A multi-mode radar fusion-based precipitation observation method and system

Abstract

The application provides a multi-mode radar fusion precipitation observation method and system, and relates to the technical field of meteorological analysis. The method comprises the following steps: performing Fourier transform on two-band space-time data sets to the frequency domain to obtain two-band frequency domain space-time data sets; for any grid in the city, determining the grid type in the two bands based on the fluctuation intensity of the amplitude spectrum in the two-band frequency domain space-time data sets over time; determining the enhanced power spectrum of the X-band high frequency based on the amplitude spectrum under the X-band of the X-band high frequency grid; determining the enhanced power spectrum of the C-band low frequency based on the amplitude spectrum under the C-band of the C-band low frequency grid; determining the power spectrum of the intermediate frequency according to the intermediate frequency state of the grid in the two bands; for any grid, fusing the power spectrum and the phase spectrum to obtain the fusion spectrum of the city; and performing inverse Fourier transform on the fusion spectrum to the spatial domain, and observing the precipitation of the city based on the obtained fusion space-time data set. The application aims to improve the precipitation observation accuracy of radar data.

Description

Technical Field

[0001] This application relates to the field of meteorological analysis technology, specifically to a precipitation observation method and system based on multi-mode radar fusion. Background Technology

[0002] Urban disaster prevention and mitigation, as well as weather forecasting, are of paramount importance. Radar technology, as a crucial remote sensing tool, plays a vital role in weather monitoring and disaster early warning. Multi-mode radar systems, in particular (such as X-band and C-band radars), each possess distinct advantages: X-band radar offers high spatial resolution, suitable for capturing subtle local changes, while C-band radar boasts a wider coverage area, ideal for monitoring broad-area precipitation or weather systems. However, due to the differences in radar bands, the data from these two types of radar exhibit significant variations in spatial resolution and signal strength. Effectively fusing this data to improve overall accuracy and usability remains a major challenge in radar data processing.

[0003] Current mainstream radar data fusion methods typically focus on the extraction and fusion of spatial domain features, such as aligning and fusing images from different radar bands through techniques like image registration and spatial filtering. These methods attempt to fuse information from different bands through spatial domain similarity processing. However, these methods have the following shortcomings: Firstly, existing radar data fusion methods often emphasize spatial domain feature extraction and fail to fully utilize frequency domain features, resulting in insufficient capture of high-frequency information and consequently neglecting the information contained in radar frequency domain features. Secondly, these methods usually rely on global image processing, limiting their ability to capture local details, especially when processing high-frequency signals in radar echoes. Traditional spatial domain methods tend to ignore or lose subtle local features, leading to insufficient local feature capture capabilities. This makes existing fusion methods lacking in accuracy and resolution for frequently changing precipitation systems in complex urban environments. Summary of the Invention

[0004] In view of this, this application provides a multi-mode radar fusion precipitation observation method and system, aiming to solve or partially solve the problems existing in the background art.

[0005] The first aspect of this application provides a multi-mode radar fusion method for precipitation observation, the method comprising:

[0006] Obtain the spatiotemporal datasets of the X-band radar and C-band radar currently aligned in spatial resolution for the target city;

[0007] The spatiotemporal datasets of the two bands are transformed from the spatial domain to the frequency domain by Fourier transform, resulting in the first frequency domain spatiotemporal dataset of the X-band and the second frequency domain spatiotemporal dataset of the C-band.

[0008] For any grid in the target city, based on the fluctuation intensity of the amplitude spectrum of the first frequency domain spatiotemporal data of the grid over time, the grid type of the grid in the current X-band is determined, and based on the fluctuation intensity of the amplitude spectrum of the second frequency domain spatiotemporal data of the grid over time, the grid type of the grid in the current C-band is determined, wherein the grid type includes high-frequency grid, mid-frequency grid and low-frequency grid;

[0009] Based on the amplitude spectrum of the first frequency domain spatiotemporal data of the X-band high-frequency grid, the enhanced power spectrum of the X-band high-frequency grid in the X-band high frequency is determined.

[0010] Based on the amplitude spectrum of the second frequency domain spatiotemporal data of the C-band low-frequency grid, the enhanced power spectrum of the C-band low-frequency grid at the C-band low frequency is determined.

[0011] Based on the intermediate frequency (IF) state of the grid in the two bands, the amplitude spectrum of the corresponding frequency domain spatiotemporal data is selected to determine the power spectrum of the grid at the IF. Specifically, when both bands are IF, the amplitude spectrum of the first frequency domain spatiotemporal data and the amplitude spectrum of the second frequency domain spatiotemporal data of the grid are selected; when only the X band is IF, the amplitude spectrum of the first frequency domain spatiotemporal data of the grid is selected; when only the C band is IF, the amplitude spectrum of the second frequency domain spatiotemporal data of the grid is selected.

[0012] For any grid in the target city, the power spectra of each grid are fused, and the phase spectra of the first frequency domain spatiotemporal data and the second frequency domain spatiotemporal data of the grid are fused to obtain the fused spectrum of the target city in units of grid.

[0013] The fused spectrum of the target city is transformed from the frequency domain to the spatial domain by using a two-dimensional inverse discrete Fourier transform to obtain a fused spatiotemporal dataset.

[0014] Based on the fused spatiotemporal dataset, the precipitation situation of the target city was observed.

[0015] A second aspect of this application provides a multi-mode radar fusion precipitation observation system, the system comprising:

[0016] The data acquisition module is used to acquire the spatiotemporal datasets of the X-band radar and the C-band radar of the target city, which are currently aligned in spatial resolution.

[0017] The Fourier transform module is used to transform the spatiotemporal datasets of two bands from the spatial domain to the frequency domain through Fourier transform, so as to obtain the first frequency domain spatiotemporal dataset of the X-band and the second frequency domain spatiotemporal dataset of the C-band.

[0018] The grid type determination module is used to determine the current grid type of any grid in the target city in the X-band based on the fluctuation intensity of the amplitude spectrum of the grid's past first frequency domain spatiotemporal data over time, and to determine the current grid type of the grid in the C-band based on the fluctuation intensity of the amplitude spectrum of the grid's past second frequency domain spatiotemporal data over time. The grid type includes high-frequency grid, mid-frequency grid, and low-frequency grid.

[0019] The first power spectrum determination module is used to determine the enhanced power spectrum of the X-band high-frequency grid in the X-band high frequency based on the amplitude spectrum of the first frequency domain spatiotemporal data of the X-band high-frequency grid.

[0020] The second power spectrum determination module is used to determine the enhanced power spectrum of the C-band low-frequency grid at the C-band low frequency based on the amplitude spectrum of the second frequency domain spatiotemporal data of the C-band low-frequency grid.

[0021] The third power spectrum determination module is used to select the amplitude spectrum of the corresponding frequency domain spatiotemporal data based on the intermediate frequency state of the grid in two bands, and determine the power spectrum of the grid at the intermediate frequency. Specifically, when both bands are intermediate frequencies, the amplitude spectrum of the first frequency domain spatiotemporal data and the amplitude spectrum of the second frequency domain spatiotemporal data of the grid are selected; when only the X band is an intermediate frequency, the amplitude spectrum of the first frequency domain spatiotemporal data of the grid is selected; when only the C band is an intermediate frequency, the amplitude spectrum of the second frequency domain spatiotemporal data of the grid is selected.

[0022] The power spectrum fusion module is used to fuse the power spectra of each grid in any grid in the target city, and to fuse the phase spectrum of the first frequency domain spatiotemporal data and the phase spectrum of the second frequency domain spatiotemporal data of the grid to obtain the fused spectrum of the target city on a grid-by-grid basis.

[0023] The inverse Fourier transform module is used to transform the fused spectrum of the target city from the frequency domain to the spatial domain through a two-dimensional inverse discrete Fourier transform to obtain a fused spatiotemporal dataset.

[0024] The precipitation observation module is used to observe the precipitation situation of the target city based on the fused spatiotemporal dataset.

[0025] The precipitation observation method based on multi-mode radar fusion provided in this application has the following advantages:

[0026] The precipitation observation method provided in this application is a radar band data fusion method based on Fourier transform, which combines radar frequency domain transformation and multi-scale weighted fusion. It converts X-band and C-band radar data into the frequency domain and then performs weighted fusion considering multi-scale characteristics. This method not only fully preserves high-frequency information (such as details of local precipitation units) but also smoothly fuses low-frequency information (such as large-scale precipitation background). Through this method, this application can effectively improve the application performance of radar data in complex urban environments, providing more accurate technical support for meteorological monitoring, disaster early warning, and other fields. Attached Figure Description

[0027] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 The flowchart illustrates a multi-mode radar fusion precipitation observation method according to one embodiment of this application;

[0029] Figure 2 This is a schematic diagram illustrating a multi-mode radar fusion precipitation observation method according to one embodiment of this application;

[0030] Figure 3 This is a schematic diagram of a multi-mode radar fusion precipitation observation system according to one embodiment of this application. Detailed Implementation

[0031] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0032] refer to Figure 1 , Figure 1 This is a flowchart illustrating a multi-mode radar fusion precipitation observation method according to one embodiment of this application. Figure 1 As shown, the method includes:

[0033] Step S01: Obtain the spatiotemporal datasets of the X-band radar and the C-band radar of the target city, which are currently aligned in spatial resolution.

