Method and system for step-by-step filtering of false signals of lightning based on principal component analysis

Abstract

The embodiment of the application discloses a lightning false signal step-by-step filtering method and system based on principal component analysis, relates to the meteorological remote sensing technical field, and comprises the following steps: S1, obtaining preprocessed transient light signals; S2, performing transient light signal level pseudo-continuous false signal filtering on the transient light signals to be processed; S3, further performing transient light signal clustering to generate transient light signal groups; S4, performing transient light signal group level pseudo-continuous false signal filtering to obtain the preprocessed transient light signal groups; S5, performing space-time clustering analysis on the transient light signal groups to obtain the lightning corresponding to the transient light signals to be processed; and S6, sequentially performing lightning level pseudo-continuous false signal filtering and lightning discrete false signal filtering on the lightning corresponding to the transient light signals to be processed. The above steps S2-S6 are performed on the transient light signals to be processed until the number of transient light signals before filtering is consistent with the number of transient light signals after filtering, the lightning false signal step-by-step filtering is completed, and the real lightning signal is output.

Description

Technical Field

[0001] This specification relates to the field of meteorological remote sensing technology, and in particular to a method and system for stepwise filtering of false lightning signals based on principal component analysis. Background Technology

[0002] Lightning is an atmospheric electrical discharge phenomenon, primarily originating from the charge separation process driven by strong updrafts in cumulonimbus clouds. It is often accompanied by severe convective weather such as heavy rain, hail, and tornadoes. Therefore, strengthening real-time, continuous, and large-scale monitoring of lightning is crucial for deeply understanding the patterns of lightning activity and improving the monitoring and early warning capabilities for severe convective weather.

[0003] Currently, lightning monitoring primarily relies on two core methods: ground-based positioning systems and space-based detection payloads. Compared to ground-based observation, lightning imagers mounted on geostationary orbit (GEO) meteorological satellites offer significant advantages in achieving large-scale, continuous lightning monitoring. The peak power of a lightning discharge typically reaches 10... 9 -10 10 On the order of W, satellite lightning imagers primarily locate lightning by detecting the emission lines of neutral oxygen atoms in the near-infrared band (777.4 nm). However, since this band still falls within the continuous spectrum of solar radiation, strong cloud top reflections often introduce complex background noise during daytime observations; especially when the sun directly enters the field of view or the satellite's attitude is jittery, halos or streaks often appear in the images. Furthermore, the inherent shot noise of photoelectric detection devices and bombardment by high-energy particles in space are unavoidable sources of interference. These factors result in a large amount of non-lightning signals in the raw satellite observation data. Statistics show that false signals account for 80%–90% of lightning data from polar-orbiting satellites and around 60% for geostationary satellites. The presence of these false signals not only severely restricts the efficiency of satellite lightning detection but also greatly reduces the effectiveness of lightning data in monitoring and warning of severe convective weather.

[0004] Currently, various spurious signal filtering algorithms have been proposed internationally for different types of non-lightning signals, such as S / C motion noise, solar reflection noise, saturation overflow noise, explosion noise, shot noise, ghost noise, and impact / line noise. These include ghost removal algorithms, coherence algorithms, and solar reflection algorithms. However, these algorithms are usually designed based on simulated single-source noise, ignoring the complexity of superimposed spurious signals from multiple different sources. For example, shot noise, the most common and prevalent type of lightning signal, theoretically follows a Poisson distribution. Its biggest difference from real lightning is that real lightning signals have strong temporal and spatial correlations, while shot noise generally manifests as random, isolated events in time and space. However, sometimes, due to the influence of CCD charge transfer processes, some originally isolated shot noise points exhibit a pseudo-continuity, displaying "short lines" or "spots" of structural characteristics, approximating S / C motion noise or impact / line noise. Therefore, using highly targeted noise algorithms may result in either missing spurious signals or over-filtering real lightning signals. Current highly targeted satellite lightning noise filtering algorithms are not efficient at filtering out spurious signals in real satellite lightning observation data.

