiment of the present application, multiple meteorological spectrometers are used to synchronously acquire multi-band spectral data of the target area; the multi-band spectral data of the target area are fused by integrating the multi-band spectral data of the multiple meteorological spectrometers, and the technical limitations of a single sensor or a single frequency band are broken through through the synergistic complementarity of multi-source data, thereby significantly improving the accuracy, timeliness and application breadth of meteorological monitoring, and generating corresponding visualized meteorological dynamic monitoring results based on the inversion of the meteorological data of the target area according to the multi-band spectral data, thereby improving the accuracy of meteorological detection results.

[0044] Next, a multi-band visual meteorological dynamic monitoring device based on a meteorological spectrometer proposed according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0045] Figure 2 It is a block diagram of a multi-band visual meteorological dynamic monitoring device based on a meteorological spectrometer according to an embodiment of the present application.

[0046] like Figure 2 As shown, the multi-band visualized meteorological dynamic monitoring device 10 based on a meteorological spectrometer includes: an acquisition module 100 , a fusion module 200 and an inversion module 300 .

[0047] Among them, the acquisition module 100 is used to use multiple meteorological spectrometers to synchronously acquire multi-band spectral data of the target area; the fusion module 200 is used to fuse the multi-band spectral data of multiple meteorological spectrometers to generate multi-band spectral data of the target area; the inversion module 300 is used to invert the meteorological data of the target area according to the multi-band spectral data to generate corresponding visualized meteorological dynamic monitoring results.

[0048] It should be noted that the aforementioned explanation of the embodiment of the multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer is also applicable to the multi-band visual meteorological dynamic monitoring device based on a meteorological spectrometer in this embodiment, and will not be repeated here.

[0049] According to the multi-band visual meteorological dynamic monitoring device based on meteorological spectrometer proposed in the embodiment of the present application, multiple meteorological spectrometers are used to synchronously acquire multi-band spectral data of a target area; the multi-band spectral data of multiple meteorological spectrometers are fused to generate multi-band spectral data of the target area, and through the synergistic complementarity of multi-source data, the technical limitations of a single sensor or a single frequency band are broken through, and the accuracy, timeliness and application breadth of meteorological monitoring are significantly improved. The meteorological data of the target area is inverted according to the multi-band spectral data to generate corresponding visual meteorological dynamic monitoring results, thereby improving the accuracy of meteorological detection results.

[0050] FIG3 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include: A memory 301 , a processor 302 , and a computer program stored in the memory 301 and executable on the processor 302 .

[0051] When the processor 302 executes the program, the multi-band visual meteorological dynamic monitoring method based on the meteorological spectrometer provided in the above embodiment is implemented.

[0052] Furthermore, the electronic device further comprises: The communication interface 303 is used for communication between the memory 301 and the processor 302 .

[0053] The memory 301 is used to store computer programs that can be run on the processor 302 .

[0054] The memory 301 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0055] If the memory 301, the processor 302 and the communication interface 303 are implemented independently, the communication interface 303, the memory 301 and the processor 302 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0056] Optionally, in a specific implementation, if the memory 301, the processor 302 and the communication interface 303 are integrated on a chip, the memory 301, the processor 302 and the communication interface 303 can communicate with each other through an internal interface.

[0057] The processor 302 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0058] An embodiment of the present application also provides a computer-readable storage medium having a computer program or instruction stored thereon. When the computer program or instruction is executed by a processor, the multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer is implemented as described above.

[0059] An embodiment of the present application also provides a computer program product, including a computer program or instructions, characterized in that when the computer program or instructions are executed, the multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer is implemented as described above.