[0034] In this embodiment, the radar data used for multi-mode radar fusion includes radar data from two bands: spatiotemporal datasets using X-band radar and C-band radar. The X-band radar spatiotemporal dataset provides high spatial resolution data, suitable for capturing small-scale precipitation systems (such as localized showers or precipitation units), while the C-band radar spatiotemporal dataset provides a larger coverage area, suitable for monitoring large-scale precipitation systems (such as wide-area precipitation or weather systems). The spatiotemporal data in the spatiotemporal dataset is stored in raster data format. Each cell in the raster spatiotemporal data is a grid, and the spatiotemporal data of each grid represents the radar echo intensity of the corresponding urban area at the corresponding time. This includes a large amount of reflectivity data; that is, the spatiotemporal data recorded in this spatiotemporal dataset corresponding to the band is the radar data of the target city in the corresponding spatiotemporal time, representing the current radar data of each grid in the target city in the corresponding band. The target city can be any city.

[0035] In this embodiment, the initial spatiotemporal datasets of X-band radar and C-band radar typically have different spatial resolutions and coordinate systems. For example, the radar data resolution of X-band radar is higher, while that of C-band radar is lower. Therefore, before using the initially acquired initial spatiotemporal datasets of X-band radar and C-band radar for subsequent fusion, it is necessary to preprocess these two initial spatiotemporal datasets to ensure their uniformity and consistency. This preprocessing includes at least aligning the initial spatiotemporal datasets of X-band radar and C-band radar to a unified coordinate system and spatial resolution. By preprocessing the initial spatiotemporal datasets of the target city in both bands, a spatiotemporal dataset of X-band radar and C-band radar that is spatially aligned and has a unified coordinate system is obtained for subsequent fusion.

[0036] Step S02: Transform the spatiotemporal datasets of the two bands from the spatial domain to the frequency domain using Fourier transform to obtain the first frequency domain spatiotemporal dataset of the X-band and the second frequency domain spatiotemporal dataset of the C-band.

[0037] In this embodiment, after obtaining the spatiotemporal datasets of the X-band radar and the C-band radar, which are spatially aligned and have the same coordinate system for the target city, through step S01, the spatiotemporal datasets of the two bands are transformed from the spatial domain to the frequency domain using two-dimensional discrete Fourier transform, resulting in a first frequency domain spatiotemporal dataset for the X-band and a second frequency domain spatiotemporal dataset for the C-band. The transformation expressions are as follows:

[0038]

[0039]

[0040] Where (m,n,t) are the spatiotemporal coordinates of the grid (m,n) in the spatial domain, t represents time, (u,v,t) are the spatiotemporal coordinates of the grid (u,v) in the frequency domain, M and N are the number of raster rows and columns, respectively, and j is the imaginary unit. This represents the amplitude spectrum of the grid (u,v) in the first frequency domain spatiotemporal data of the X-band. This represents the amplitude spectrum of the grid (u,v) in the second frequency domain spatiotemporal data of the C-band. This refers to the spatiotemporal data of a grid (m,n) in the X-band within the spatiotemporal domain. This represents the spatiotemporal data of a grid (m,n) in the C-band under the spatiotemporal domain.

[0041] In this embodiment, the result obtained after performing a two-dimensional discrete Fourier transform also includes the X-band phase spectrum corresponding to the spatiotemporal data of the grid in the X-band. And, the C-band phase spectrum corresponding to the spatiotemporal data of the grid in the C-band. The first frequency domain spatiotemporal data in the first frequency domain spatiotemporal dataset, corresponding to the grid, includes two parts: the amplitude spectrum corresponding to the grid. and the phase spectrum corresponding to the grid That is, the first frequency domain spatiotemporal data is represented as Similarly, the second frequency domain spatiotemporal data corresponding to the grid in the second frequency domain spatiotemporal dataset includes two parts: the amplitude spectrum corresponding to the grid. and the phase spectrum corresponding to the grid That is, the second frequency domain spatiotemporal data is represented as .

[0042] Step S03: For any grid in the target city, based on the fluctuation intensity of the amplitude spectrum of the first frequency domain spatiotemporal data of the grid over time, determine the grid type of the grid in the X-band; and based on the fluctuation intensity of the amplitude spectrum of the second frequency domain spatiotemporal data of the grid over time, determine the grid type of the grid in the C-band. The grid type includes high-frequency grid, mid-frequency grid and low-frequency grid.

[0043] In this embodiment, for any grid cell in the target city, based on the specific values ​​of the amplitude spectrum of the grid cell's most recent consecutive first frequency domain spatiotemporal data, the fluctuation intensity of the first frequency domain spatiotemporal data amplitude spectrum over time in the most recent past period is determined. If the fluctuation intensity of the grid cell's first frequency domain spatiotemporal data amplitude spectrum over time is large, the grid cell is determined to currently belong to the X-band high-frequency grid; if the fluctuation intensity of the grid cell's first frequency domain spatiotemporal data amplitude spectrum over time is small, the grid cell is determined to currently belong to the X-band low-frequency grid; if the fluctuation intensity of the grid cell's first frequency domain spatiotemporal data amplitude spectrum over time is moderate, the grid cell is determined to currently belong to the X-band mid-frequency grid. Through the same implementation method, each grid cell in the target city can calculate whether it currently belongs to a high-frequency grid, a mid-frequency grid, or a low-frequency grid in the X-band.

[0044] In this embodiment, for any grid cell in the target city, based on the specific values ​​of the amplitude spectrum of the grid cell's most recent consecutive second-frequency-domain spatiotemporal data over the past, the fluctuation intensity of the grid cell's amplitude spectrum over the most recent past period is determined. If the fluctuation intensity of the grid cell's amplitude spectrum over the most recent past period is large, the grid cell is determined to belong to a C-band high-frequency grid; if the fluctuation intensity is small, the grid cell is determined to belong to a C-band low-frequency grid; if the fluctuation intensity is moderate, the grid cell is determined to belong to a C-band mid-frequency grid. Through the same implementation method, each grid cell in the target city can calculate whether it currently belongs to a high-frequency, mid-frequency, or low-frequency grid in the C-band.

[0045] One preferred implementation of the fluctuation intensity of the amplitude spectrum of the first frequency domain spatiotemporal data of the grid in the most recent past period is as follows: the overall standard deviation of the amplitude spectrum of multiple consecutive first frequency domain spatiotemporal data of the grid in the most recent past period is determined as the corresponding fluctuation intensity.

[0046] One possible implementation of the fluctuation intensity of the amplitude spectrum of the first frequency domain spatiotemporal data of the grid over the most recent past period is as follows: determine the average absolute deviation of the amplitude spectrum of multiple consecutive first frequency domain spatiotemporal data of the grid over the most recent past period as the corresponding fluctuation intensity. The average absolute deviation is calculated by calculating the mean of the amplitude spectra of multiple first frequency domain spatiotemporal data; calculating the absolute value of the difference between the amplitude spectrum of each first frequency domain spatiotemporal data and the mean; and calculating the arithmetic mean of the absolute values.

[0047] One preferred implementation of the fluctuation intensity of the amplitude spectrum of the second frequency domain spatiotemporal data of the grid in the most recent past period is as follows: the overall standard deviation of the amplitude spectrum of multiple consecutive second frequency domain spatiotemporal data of the grid in the most recent past period is determined as the corresponding fluctuation intensity.

[0048] One possible implementation of the fluctuation intensity of the amplitude spectrum of second-frequency domain spatiotemporal data of a grid over the most recent past period is as follows: The average absolute deviation of the amplitude spectra of multiple consecutive second-frequency domain spatiotemporal data of the grid over the most recent past period is determined as the corresponding fluctuation intensity. The average absolute deviation is calculated by: calculating the mean of the amplitude spectra of multiple second-frequency domain spatiotemporal data; calculating the absolute value of the difference between the amplitude spectrum of each second-frequency domain spatiotemporal data and the mean; and calculating the arithmetic mean of the absolute values.

[0049] In this embodiment, for any grid in the target city, there are two types: grid type under X-band and grid type under C-band.

[0050] Step S04: Based on the amplitude spectrum of the first frequency domain spatiotemporal data of the X-band high-frequency grid, determine the enhanced power spectrum of the X-band high-frequency grid in the X-band high frequency.