[0005] Therefore, a method and system for stepwise filtering of false lightning signals based on principal component analysis is needed. Summary of the Invention

[0006] This specification provides a method and system for step-by-step filtering of lightning spurious signals based on principal component analysis, to address the following technical problem: Currently, various spurious signal filtering algorithms have been proposed internationally for different types of non-lightning signals, such as S / C motion noise, solar reflection noise, saturation overflow noise, explosion noise, shot noise, ghost noise, and impact / line noise, including ghosting algorithms, coherence algorithms, and solar reflection algorithms. However, these algorithms are usually designed based on simulated single-source noise, ignoring the complexity of the superposition of spurious signals from multiple different sources. For example, shot noise, the most common and predominant type of lightning signal, theoretically follows a Poisson distribution. Its biggest difference from real lightning is that real lightning signals have strong correlation in both time and space, while shot noise generally manifests as random, isolated events in time and space. However, sometimes, due to the influence of CCD charge transfer processes, some originally isolated shot noise points exhibit a pseudo-continuity, displaying "short lines" or "spots" of structural characteristics, approximating S / C motion noise or impact / line noise. Therefore, using highly targeted noise removal algorithms may result in either missing false signals or over-filtering out genuine lightning signals. Current highly targeted satellite lightning noise removal algorithms are not efficient at filtering out false signals from real satellite lightning observation data.

[0007] To solve the above-mentioned technical problems, the embodiments in this specification are implemented as follows:

[0008] This specification provides a step-by-step filtering method for lightning spurious signals based on principal component analysis. This method is applied to lightning observation data from geostationary meteorological satellites and includes:

[0009] S1: Acquire the transient optical signal to be processed;

[0010] S2: Based on the aspect ratio and energy variation coefficient of the transient light signal within the sliding time window of the transient light signal to be processed, perform pseudo-continuous false signal filtering at the transient light signal level to obtain the preprocessed transient light signal.

[0011] S3: After determining the clustering core based on the spatial index grid of the transient light signal determined by the preprocessed transient light signal, a first sliding window is constructed, and the transient light signals that meet the first preset conditions within the first sliding window are clustered to generate a transient light signal group.

[0012] S4: Based on the aspect ratio of the transient optical signal group, the first two principal components of the transient optical signal group principal component analysis, and the energy variation coefficient of the transient optical signal group, pseudo-continuous false signals are filtered out at the transient optical signal group level to obtain the preprocessed transient optical signal group.

[0013] S5: Using the spatial index grid of the transient light signal group determined by the preprocessed transient light signal group, construct a second sliding window, and perform spatiotemporal clustering analysis on the transient light signal groups that meet the second preset conditions within the second sliding window to obtain the lightning corresponding to the transient light signal to be processed;

[0014] S6: Perform lightning-level pseudo-continuous false signal filtering and lightning discrete false signal filtering sequentially on the lightning corresponding to the transient light signal to be processed;

[0015] Repeat steps S2-S6 for the transient light signal to be processed until the number of transient light signals before and after filtering is the same, thus completing the filtering of false lightning signals and outputting the real lightning signal.

[0016] This specification also provides a step-by-step lightning spurious signal filtering system based on principal component analysis. This system is applied to lightning observation data from geostationary meteorological satellites and includes:

[0017] The acquisition module acquires the transient optical signal to be processed.

[0018] The transient optical signal preprocessing module performs pseudo-continuous false signal filtering at the transient optical signal level based on the aspect ratio and energy variation coefficient of the transient optical signal within the sliding time window of the transient optical signal to be processed, thereby obtaining the preprocessed transient optical signal.

[0019] The transient optical signal group generation module determines the clustering core based on the spatial index grid of the transient optical signal determined by the preprocessed transient optical signal, constructs a first sliding window, and clusters the transient optical signals that meet the first preset conditions within the first sliding window to generate transient optical signal groups.