[0060] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0061] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0062] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0063] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiment, N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of multiple of the following technologies known in the art: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0064] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

Claims

1. A multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer, characterized in that: The following steps are involved: Use multiple meteorological spectrometers to synchronously acquire multi-band spectral data of the target area; The multi-band spectral data of the plurality of meteorological spectrometers are fused to generate the multi-band spectral data of the target area, wherein the multi-band spectral data of the plurality of meteorological spectrometers are input into a pre-trained target neural network model, and the target neural network model outputs the fused multi-band spectral data of the target area, wherein the target neural network model comprises: a convolutional neural network layer, a recurrent neural network layer, a graph neural network layer, an attention mechanism layer, a data fusion layer and an output layer; wherein the convolutional neural network layer is used to process the multi-spectral data in space, the recurrent neural network layer is used to process the spectral data of the time series, the graph neural network layer is used to process the spectral data of the irregular space, the attention mechanism layer is used to extract the key spectral data of the plurality of neural network layers, and adjust the weights of the spectral data corresponding to different bands according to the key spectral data, the data fusion layer is used to fuse the multi-band spectral data of the target area according to the weights of the spectral data corresponding to different bands, and the output layer outputs the fused multi-band spectral data of the target area; The meteorological data of the target area is inverted according to the multi-band spectral data to generate corresponding visualized meteorological dynamic monitoring results.

2. The multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer according to claim 1 is characterized in that: The step of adjusting the weights of the spectral data corresponding to different bands according to the key spectral data includes: Use equal probability sampling to randomly select samples and bands from historical data to build PLS models; Update the selection probability of different bands according to the regression coefficient of the PLS model; Introduce exponential decay function to dynamically reduce the probability of being selected and remove redundant bands; The PLS model is optimized through multiple rounds of sampling iterations, and the spectral data corresponding to the different bands are input into the PLS model to determine the weight corresponding to each band.

3. The multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer according to claim 1 is characterized in that: After using multiple meteorological spectrometers to synchronously acquire multi-band spectral data of the target area, including: The multi-band spectral data is preprocessed, wherein the preprocessing includes radiation calibration, geometric correction and atmospheric correction.

4. The multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer according to claim 1 is characterized in that: The generating of a visualized meteorological dynamic monitoring result by inverting the meteorological data of the target area according to the multi-band spectral data includes: Extracting target data related to meteorology from multi-band spectral data, where the target data includes reflectivity, transmittance and absorptivity in different bands; Inputting a target inversion model according to the target data, the target inversion model outputs meteorological data corresponding to the target area; Generate visual meteorological dynamic monitoring results based on the meteorological data corresponding to the target area.

5. The multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer according to claim 4 is characterized in that: The calculation formula of the target inversion model is: ;in, Expressed as the attenuation coefficient corresponding to the i-th band, Represented as the i-th band.

6. A multi-band visual meteorological dynamic monitoring device based on a meteorological spectrometer, used to implement the multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer according to any one of claims 1 to 5, characterized in that: include: An acquisition module, used to synchronously acquire multi-band spectral data of a target area using multiple meteorological spectrometers; A fusion module, used for fusing the multi-band spectral data of the multiple meteorological spectrometers to generate multi-band spectral data of a target area, wherein the multi-band spectral data of the multiple meteorological spectrometers are input into a pre-trained target neural network model, and the target neural network model outputs the fused multi-band spectral data of the target area, wherein the target neural network model includes: a convolutional neural network layer, a recurrent neural network layer, a graph neural network layer, an attention mechanism layer, a data fusion layer and an output layer; wherein the convolutional neural network layer is used to process multi-spectral data in space, the recurrent neural network layer is used to process spectral data in time series, the graph neural network layer is used to process spectral data in irregular space, the attention mechanism layer is used to extract key spectral data of multiple neural network layers, and adjust the weights of spectral data corresponding to different bands according to the key spectral data, the data fusion layer is used to fuse the multi-band spectral data of the target area according to the weights of spectral data corresponding to different bands, and the output layer outputs the fused multi-band spectral data of the target area; The inversion module is used to invert the meteorological data of the target area according to the multi-band spectral data to generate corresponding visualized meteorological dynamic monitoring results.

7. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer as described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer as described in any one of claims 1 to 5.

9. A computer program product, characterized in that It includes a computer program, which, when executed by a processor, is used to implement the multi-band visual meteorological dynamic monitoring method based on a meteorological spectrometer as described in any one of claims 1 to 5.

Images