[0051] In this embodiment, for any grid cell in the target city, it is determined whether the grid cell belongs to an X-band high-frequency grid. If the grid cell belongs to an X-band high-frequency grid, its power spectrum in the X-band is enhanced. The enhanced amplitude spectrum of the grid cell is obtained by multiplying its first frequency domain spatiotemporal data amplitude spectrum by an enhancement coefficient. Then, based on the enhanced amplitude spectrum of the grid cell, its enhanced power spectrum in the X-band high frequency is calculated. Using the same implementation method, each grid cell in the target city that currently belongs to an X-band high-frequency grid can calculate its own current enhanced power spectrum in the X-band high frequency.

[0052] Step S05: Based on the amplitude spectrum of the second frequency domain spatiotemporal data of the C-band low-frequency grid, determine the enhanced power spectrum of the C-band low-frequency grid at the C-band low frequency.

[0053] In this embodiment, for any grid cell in the target city, it is determined whether the grid cell belongs to a C-band low-frequency grid. If the grid cell belongs to a C-band low-frequency grid, the power spectrum of the grid cell in the C-band is enhanced. Based on the amplitude spectrum of the second frequency domain spatiotemporal data of the grid cell, the power spectrum of the grid cell in the C-band low frequency is calculated. Then, the power spectrum is multiplied by an enhancement factor to obtain the enhanced power spectrum of the grid cell in the C-band low frequency. Through the same implementation method, each grid cell in the target city that currently belongs to a C-band low-frequency grid can calculate its own enhanced power spectrum in the C-band low frequency.

[0054] Step S06: Based on the intermediate frequency (IF) state of the grid in the two bands, select the amplitude spectrum of the corresponding frequency domain spatiotemporal data to determine the power spectrum of the grid at the IF. Specifically, when both bands are IF, select the amplitude spectrum of the first frequency domain spatiotemporal data and the amplitude spectrum of the second frequency domain spatiotemporal data of the grid. When only the X band is IF, select the amplitude spectrum of the first frequency domain spatiotemporal data of the grid. When only the C band is IF, select the amplitude spectrum of the second frequency domain spatiotemporal data of the grid.

[0055] In this embodiment, for any grid in the target city, if the grid belongs to the intermediate frequency (IF) grid in both frequency bands, the power spectrum of the grid at the IF is calculated based on the amplitude spectrum of the first frequency domain spatiotemporal data and the amplitude spectrum of the second frequency domain spatiotemporal data of the grid. If the grid belongs to the IF grid only in the X-band and not in the C-band, the power spectrum of the grid at the IF is calculated based on the amplitude spectrum of the first frequency domain spatiotemporal data of the grid. If the grid belongs to the IF grid only in the C-band and not in the X-band, the power spectrum of the grid at the IF is calculated based on the amplitude spectrum of the second frequency domain spatiotemporal data of the grid.

[0056] Step S07: For any grid in the target city, fuse the power spectra of each grid, and fuse the phase spectrum of the first frequency domain spatiotemporal data and the phase spectrum of the second frequency domain spatiotemporal data of the grid to obtain the fused spectrum of the target city on a grid-by-grid basis.

[0057] In this embodiment, through steps S01 to S06, each grid cell in the target city can calculate its own power spectrum. Since a grid cell in the target city will have corresponding grid types in two bands, and different bands may have corresponding power spectrum calculation results, a grid cell in the target city may have multiple power spectra. This application performs a fusion calculation on the power spectra of any grid cell in the target city to obtain the fused power spectrum of the grid cell, and simultaneously performs a fusion calculation on the phase spectra of the two bands belonging to the grid cell to obtain the fused phase spectrum of the grid cell. The fused power spectrum and fused phase spectrum of the grid cell constitute the fused spectrum of the grid cell. Through the same implementation method, each grid cell in the target city can calculate its own corresponding fused spectrum, and the fused spectra of all grid cells in the target city together constitute the fused spectrum of the target city on a grid cell basis. One optional implementation method for power spectrum fusion calculation is: setting corresponding weights for each power spectrum of the grid cell, and then performing a weighted summation of the power spectra of the grid cell to obtain the fused power spectrum of the grid cell. One possible implementation of phase spectrum fusion calculation is to add the phase spectra of the grid in the two bands and take the average value as the fused power spectrum of the grid.

[0058] Step S08: Transform the fused spectrum of the target city from the frequency domain to the spatial domain using a two-dimensional inverse discrete Fourier transform to obtain a fused spatiotemporal dataset.

[0059] In this embodiment, after obtaining the fused spectrum of the target city in grid units through steps S01 to S07, the fused spectrum of the target city in grid units is transformed back to the spatial domain through a two-dimensional inverse discrete Fourier transform. First, the complex spectrum is reconstructed, expressed as: Then, the inverse Fourier transform is used to transform the frequency domain back to the spatial domain, expressed as follows: ,in, For spatial domain spatiotemporal coordinates, For frequency domain spatiotemporal coordinates, For grid-based fused power spectrum The obtained fusion amplitude spectrum, The fused phase spectrum of the grid, For the fusion of spatiotemporal data from the grid, and These represent the number of raster rows and columns, respectively. The virtual unit is used. By transforming the fused spectrum of the target city from the frequency domain to the spatial domain through a two-dimensional inverse discrete Fourier transform, the fused spatiotemporal dataset of the target city can be obtained.

[0060] Step S09: Observe the precipitation situation of the target city based on the fused spatiotemporal dataset.

[0061] In this embodiment, after obtaining the fused spatiotemporal dataset of the target city in grid units through steps S01 to S07, since the original radar observation is the reflectivity of precipitation particles in the atmosphere, it is converted into ground precipitation through quantitative precipitation estimation based on the fused spatiotemporal dataset. Since converting radar data into precipitation data is a conventional calculation in this field, it will not be described in detail here. Thus, the precipitation situation of the target city can be obtained.

[0062] The precipitation observation method provided in this application is a radar band data fusion method based on Fourier transform, which combines radar frequency domain transformation and multi-scale weighted fusion. It converts X-band and C-band radar data into the frequency domain and then performs weighted fusion considering multi-scale characteristics. This method not only fully preserves high-frequency information (such as details of local precipitation units) but also smoothly fuses low-frequency information (such as large-scale precipitation background). Through this method, this application can effectively improve the application performance of radar data in complex urban environments, providing more accurate technical support for meteorological monitoring, disaster early warning, and other fields.

[0063] In conjunction with the above embodiments, in one implementation, this application also provides a multi-mode radar fusion precipitation observation method. In this multi-mode radar fusion precipitation observation method, step S03 may include:

[0064] Step S03_1a: For any grid in the target city, obtain the amplitude spectrum of the first frequency domain spatiotemporal data of the grid preceding the current time for a predetermined number of consecutive times.

[0065] In this embodiment, for any grid cell in the target city, the amplitude spectrum of the grid cell in the first frequency domain spatiotemporal data of the most recent past period is obtained.

[0066] Step S03_2a: Based on the amplitude spectrum of the acquired continuous preset number of first frequency domain spatiotemporal data, calculate the fluctuation intensity of the amplitude spectrum of the first frequency domain spatiotemporal data of the grid over time using the X-band fluctuation intensity algorithm; the X-band fluctuation intensity algorithm is as follows: Where T represents the number of amplitude spectra of the first frequency domain spatiotemporal data of the acquired grid prior to the current time. This represents the amplitude spectrum of the first frequency domain spatiotemporal data of the grid at time t. This represents the mean of the amplitude spectrum of the corresponding number of first-frequency-domain spatiotemporal data obtained. This represents the amplitude spectrum fluctuation intensity of the first frequency domain spatiotemporal data of the grid over time.

[0067] In this embodiment, the amplitude spectrum of the first frequency domain spatiotemporal data of the grid over the most recent past period is calculated using the X-band wave intensity algorithm, thus obtaining the value of the wave intensity of the amplitude spectrum of the first frequency domain spatiotemporal data of the grid over time. In the X-band wave intensity algorithm... This represents the intensity of the amplitude spectrum fluctuation over time in the first frequency domain spatiotemporal data of the grid in the most recent past period. The subscript for time t is omitted to avoid conflict with the time t used for summation in the X-band wave intensity algorithm. T represents the number of amplitude spectra of the first frequency domain spatiotemporal data obtained in a predetermined number of consecutive steps.

[0068] Step S03_3a: If the fluctuation intensity is greater than the high frequency threshold, determine that the grid is currently a high frequency grid in the X band.

[0069] In this embodiment, if the fluctuation intensity obtained through step S03_2a is greater than the high-frequency threshold, it is determined that the grid is currently a high-frequency grid in the X-band. That is, this application determines the grid type of the grid in the X-band based on the amplitude spectrum of the grid over a past period, such as whether it is a high-frequency grid, a mid-frequency grid, or a low-frequency grid in the X-band.

[0070] Step S03_4a: If the fluctuation intensity is less than or equal to the low frequency threshold, determine that the grid is currently a low frequency grid in the X band.

[0071] In this embodiment, if the fluctuation intensity obtained through step S03_2a is less than or equal to the low-frequency threshold, the grid is determined to be a low-frequency grid in the X-band.