[0020] The transient optical signal group preprocessing module performs pseudo-continuous false signal filtering at the transient optical signal group level based on the aspect ratio of the transient optical signal group, the first two principal components of the transient optical signal group principal component analysis, and the energy variation coefficient of the transient optical signal group, to obtain the preprocessed transient optical signal group.

[0021] The lightning generation module constructs a second sliding window using the spatial index grid of the transient light signal group determined by the preprocessed transient light signal group. It then performs spatiotemporal clustering analysis on the transient light signal groups that meet the second preset conditions within the second sliding window to obtain the lightning corresponding to the transient light signal to be processed.

[0022] The lightning false signal filtering module sequentially performs lightning-level pseudo-continuous false signal filtering and lightning discrete false signal filtering on the lightning corresponding to the transient light signal to be processed.

[0023] The output module repeatedly executes the steps of the transient light signal preprocessing module - lightning false signal filtering module on the transient light signal to be processed until the number of transient light signals before and after filtering is the same, thus completing the lightning false signal filtering and outputting the real lightning signal.

[0024] The principal component analysis-based method and system for progressively filtering lightning spurious signals provided in this specification not only achieves efficient filtering of spurious signals from geostationary meteorological satellites but also effectively overcomes the limitation of traditional algorithms that are prone to mistakenly deleting real signals, significantly improving the accuracy of lightning detection and the signal-to-noise ratio. This improvement optimizes the operational availability of satellite lightning products and lays a data foundation for accurate monitoring and early warning of severe convective weather. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 A schematic diagram of the system architecture for a step-by-step filtering method for lightning spurious signals based on principal component analysis, provided in the embodiments of this specification;

[0027] Figure 2 A flowchart illustrating a step-by-step filtering method for spurious lightning signals based on principal component analysis, provided as an embodiment of this specification;

[0028] Figure 3 A schematic diagram illustrating an embodiment of a stepwise filtering method for spurious lightning signals based on principal component analysis provided in this specification.

[0029] Figure 4 The image shows a 10.8µm infrared channel cloud image overlaid with the LMI lightning localization results from the FY-4A geostationary meteorological satellite lightning imager, without false signal filtering. (The red asterisks in the image represent the distribution of "lightning" obtained without false signal filtering; the colors in the cloud image, from light to dark, represent the cloud top brightness temperature (TBB) in K; the black trapezoidal boxes indicate the field of view of the lightning imager.)

[0030] Figure 5 The image shows a 10.8µm infrared channel cloud image observed by an imager overlaying a superimposed image of lightning distribution detected by the World Wide LightningLocation Network (WWLLN) within the field of view (red asterisks in the image represent lightning distribution detected by the WWLLN).

[0031] Figure 6 The image shows a 10.8µm infrared channel cloud image observed by the "flashes distribution overlay imager" in the official product (the yellow pentagram in the image represents the "flashes" distribution).

[0032] Figure 7 The distribution of lightning calculated by the method of this application is superimposed on the 10.8µm infrared channel cloud image observed by the imager (the yellow pentagram in the figure represents the distribution of "lightning");

[0033] Figure 8 This is a schematic diagram of a lightning spurious signal filtering system based on principal component analysis, provided as an embodiment of this specification.

[0034] Explanation of reference numerals in the attached figures:

[0035] The system architecture includes 100, terminal device 101, network 102, server 103, acquisition module 801, transient optical signal preprocessing module 803, transient optical signal group generation module 805, transient optical signal group preprocessing module 807, lightning generation module 809, lightning false signal filtering module 811, and output module 813. Detailed Implementation

[0036] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0037] Lightning imagers on geostationary orbit (GEO) meteorological satellites offer significant advantages in enabling large-scale, continuous lightning monitoring. This is of great scientific importance and practical application for deeply understanding the patterns of lightning activity and improving the monitoring and early warning capabilities for severe convective weather. However, due to the influence of photoelectric detection devices and the outer space environment, raw satellite lightning observation data often contains a large number of non-lightning signals. The presence of these false signals not only severely restricts the efficiency of satellite lightning detection but also greatly reduces the effectiveness of lightning data in monitoring and early warning of severe convective weather.