[0072] Step S03_5a: If the fluctuation intensity is less than or equal to the high-frequency threshold and greater than the low-frequency threshold, determine that the grid is currently a mid-frequency grid in the X-band;

[0073] In this embodiment, if the fluctuation intensity obtained through step S03_2a is less than or equal to the high-frequency threshold and greater than the low-frequency threshold, the grid is determined to be a mid-frequency grid in the X-band. For example, if the high-frequency threshold is... The low-frequency threshold is The wave intensity obtained through step S03_2a satisfy At that time, it was determined that the current grid is a mid-frequency grid in the X-band.

[0074] In conjunction with the above embodiments, in one implementation, this application also provides a multi-mode radar fusion precipitation observation method. In this multi-mode radar fusion precipitation observation method, step S03 may include:

[0075] Step S03_1b: For any grid in the target city, obtain the amplitude spectrum of the second frequency domain spatiotemporal data of the grid preceding the current time for a predetermined number of consecutive times.

[0076] In this embodiment, for any grid cell in the target city, the amplitude spectrum of the grid cell in the second frequency domain spatiotemporal data of the most recent past period is obtained.

[0077] Step S03_2b: Based on the amplitude spectrum of the acquired consecutive preset number of second frequency domain spatiotemporal data, calculate the fluctuation intensity of the amplitude spectrum of the grid's past second frequency domain spatiotemporal data over time using the C-band fluctuation intensity algorithm; the C-band fluctuation intensity algorithm is as follows: Where T represents the number of amplitude spectra of the second frequency domain spatiotemporal data obtained from the grid up to the present. This represents the amplitude spectrum of the second frequency domain spatiotemporal data of the grid at time t. This represents the mean of the amplitude spectrum of the corresponding number of second-frequency-domain spatiotemporal data obtained. This represents the intensity of the amplitude spectrum fluctuation over time in the second frequency domain spatiotemporal data of the grid in the past.

[0078] In this embodiment, the amplitude spectrum of the second frequency domain spatiotemporal data of the grid over the most recent past period is calculated using the C-band wave intensity algorithm, thus obtaining the value of the wave intensity of the amplitude spectrum of the second frequency domain spatiotemporal data of the grid over time. In the C-band wave intensity algorithm... This represents the intensity of the amplitude spectrum fluctuation over time in the second frequency domain spatiotemporal data of the grid over the most recent past period. The subscript for time t is omitted to avoid conflict with the time t used for summation in the C-band wave intensity algorithm. T represents the number of amplitude spectra of the acquired consecutive preset number of second frequency domain spatiotemporal data.

[0079] Step S03_3b: If the fluctuation intensity is greater than the high frequency threshold, determine that the grid is currently a high frequency grid in the C band.

[0080] In this embodiment, if the fluctuation intensity obtained through step S03_2b is greater than the high-frequency threshold, it is determined that the grid is currently a high-frequency grid in the C-band. That is, this application determines the grid type of the grid in the C-band based on the amplitude spectrum of the grid over a past period, such as whether it is a high-frequency grid, a mid-frequency grid, or a low-frequency grid in the C-band.

[0081] Step S03_4b: If the fluctuation intensity is less than or equal to the low-frequency threshold, determine that the grid is currently a low-frequency grid in the C-band.

[0082] In this embodiment, if the fluctuation intensity obtained through step S03_2b is less than or equal to the low-frequency threshold, the grid is determined to be a low-frequency grid in the C-band.

[0083] Step S03_5b: If the fluctuation intensity is less than or equal to the high-frequency threshold and greater than the low-frequency threshold, determine that the grid is currently a mid-frequency grid in the C-band.

[0084] In this embodiment, if the fluctuation intensity obtained through step S03_2b is less than or equal to the high-frequency threshold and greater than the low-frequency threshold, the grid is determined to be a mid-frequency grid in the C-band. For example, if the high-frequency threshold is... The low-frequency threshold is The wave intensity obtained through step S03_2b satisfy At that time, it was determined that the current grid is a mid-frequency grid in the C-band.

[0085] In conjunction with the above embodiments, in one implementation, this application also provides a multi-mode radar fusion precipitation observation method. In this multi-mode radar fusion precipitation observation method, step S04 may include:

[0086] Step S04_1: Enhance the amplitude spectrum of the first frequency domain spatiotemporal data of the X-band high-frequency grid using the first enhancement algorithm to obtain the first enhanced amplitude spectrum of the first enhanced frequency domain spatiotemporal data of the X-band high-frequency grid; the first enhancement algorithm is: ;in, This represents the amplitude spectrum of the first frequency domain spatiotemporal data of the grid at time t, where γ represents the high-frequency enhancement coefficient of the X-band. This represents the fluctuation intensity of the grid at time t in the X-band. For high-frequency thresholds, This is the low-frequency threshold.

[0087] In this embodiment, for the amplitude spectrum of the first frequency domain spatiotemporal data of any grid belonging to the X-band high-frequency grid, a first enhancement algorithm is used to perform enhancement calculations on the amplitude spectrum of the first frequency domain spatiotemporal data of that grid, resulting in the first enhanced amplitude spectrum of the first enhanced frequency domain spatiotemporal data of that grid. Using the same implementation method, the first enhanced amplitude spectrum of each grid belonging to the X-band high-frequency grid can be calculated for itself in the X-band high frequency range. The first enhancement algorithm... This is an enhancement coefficient that is multiplied by the amplitude spectrum of the first frequency domain spatiotemporal data of the grid, as mentioned in the above implementation method.

[0088] Step S04_2: Based on the first enhanced amplitude spectrum of the first enhanced frequency domain spatiotemporal data of the X-band high-frequency grid, determine the enhanced power spectrum of the X-band high-frequency grid in the X-band high frequency.

[0089] In this embodiment, for any grid belonging to the X-band high-frequency grid, the first enhancement amplitude spectrum of the grid in the X-band high frequency is calculated to obtain the enhanced power spectrum of the grid in the X-band high frequency. The calculation expression is as follows: Using the same implementation method, the enhanced power spectrum of each grid belonging to the X-band high-frequency grid can be calculated in the X-band high frequency.

[0090] In this embodiment, in radar signal processing, frequency domain enhancement, by directly adjusting the frequency distribution coefficient of the signal, can effectively highlight the structural details of a specific frequency band. Since X-band radar signals are inherently rich in high-frequency detail information (related to spatial resolution), frequency domain enhancement is more conducive to improving the representation of local details. Power spectrum enhancement, on the other hand, is mainly used to adjust the energy distribution characteristics of the signal and is extremely effective for energy statistical characteristics, but it cannot directly affect the frequency coefficient itself. Therefore, this application uses frequency domain enhancement (i.e., enhancement of the amplitude spectrum, not the power spectrum) instead of power spectrum enhancement for the X-band, which can improve the distinguishability and detail representation of high-frequency local information, thereby more fully utilizing the high-resolution characteristics of the X-band.

[0091] In conjunction with the above embodiments, in one implementation, this application also provides a multi-mode radar fusion precipitation observation method. In this multi-mode radar fusion precipitation observation method, step S05 may include:

[0092] Step S05_1: Based on the amplitude spectrum of the second frequency domain spatiotemporal data of the C-band low-frequency grid, determine the power spectrum of the C-band low-frequency grid at the C-band low frequency.

[0093] In this embodiment, for the amplitude spectrum of the second frequency domain spatiotemporal data of any grid belonging to the C-band low-frequency grid, the amplitude spectrum of the second frequency domain spatiotemporal data of the grid is calculated to obtain the power spectrum of the grid in the C-band low frequency. The calculation expression is as follows: , For the C-band amplitude spectrum of the corresponding grid, This is the C-band power spectrum for the corresponding grid. Using the same implementation method, the power spectrum of each grid belonging to the C-band low-frequency grid can be calculated in the C-band low frequency range.

[0094] Step S05_2: The power spectrum of the C-band low-frequency grid at C-band low frequencies is enhanced using the second enhancement algorithm to obtain the enhanced power spectrum of the C-band low-frequency grid at C-band low frequencies; the second enhancement algorithm is as follows: ;in, This represents the power spectrum of the grid at time t. This represents the amplitude spectrum of the second frequency domain spatiotemporal data of the grid at time t. Indicates the C-band low-frequency enhancement factor. This represents the fluctuation intensity of the grid at time t in the C-band. For high-frequency thresholds, This is the low-frequency threshold.

[0095] In this embodiment, for the power spectrum of any grid belonging to the C-band low-frequency grid at C-band low frequencies, a second enhancement algorithm is used to enhance the power spectrum of that grid at C-band low frequencies, resulting in the enhanced power spectrum of that grid at C-band low frequencies. The second enhancement algorithm... This is the enhancement factor mentioned in the above embodiments, which is the power spectrum of the grid at low frequencies in the C-band.

[0096] In conjunction with the above embodiments, in one implementation, this application also provides a multi-mode radar fusion precipitation observation method. In this multi-mode radar fusion precipitation observation method, step S06 may include:

[0097] Step S06_1a: For any grid in the target city, where the grid is an intermediate frequency grid in both bands, calculate the power spectrum of the grid in the X-band intermediate frequency based on the amplitude spectrum of the first frequency domain spatiotemporal data of the grid, and calculate the power spectrum of the grid in the C-band intermediate frequency based on the amplitude spectrum of the second frequency domain spatiotemporal data of the grid.

[0098] In this embodiment, for any grid in the target city, if the grid is an intermediate frequency grid in both bands, the power spectrum of the grid in the X-band intermediate frequency is calculated based on the amplitude spectrum of the first frequency domain spatiotemporal data of the grid, and the power spectrum of the grid in the C-band intermediate frequency is calculated based on the amplitude spectrum of the second frequency domain spatiotemporal data of the grid.

[0099] Step S06_1b: Based on the relationship between the power spectrum of the grid at the intermediate frequency of a single band and the sum of the power spectra of the grid at the intermediate frequencies of two bands, determine the relative contribution weight of the grid in each band.

[0100] In this embodiment, the power spectrum of the grid at the intermediate frequency (IF) in the X-band and the power spectrum at the IF in the C-band are then summed to determine the ratio of the power spectrum of the grid at the IF in the X-band to the sum of the power spectra of the grid in both bands. This ratio is then determined as the relative contribution weight of the grid in the X-band. Simultaneously, the ratio of the power spectrum of the grid at the IF in the C-band to the sum of the power spectra of the grid in both bands is determined, and this ratio is also determined as the relative contribution weight of the grid in the C-band.

[0101] Step S06_1c: Based on the relative contribution weights of the grid in the two bands, the power spectrum of the grid at the intermediate frequency in the two bands is weighted and summed to obtain the power spectrum of the grid at the intermediate frequency.

[0102] Finally, based on the relative contribution weights of the grid in the two bands, the power spectra of the grid at the intermediate frequency (IF) in both bands are weighted and summed to obtain the power spectrum of the grid at the IF. The weighted summation method is to multiply the relative contribution weight of the grid in the X band by the power spectrum at the IF in the X band, and the relative contribution weight of the grid in the C band by the power spectrum at the IF in the C band. The two products are then added together to obtain the power spectrum of the grid at the IF. Using the same implementation method, each grid in the target city that is an IF grid in both bands can calculate its own power spectrum at the IF.

[0103] Step S06_2: For any grid in the target city, if the grid is an intermediate frequency grid only in the X-band, calculate the power spectrum of the grid in the intermediate frequency of the X-band based on the amplitude spectrum of the first frequency domain spatiotemporal data of the grid, and determine the power spectrum of the grid in the intermediate frequency of the X-band as the power spectrum of the grid in the intermediate frequency.

[0104] In this embodiment, for any grid cell in the target city, if the grid cell is an intermediate frequency (IF) grid in the X-band but not in the C-band, then based on the amplitude spectrum of the grid cell's first frequency domain spatiotemporal data, the power spectrum of the grid cell at the IF in the X-band is calculated, and the power spectrum of the grid cell at the IF in the X-band is directly determined as the power spectrum of the grid cell at the IF. Through the same implementation method, each grid cell in the target city that is an IF grid cell only in the X-band can calculate its own power spectrum at the IF.

[0105] Step S06_3: For any grid in the target city, if the grid is an intermediate frequency grid only in the C-band, calculate the power spectrum of the grid in the intermediate frequency of the C-band based on the amplitude spectrum of the second frequency domain spatiotemporal data of the grid, and determine the power spectrum of the grid in the intermediate frequency of the C-band as the power spectrum of the grid in the intermediate frequency.

[0106] In this embodiment, for any grid cell in the target city, if the grid cell is an intermediate frequency (IF) grid in the C-band but not in the X-band, then based on the amplitude spectrum of the grid cell's second frequency domain spatiotemporal data, the power spectrum of the grid cell at the C-band IF is calculated, and the power spectrum of the grid cell at the C-band IF is directly determined as the power spectrum of the grid cell at the IF. Through the same implementation method, each grid cell in the target city that is an IF grid cell only in the C-band can calculate its own power spectrum at the IF.

[0107] In conjunction with the above embodiments, in one implementation, this application also provides a multi-mode radar fusion precipitation observation method. In this multi-mode radar fusion precipitation observation method, step S07 may include:

[0108] Step S07_1: For any grid in the target city, if the grid is a high-frequency grid in the X-band and a high-frequency grid in the C-band, determine the enhanced power spectrum of the grid in the high frequency of the X-band as the fused power spectrum of the grid.

[0109] In this embodiment, as Figure 2 As shown, Figure 2 The data fusion process of multiple bands within the same grid at the same time is illustrated. For any grid in the target city, if the grid is a high-frequency grid in both the X-band and C-band, then the enhanced power spectrum of that grid in the high-frequency X-band is directly determined as the fused power spectrum of that grid.

[0110] Step S07_2: For any grid in the target city, if the grid is a high-frequency grid in the X-band and a mid-frequency grid in the C-band, fuse the enhanced power spectrum of the grid in the high-frequency X-band with the power spectrum of the grid in the mid-frequency to obtain the fused power spectrum of the grid.

[0111] In this embodiment, for any grid cell in the target city, if the grid cell is a high-frequency grid in the X-band and a mid-frequency grid in the C-band, then the enhanced power spectrum of the grid cell in the high-frequency X-band and the power spectrum of the grid cell in the mid-frequency are fused to obtain the fused power spectrum of the grid cell. Specifically, for a grid cell that is a high-frequency grid in the X-band and a mid-frequency grid in the C-band, a weight is assigned to the enhanced power spectrum of the grid cell in the high-frequency X-band, and a weight is assigned to the power spectrum of the grid cell in the mid-frequency. Then, the enhanced power spectrum of the grid cell in the high-frequency X-band and the power spectrum of the grid cell in the mid-frequency are weighted and summed to obtain the fused power spectrum of the grid cell. In this embodiment, the sum of the two weights is 1, and the weight corresponding to the enhanced power spectrum of the grid cell in the high-frequency X-band is greater than the weight corresponding to the power spectrum of the grid cell in the mid-frequency.

[0112] Step S07_3: For any grid in the target city, if the grid is a high-frequency grid in the X-band and a low-frequency grid in the C-band, fuse the enhanced power spectrum of the grid in the high-frequency X-band with the enhanced power spectrum of the grid in the low-frequency C-band to obtain the fused power spectrum of the grid.

[0113] In this embodiment, for any grid cell in the target city, if the grid cell is a high-frequency grid in the X-band and a low-frequency grid in the C-band, then the enhanced power spectrum of the grid cell in the high-frequency X-band and the enhanced power spectrum of the grid cell in the low-frequency C-band are fused to obtain the fused power spectrum of the grid cell. Specifically, for a grid cell that is a high-frequency grid in the X-band and a low-frequency grid in the C-band, a weight is assigned to the enhanced power spectrum of the grid cell in the high-frequency X-band, and a weight is assigned to the enhanced power spectrum of the grid cell in the low-frequency C-band. Then, the enhanced power spectrum of the grid cell in the high-frequency X-band and the enhanced power spectrum of the grid cell in the low-frequency C-band are weighted and summed to obtain the fused power spectrum of the grid cell. In this embodiment, the sum of the two weights is 1.

[0114] Step S07_4: For any grid in the target city, if the grid is a mid-frequency grid in the X-band and a high-frequency grid in the C-band, determine the power spectrum of the grid at the mid-frequency as the fused power spectrum of the grid.

[0115] In this embodiment, for any grid in the target city, if the grid is a mid-frequency grid in the X-band and a high-frequency grid in the C-band, then the power spectrum of the grid in the mid-frequency range is directly determined as the fused power spectrum of the grid.

[0116] Step S07_5: For any grid in the target city, if the grid is an intermediate frequency grid in the X-band and an intermediate frequency grid in the C-band, determine the power spectrum of the grid at the intermediate frequency as the fused power spectrum of the grid.

[0117] In this embodiment, for any grid in the target city, if the grid is an intermediate frequency grid in the X-band and an intermediate frequency grid in the C-band, then the power spectrum of the grid in the intermediate frequency is directly determined as the fused power spectrum of the grid.

[0118] Step S07_6: For any grid in the target city, if the grid is a mid-frequency grid in the X-band and a low-frequency grid in the C-band, fuse the power spectrum of the grid at the mid-frequency with the enhanced power spectrum of the grid at the low frequency of the C-band to obtain the fused power spectrum of the grid.