[0038] Current operational geostationary meteorological satellite lightning location primarily employs various spurious signal filtering algorithms targeting specific noise sources. These methods are typically designed based on simulated single-source noise, making it difficult to handle complex situations involving the superposition of spurious signals from multiple different sources. To improve the spurious signal filtering performance in real satellite lightning observation data, this specification provides a step-by-step lightning spurious signal filtering method and system based on principal component analysis.

[0039] Figure 1 This diagram illustrates the system architecture of a step-by-step filtering method for spurious lightning signals based on principal component analysis, as provided in the embodiments of this specification. Figure 1 As shown, system architecture 100 may include at least one terminal device 101, a network 102, and a server 103. Network 102 serves as the medium for providing a communication link between terminal device 101 and server 103. Network 102 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0040] Terminal device 101 interacts with server 103 via network 102 to receive or send messages, etc. Various client applications can be installed on terminal device 101, such as dedicated programs like a step-by-step filtering method for lightning spurious signals based on principal component analysis.

[0041] Terminal device 101 can be hardware or software. When terminal device 101 is hardware, it can be various dedicated or general-purpose electronic devices, including but not limited to smartphones, tablets, laptops, and desktop computers. When terminal device 101 is software, it can be installed in the electronic devices listed above. It can be implemented as multiple software programs or software modules (e.g., multiple software programs or software modules used to provide distributed services) or as a single software program or software module.

[0042] Server 103 can be a server that provides various services, such as a backend server that provides services to client applications installed on terminal device 101. For example, the server can perform stepwise filtering of lightning spoofing signals based on principal component analysis so that the results of stepwise filtering of lightning spoofing signals based on principal component analysis can be displayed on terminal device 101.

[0043] Server 103 can be either hardware or software. When server 103 is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When server 103 is software, it can be implemented as multiple software programs or software modules (e.g., multiple software programs or software modules used to provide distributed services), or as a single software program or software module.

[0044] Figure 2 This is a flowchart illustrating a step-by-step filtering method for spurious lightning signals based on principal component analysis, provided in an embodiment of this specification. From a programming perspective, the execution entity of the process can be a program hosted on an application server or application terminal. It is understood that this method can be executed by any device, equipment, platform, or cluster of devices with computing and processing capabilities. Figure 2 As shown, the step-by-step filtering method for false lightning signals includes:

[0045] Step S201: Acquire the transient optical signal to be processed.

[0046] In the embodiments of this specification, the transient light signal to be processed is lightning observation data from a geostationary meteorological satellite, and the transient light signal to be processed includes: latitude and longitude data, energy, and observation time of the transient light signal.

[0047] Step S203: Based on the aspect ratio and energy variation coefficient of the transient optical signal within the sliding time window of the transient optical signal to be processed, perform pseudo-continuous false signal filtering at the transient optical signal level to obtain the preprocessed transient optical signal.

[0048] In the embodiments of this specification, the step of filtering out pseudo-continuities and spurious signals at the transient light signal level based on the aspect ratio and energy variation coefficient of the transient light signal within a sliding time window of the transient light signal to be processed, to obtain a preprocessed transient light signal, specifically includes:

[0049] The transient optical signals to be processed are sorted according to their occurrence time to obtain a transient optical signal time series;

[0050] On the transient optical signal time series, several sliding time windows are obtained with a preset time as the sliding time window;

[0051] Principal component analysis is performed on the latitude and longitude of the transient light signal within any sliding time window as a feature variable to obtain the aspect ratio and energy variation coefficient of the transient light signal within any sliding time window.