[0119] In this embodiment, for any grid cell in the target city, if the grid cell is a mid-frequency grid in the X-band and a low-frequency grid in the C-band, then the power spectrum of the grid cell at the mid-frequency level is fused with the enhanced power spectrum of the grid cell at the low frequency in the C-band to obtain the fused power spectrum of the grid cell. Specifically, for a grid cell that is a mid-frequency grid in the X-band and a low-frequency grid in the C-band, a weight is assigned to the power spectrum of the grid cell at the mid-frequency level, and a weight is assigned to the enhanced power spectrum of the grid cell at the low frequency in the C-band. Then, the power spectrum of the grid cell at the mid-frequency level and the enhanced power spectrum of the grid cell at the low frequency in the C-band are weighted and summed to obtain the fused power spectrum of the grid cell. In this embodiment, the sum of the two weights is 1, and the weight corresponding to the power spectrum of the grid cell at the mid-frequency level is less than the weight corresponding to the enhanced power spectrum of the grid cell at the low frequency in the C-band.

[0120] Step S07_7: For any grid in the target city, if the grid is a low-frequency grid in the X-band and a high-frequency grid in the C-band, determine that the fused power spectrum of the grid is zero.

[0121] In this embodiment, for any grid cell in the target city, if the grid cell is a low-frequency grid in the X-band and a high-frequency grid in the C-band, the fused power spectrum of the grid cell is directly set to zero. This is because the grid cell performs poorly at low frequencies in the X-band and also performs poorly at high frequencies in the C-band; therefore, it is directly set to zero. Subsequently, the power spectrum of the grid cell is obtained by interpolation based on the power spectra of surrounding grid cells.

[0122] Step S07_8: For any grid in the target city, if the grid is a low-frequency grid in the X-band and a mid-frequency grid in the C-band, determine the power spectrum of the grid at the mid-frequency as the fused power spectrum of the grid.

[0123] In this embodiment, for any grid in the target city, if the grid is a low-frequency grid in the X-band and a mid-frequency grid in the C-band, then the power spectrum of the grid in the mid-frequency band is directly determined as the fused power spectrum of the grid.

[0124] Step S07_9: For any grid in the target city, if the grid is a low-frequency grid in the X-band and a low-frequency grid in the C-band, determine the enhanced power spectrum of the grid at the low frequency of the C-band as the fused power spectrum of the grid.

[0125] In this embodiment, for any grid in the target city, if the grid is a low-frequency grid in both the X-band and C-band, then the enhanced power spectrum of the grid in the low-frequency C-band is directly determined as the fused power spectrum of the grid.

[0126] Step S07_10: For any grid in the target city, fuse the phase spectrum of the first frequency domain spatiotemporal data and the phase spectrum of the second frequency domain spatiotemporal data of the grid to obtain the fused phase spectrum of the grid.

[0127] In this embodiment, for any grid in the target city, the phase spectrum of the first frequency domain spatiotemporal data and the phase spectrum of the second frequency domain spatiotemporal data of the grid are fused by averaging to obtain the fused phase spectrum of the grid.

[0128] Step S07_11: Based on the fused power spectrum and fused phase spectrum of each grid in the target city, obtain the fused spectrum of the target city in units of grids.

[0129] In this embodiment, the fusion power spectrum and fusion phase spectrum of each grid in the target city can be obtained through the above steps S07_1 to S07_10. The fusion power spectrum and fusion phase spectrum of a grid constitute the fusion frequency of that grid. That is, each grid in the target city can calculate its own current fusion spectrum, thereby forming the fusion spectrum of the target city on a grid-by-grid basis.

[0130] In conjunction with the above embodiments, in one implementation, this application also provides a precipitation observation method based on multi-mode radar fusion. This multi-mode radar fusion precipitation observation method further includes: identifying grid cells with a value of zero in the fusion power spectrum of a target city that need correction; and spatially interpolating and completing the fusion power spectrum of the grid cells to be corrected based on the fusion power spectra of other grid cells surrounding the grid cells to be corrected.

[0131] In this embodiment, after obtaining the fused spectrum of each grid in the target city, the grids with a fused power spectrum of zero are identified and referred to as grids to be corrected. For any grid to be corrected, spatial interpolation is performed on the fused power spectrum of the grid to be corrected based on the fused power spectra of the surrounding grids to obtain the value of the fused power spectrum of the grid to be corrected. Through the same implementation method, a corresponding fused power spectrum can be calculated for each grid to be corrected in the target city. Optional spatial interpolation methods include Kriging interpolation, nearest neighbor interpolation, etc.

[0132] In conjunction with the above embodiments, in one implementation, this application also provides a precipitation observation method based on multi-mode radar fusion. This multi-mode radar fusion precipitation observation method further includes: collecting spatiotemporal datasets of X-band radar and C-band radar for a target city during the same period; performing noise removal processing on the spatiotemporal data in the two band datasets; performing coordinate transformation on the noise-removed spatiotemporal data in the two band datasets to transform the two band datasets to the same coordinate system; and resampling the spatiotemporal data of the C-band radar after coordinate system one using nearest neighbor interpolation, based on the spatial resolution of the X-band radar spatiotemporal data after coordinate system one, to spatially align the spatiotemporal data of the two bands after coordinate system one, resulting in spatially resolution-aligned X-band radar spatiotemporal datasets and C-band radar spatiotemporal datasets.

[0133] In this embodiment, the spatiotemporal data of each band radar may be interfered with by various noise sources during the acquisition process, such as system noise and atmospheric noise. To improve data quality, this application performs noise removal processing on the acquired spatiotemporal data of each band radar. One optional approach is to use a median filtering algorithm to filter the spatiotemporal data of the two band radars, removing small outliers, smoothing the data, and retaining important precipitation signals. For the removed spatiotemporal data, spatial interpolation is used to complete it, to avoid some grids lacking spatiotemporal data for the corresponding band radar. Secondly, coordinate system transformation is performed on the spatiotemporal data of the two band radars after noise removal to unify the spatiotemporal data of the two band radars to the same coordinate system. Finally, since the spatial resolution of X-band radar and C-band radar is generally different, with X-band usually having a higher resolution, this application uses the nearest neighbor interpolation method to resample the spatiotemporal data of C-band radar after unifying the coordinate system, to align the spatial resolution of the C-band radar spatiotemporal data with that of X-band radar.

[0134] In conjunction with the above embodiments, in one implementation, this application also provides a multi-mode radar fusion precipitation observation method. In this multi-mode radar fusion precipitation observation method, the method further includes:

[0135] Step S09a: Normalize the fused spatiotemporal dataset using a normalization algorithm to obtain the target fused spatiotemporal dataset; the expression for the normalization algorithm is: ;in, This represents the target fusion spatiotemporal data of the grid (x,y) in the spatial domain at time t. This represents the fused spatiotemporal data of the grid (x,y) in the spatial domain at time t. This represents the maximum spatiotemporal data in the initial C-band radar spatiotemporal dataset. This represents the minimum spatiotemporal data in the initial spatiotemporal dataset of the C-band radar. This represents the largest fused spatiotemporal data point in the fused spatiotemporal dataset. This represents the smallest fused spatiotemporal data in the fused spatiotemporal dataset.

[0136] In this embodiment, due to the shorter wavelength of the X-band, the attenuation of atmospheric particles such as precipitation and clouds is significantly stronger than that of the C-band. Therefore, the observation values ​​of X-band radar at long distances or in areas of heavy precipitation may be significantly lower. If C-band data is adjusted based on the X-band, it may lead to overcompensation of C-band data at long distances, introducing false high values. C-band radar is typically designed for medium- to long-range, large-area monitoring, and its dynamic range is optimized to cover the echo intensity of typical meteorological targets; X-band radar is often used for refined, short-range observation, and its original dynamic range may be narrower. Therefore, aligning the X-band to the C-band ensures that the fused data maintains a consistent intensity scale throughout the entire observation area. Based on this, after obtaining the fused spatiotemporal dataset of the target city, this application normalizes it to ensure that the fusion result conforms to the radar's physical characteristics, making its dynamic range consistent with that of the C-band data and avoiding overall intensity deviation caused by frequency domain weighting. Specifically, the fused spatiotemporal dataset is normalized using a normalization algorithm to obtain the target fused spatiotemporal dataset. This represents the maximum spatiotemporal data in the initial spatiotemporal dataset of the C-band radar, which is the maximum spatiotemporal data in the spatiotemporal dataset of the C-band radar that was first acquired and aligned to a spatial resolution. This represents the minimum spatiotemporal data in the initial spatiotemporal dataset of the C-band radar, which is the minimum spatiotemporal data in the spatiotemporal dataset of the C-band radar that was first acquired and aligned to a spatial resolution.

[0137] In this application, if the method further includes step S09a, step S09 may include: observing the precipitation of the target city based on the target fused spatiotemporal dataset.

[0138] In this embodiment, if the method further includes step S09a, then step S09 observes the precipitation situation of the target city based on the normalized target fusion spatiotemporal dataset.

[0139] Based on the same inventive concept, this application provides a multi-mode radar fusion precipitation observation system, such as... Figure 3 As shown, the system 300 includes:

[0140] The data acquisition module 301 is used to acquire the spatiotemporal datasets of the X-band radar and the C-band radar of the target city, which are currently aligned in spatial resolution.

[0141] Fourier transform module 302 is used to transform the spatiotemporal datasets of two bands from the spatial domain to the frequency domain through Fourier transform, so as to obtain the first frequency domain spatiotemporal dataset of X band and the second frequency domain spatiotemporal dataset of C band.

[0142] The grid type determination module 303 is used to determine the grid type of any grid in the target city in the X-band based on the fluctuation intensity of the amplitude spectrum of the first frequency domain spatiotemporal data of the grid over time, and to determine the grid type of the grid in the C-band based on the fluctuation intensity of the amplitude spectrum of the second frequency domain spatiotemporal data of the grid over time. The grid type includes high-frequency grid, mid-frequency grid and low-frequency grid.

[0143] The first power spectrum determination module 304 is used to determine the enhanced power spectrum of the X-band high-frequency grid in the X-band high frequency based on the amplitude spectrum of the first frequency domain spatiotemporal data of the X-band high-frequency grid.

[0144] The second power spectrum determination module 305 is used to determine the enhanced power spectrum of the C-band low-frequency grid at the C-band low frequency based on the amplitude spectrum of the second frequency domain spatiotemporal data of the C-band low-frequency grid.

[0145] The third power spectrum determination module 306 is used to determine the power spectrum of the grid at the intermediate frequency by selecting the amplitude spectrum of the corresponding frequency domain spatiotemporal data based on the intermediate frequency state of the grid in two bands; wherein, when both bands are intermediate frequencies, the amplitude spectrum of the first frequency domain spatiotemporal data and the amplitude spectrum of the second frequency domain spatiotemporal data of the grid are selected; when only the X band is an intermediate frequency, the amplitude spectrum of the first frequency domain spatiotemporal data of the grid is selected; when only the C band is an intermediate frequency, the amplitude spectrum of the second frequency domain spatiotemporal data of the grid is selected.

[0146] The power spectrum fusion module 307 is used to fuse the power spectra of each grid in any grid in the target city, and to fuse the phase spectrum of the first frequency domain spatiotemporal data and the phase spectrum of the second frequency domain spatiotemporal data of the grid to obtain the fused spectrum of the target city in units of grids.

[0147] The inverse Fourier transform module 308 is used to transform the fusion spectrum of the target city from the frequency domain to the spatial domain through a two-dimensional inverse discrete Fourier transform to obtain a fusion spatiotemporal dataset.

[0148] The precipitation observation module 309 is used to observe the precipitation situation of the target city based on the fused spatiotemporal dataset.

[0149] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily necessary for the embodiments of this application.

[0150] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0151] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, embodiments of this application can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of this application can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0152] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1A device that provides the functions specified in one or more boxes.

[0153] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0154] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0155] Although preferred embodiments of the present application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present application.

[0156] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0157] The above provides a detailed description of a multi-mode radar fusion precipitation observation method and system provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A precipitation observation method using multi-mode radar fusion, characterized in that, The method includes: Obtain the spatiotemporal datasets of the X-band radar and C-band radar currently aligned in spatial resolution for the target city; The spatiotemporal datasets of the two bands are transformed from the spatial domain to the frequency domain by Fourier transform, resulting in the first frequency domain spatiotemporal dataset of the X-band and the second frequency domain spatiotemporal dataset of the C-band. For any grid in the target city, based on the fluctuation intensity of the amplitude spectrum of the first frequency domain spatiotemporal data of the grid over time, the grid type of the grid in the current X-band is determined, and based on the fluctuation intensity of the amplitude spectrum of the second frequency domain spatiotemporal data of the grid over time, the grid type of the grid in the current C-band is determined, wherein the grid type includes high-frequency grid, mid-frequency grid and low-frequency grid; Based on the amplitude spectrum of the first frequency domain spatiotemporal data of the X-band high-frequency grid, the enhanced power spectrum of the X-band high-frequency grid in the X-band high frequency is determined. Based on the amplitude spectrum of the second frequency domain spatiotemporal data of the C-band low-frequency grid, the enhanced power spectrum of the C-band low-frequency grid at the C-band low frequency is determined. Based on the intermediate frequency (IF) state of the grid in the two bands, the amplitude spectrum of the corresponding frequency domain spatiotemporal data is selected to determine the power spectrum of the grid at the IF. Specifically, when both bands are IF, the amplitude spectrum of the first frequency domain spatiotemporal data and the amplitude spectrum of the second frequency domain spatiotemporal data of the grid are selected; when only the X band is IF, the amplitude spectrum of the first frequency domain spatiotemporal data of the grid is selected; when only the C band is IF, the amplitude spectrum of the second frequency domain spatiotemporal data of the grid is selected. For any grid in the target city, the power spectra of each grid are fused, and the phase spectra of the first frequency domain spatiotemporal data and the second frequency domain spatiotemporal data of the grid are fused to obtain the fused spectrum of the target city in units of grids. The fused spectrum of the target city is transformed from the frequency domain to the spatial domain by using a two-dimensional inverse discrete Fourier transform to obtain a fused spatiotemporal dataset. Based on the fused spatiotemporal dataset, observe the precipitation situation of the target city; Among them, determining the enhanced power spectrum of the X-band high-frequency grid in the X-band high frequency range based on the amplitude spectrum of the first frequency domain spatiotemporal data of the X-band high-frequency grid includes: The amplitude spectrum of the first frequency domain spatiotemporal data of the X-band high-frequency grid is enhanced by the first enhancement algorithm to obtain the first enhanced amplitude spectrum of the first enhanced frequency domain spatiotemporal data of the X-band high-frequency grid. Based on the first enhanced amplitude spectrum of the first enhanced frequency domain spatiotemporal data of the X-band high-frequency grid, the enhanced power spectrum of the X-band high-frequency grid in the X-band high frequency is determined; The first enhancement algorithm is: ; in, This represents the amplitude spectrum of the first frequency domain spatiotemporal data of the grid at time t, where γ represents the high-frequency enhancement coefficient of the X-band. This represents the fluctuation intensity of the grid at time t in the X-band. For high-frequency thresholds, Here, (u,v,t) represents the low-frequency threshold, (u,v,t) represents the spatiotemporal coordinates of the grid (u,v) in the frequency domain, and t represents the time. Among them, determining the enhanced power spectrum of the C-band low-frequency grid at C-band low frequencies based on the amplitude spectrum of the second frequency domain spatiotemporal data of the C-band low-frequency grid includes: Based on the amplitude spectrum of the second frequency domain spatiotemporal data of the C-band low-frequency grid, the power spectrum of the C-band low-frequency grid at the C-band low frequency is determined. The power spectrum of the C-band low-frequency grid at C-band low frequency is enhanced by the second enhancement algorithm to obtain the enhanced power spectrum of the C-band low-frequency grid at C-band low frequency. The second enhancement algorithm is: ; in, This represents the power spectrum of the grid at time t. This represents the amplitude spectrum of the second frequency domain spatiotemporal data of the grid at time t. Indicates the C-band low-frequency enhancement factor. This represents the fluctuation intensity of the grid at time t in the C-band. This is the low-frequency threshold.

2. The precipitation observation method based on multi-mode radar fusion according to claim 1, characterized in that, For any grid cell in the target city, based on the fluctuation intensity of the amplitude spectrum of the grid cell's past first-frequency-domain spatiotemporal data over time, the current grid cell type in the X-band is determined, including: For any grid cell in the target city, obtain the amplitude spectrum of a first frequency domain spatiotemporal data of the grid cell for a consecutive preset number of consecutive times before the current time; Based on the amplitude spectrum of the first frequency domain spatiotemporal data of the obtained continuous preset number, the fluctuation intensity of the amplitude spectrum of the first frequency domain spatiotemporal data of the grid over time is calculated by the X-band fluctuation intensity algorithm. If the fluctuation intensity is greater than the high-frequency threshold, the grid is determined to be a high-frequency grid in the X-band. If the fluctuation intensity is less than or equal to the low-frequency threshold, the grid is determined to be a low-frequency grid in the X-band. If the fluctuation intensity is less than or equal to the high-frequency threshold and greater than the low-frequency threshold, the grid is determined to be a mid-frequency grid in the X-band. The algorithm for X-band wave intensity is as follows: ; Where T represents the number of amplitude spectra of the first frequency domain spatiotemporal data of the acquired grid prior to the current time. This represents the amplitude spectrum of the first frequency domain spatiotemporal data of the grid at time t. This represents the mean of the amplitude spectrum of the corresponding number of first-frequency-domain spatiotemporal data obtained. This represents the intensity of the amplitude spectrum fluctuation over time in the first frequency domain spatiotemporal data of the grid.

3. The precipitation observation method based on multi-mode radar fusion according to claim 1, characterized in that, Based on the intermediate frequency (IF) status of the grid in the two bands, the amplitude spectrum of the corresponding frequency domain spatiotemporal data is selected to determine the power spectrum of the grid at the IF, including: For any grid in the target city, where the grid is a mid-frequency grid in both bands, the power spectrum of the grid in the X-band mid-frequency is calculated based on the amplitude spectrum of the first frequency domain spatiotemporal data of the grid, and the power spectrum of the grid in the C-band mid-frequency is calculated based on the amplitude spectrum of the second frequency domain spatiotemporal data of the grid. Based on the relationship between the power spectrum of the grid at the intermediate frequency of a single band and the sum of the power spectra of the grid at the intermediate frequencies of two bands, the relative contribution weight of the grid in each band is determined. Based on the relative contribution weights of the grid in the two bands, the power spectrum of the grid at the intermediate frequency in the two bands is weighted and summed to obtain the power spectrum of the grid at the intermediate frequency. For any grid in the target city, where the grid is an intermediate frequency grid only in the X-band, the power spectrum of the grid in the intermediate frequency of the X-band is calculated based on the amplitude spectrum of the first frequency domain spatiotemporal data of the grid, and the power spectrum of the grid in the intermediate frequency of the X-band is determined as the power spectrum of the grid in the intermediate frequency. For any grid in the target city, where the grid is an intermediate frequency grid only in the C-band, the power spectrum of the grid in the intermediate frequency of the C-band is calculated based on the amplitude spectrum of the second frequency domain spatiotemporal data of the grid, and the power spectrum of the grid in the intermediate frequency of the C-band is determined as the power spectrum of the grid in the intermediate frequency.

4. The precipitation observation method based on multi-mode radar fusion according to claim 3, characterized in that, For any grid cell in the target city, the power spectra of each grid cell are fused, and the phase spectra of the first frequency domain spatiotemporal data and the second frequency domain spatiotemporal data of the grid cell are fused to obtain the fused spectrum of the target city on a grid cell basis, including: For any grid in the target city, if the grid is a high-frequency grid in the X-band and a high-frequency grid in the C-band, the enhanced power spectrum of the grid in the high frequency of the X-band is determined as the fused power spectrum of the grid. For any grid in the target city, if the grid is a high-frequency grid in the X-band and a mid-frequency grid in the C-band, the enhanced power spectrum of the grid in the high frequency of the X-band is fused with the power spectrum of the grid in the mid-frequency to obtain the fused power spectrum of the grid. For any grid in the target city, if the grid is a high-frequency grid in the X-band and a low-frequency grid in the C-band, the enhanced power spectrum of the grid in the high frequency of the X-band is fused with the enhanced power spectrum of the grid in the low frequency of the C-band to obtain the fused power spectrum of the grid. For any grid in the target city, if the grid is a mid-frequency grid in the X-band and a high-frequency grid in the C-band, the power spectrum of the grid at the mid-frequency is determined as the fused power spectrum of the grid. For any grid in the target city, if the grid is a mid-frequency grid in the X-band and a mid-frequency grid in the C-band, the power spectrum of the grid at the mid-frequency is determined as the fused power spectrum of the grid. For any grid in the target city, if the grid is a mid-frequency grid in the X-band and a low-frequency grid in the C-band, the power spectrum of the grid at the mid-frequency is fused with the enhanced power spectrum of the grid at the low frequency of the C-band to obtain the fused power spectrum of the grid. For any grid in the target city, if the grid is a low-frequency grid in the X-band and a high-frequency grid in the C-band, the fusion power spectrum of the grid is determined to be zero. For any grid in the target city, if the grid is a low-frequency grid in the X-band and a mid-frequency grid in the C-band, the power spectrum of the grid at the mid-frequency is determined as the fused power spectrum of the grid. For any grid in the target city, if the grid is a low-frequency grid in the X-band and a low-frequency grid in the C-band, the enhanced power spectrum of the grid in the low frequency of the C-band is determined as the fused power spectrum of the grid. For any grid in the target city, the phase spectrum of the first frequency domain spatiotemporal data and the phase spectrum of the second frequency domain spatiotemporal data of the grid are fused to obtain the fused phase spectrum of the grid. Based on the fused power spectrum and fused phase spectrum of each grid in the target city, the fused spectrum of the target city is obtained on a grid-by-grid basis.

5. The precipitation observation method based on multi-mode radar fusion according to claim 4, characterized in that, The method further includes: Identify the grid cells to be corrected that have a value of zero in the fusion power spectrum of the target city; Based on the fused power spectra of other grids surrounding the grid to be corrected, spatial interpolation is performed to complete the fused power spectrum of the grid to be corrected.

6. The precipitation observation method based on multi-mode radar fusion according to claim 1, characterized in that, The method further includes: Spatiotemporal datasets of X-band radar and C-band radar were collected for the target city during the same period. Noise removal processing is performed on the spatiotemporal data in the spatiotemporal datasets of the two bands; The spatiotemporal data in the two bands after noise removal are transformed to convert the spatiotemporal data of the two bands to the same coordinate system. Based on the spatial resolution of the spatiotemporal data of the X-band radar after coordinate system one, the spatiotemporal data of the C-band radar after coordinate system one is resampled by the nearest neighbor interpolation method to align the spatial resolution of the spatiotemporal data of the two bands after coordinate system one, so as to obtain the spatial resolution aligned spatiotemporal datasets of the X-band radar and the C-band radar.

7. The precipitation observation method based on multi-mode radar fusion according to claim 1, characterized in that, The method further includes: The fused spatiotemporal dataset is normalized using a normalization algorithm to obtain the target fused spatiotemporal dataset. The step of observing the precipitation situation of the target city based on the fused spatiotemporal dataset includes: observing the precipitation situation of the target city based on the target fused spatiotemporal dataset; The expression for the normalization algorithm is: ; in, This represents the target fusion spatiotemporal data of the grid (x,y) in the spatial domain at time t. This represents the fused spatiotemporal data of the grid (x,y) in the spatial domain at time t. This represents the maximum spatiotemporal data in the initial C-band radar spatiotemporal dataset. This represents the minimum spatiotemporal data in the initial spatiotemporal dataset of the C-band radar. This represents the largest fused spatiotemporal data point in the fused spatiotemporal dataset. This represents the smallest fused spatiotemporal data in the fused spatiotemporal dataset.

8. A multi-mode radar fusion precipitation observation system, characterized in that, A precipitation observation method based on multi-mode radar fusion according to any one of claims 1 to 7, the system comprising: The data acquisition module is used to acquire the spatiotemporal datasets of the X-band radar and the C-band radar of the target city, which are currently aligned in spatial resolution. The Fourier transform module is used to transform the spatiotemporal datasets of two bands from the spatial domain to the frequency domain through Fourier transform, so as to obtain the first frequency domain spatiotemporal dataset of the X-band and the second frequency domain spatiotemporal dataset of the C-band. The grid type determination module is used to determine the current grid type of any grid in the target city in the X-band based on the fluctuation intensity of the amplitude spectrum of the grid's past first frequency domain spatiotemporal data over time, and to determine the current grid type of the grid in the C-band based on the fluctuation intensity of the amplitude spectrum of the grid's past second frequency domain spatiotemporal data over time. The grid type includes high-frequency grid, mid-frequency grid, and low-frequency grid. The first power spectrum determination module is used to determine the enhanced power spectrum of the X-band high-frequency grid in the X-band high frequency based on the amplitude spectrum of the first frequency domain spatiotemporal data of the X-band high-frequency grid. The second power spectrum determination module is used to determine the enhanced power spectrum of the C-band low-frequency grid at the C-band low frequency based on the amplitude spectrum of the second frequency domain spatiotemporal data of the C-band low-frequency grid. The third power spectrum determination module is used to select the amplitude spectrum of the corresponding frequency domain spatiotemporal data based on the intermediate frequency state of the grid in two bands, and determine the power spectrum of the grid at the intermediate frequency. Specifically, when both bands are intermediate frequencies, the amplitude spectrum of the first frequency domain spatiotemporal data and the amplitude spectrum of the second frequency domain spatiotemporal data of the grid are selected; when only the X band is an intermediate frequency, the amplitude spectrum of the first frequency domain spatiotemporal data of the grid is selected; when only the C band is an intermediate frequency, the amplitude spectrum of the second frequency domain spatiotemporal data of the grid is selected. The power spectrum fusion module is used to fuse the power spectra of each grid in any grid in the target city, and to fuse the phase spectrum of the first frequency domain spatiotemporal data and the phase spectrum of the second frequency domain spatiotemporal data of the grid to obtain the fused spectrum of the target city in units of grids. The inverse Fourier transform module is used to transform the fused spectrum of the target city from the frequency domain to the spatial domain through a two-dimensional inverse discrete Fourier transform to obtain a fused spatiotemporal dataset. The precipitation observation module is used to observe the precipitation situation of the target city based on the fused spatiotemporal dataset.

Images