[0052] If the aspect ratio of the transient optical signal within any sliding time window is greater than the first preset threshold, and the energy variation coefficient of the transient optical signal within any sliding time window is less than the first coefficient threshold, then all transient optical signals within any sliding time window are pseudo-continuous signals, and pseudo-continuous false signals are filtered out at the transient optical signal level.

[0053] The transient light signal is obtained by sequentially filtering out pseudo-continuous false signals at the transient light signal level within the several sliding time windows.

[0054] In the embodiments described in this specification, the preset time is 10ms.

[0055] In the embodiments described in this specification, the aspect ratio of the transient optical signal within any sliding time window is: Formula (1);

[0056] in,

[0057] This represents the aspect ratio of the transient optical signal within any of the sliding time windows;

[0058] This represents the principal axis within the arbitrary sliding time window;

[0059] This represents the secondary axis within the arbitrary sliding time window;

[0060] The coefficient of variation of the energy of the transient optical signal within any sliding time window is the ratio of the standard deviation of the energy of all transient optical signals within the arbitrary sliding time window to the mean energy of all transient optical signals within the arbitrary sliding time window.

[0061] Formula (2);

[0062] Formula (3);

[0063] Formula (4);

[0064] Formula (5);

[0065] Formula (6);

[0066] in, , , , These represent the latitude and longitude, longitude and longitude, latitude and longitude, and longitude and latitude variance of the transient optical signal within the sliding time window, respectively. The number of samples within the sliding time window; The covariance matrix is ​​a two-dimensional matrix composed of the latitude and longitude variances of the transient optical signal. as well as These represent the trace and determinant of the matrix, respectively.

[0067] In this illustrative embodiment, the expression for the energy variation coefficient of the transient optical signal is:

[0068] Formula (7);

[0069] in, This represents the energy variation coefficient of a transient optical signal; This represents the energy of the i-th transient light signal within the window; The standard deviation of the transient optical signal energy within the window; This represents the average energy of the transient optical signal within the sliding time window.

[0070] In the embodiments described in this specification, the first preset threshold is 15 and the first preset coefficient is 0.25.

[0071] Step S205: After determining the clustering core based on the spatial index grid of the transient light signals determined by the preprocessed transient light signals, a first sliding window is constructed, and the transient light signals that meet the first preset conditions within the first sliding window are clustered to generate a transient light signal group.

[0072] In this embodiment of the specification, after determining the clustering core based on the spatial index grid of the transient optical signals determined by the preprocessed transient optical signals, a first sliding window is constructed, and transient optical signals that meet the first preset conditions within the first sliding window are clustered to generate transient optical signal groups, specifically including:

[0073] Using the maximum latitude and longitude and the minimum latitude and longitude of the preprocessed transient light signal as the boundaries of the spatial index grid, and using the first preset distance as the grid spacing, the spatial index grid of the transient light signal is constructed.

[0074] Using each grid point in the spatial index grid of the preprocessed transient optical signal as a cluster core, and based on each cluster core, using a first preset window as the first sliding window, the transient optical signals that meet the first preset conditions within the sliding window are clustered into a transient optical signal group, thereby generating the transient optical signal group.

[0075] in,

[0076] The first preset condition is that the time distance between transient optical signals is 0ms, the spatial distance between the transient optical signals is less than the first preset distance, and the root node index numbers corresponding to the transient optical signals are different.

[0077] In this embodiment, the first preset distance is 16.5km.

[0078] The spatial index grid of transient optical signals obtained through this step includes the number of transient optical signals contained in each grid point.

[0079] In the embodiments of this specification, the formula for calculating the spatial distance between the transient optical signals is as follows:

[0080] Formula (8);

[0081] in,

[0082] This indicates the spatial distance between the transient optical signals;

[0083] Represents the radius of the Earth;

[0084] Indicates latitude;

[0085] Indicates longitude;

[0086] Represents the latitude of the j-th transient optical signal;

[0087] Represents the latitude of the i-th transient optical signal;

[0088] = - This represents the difference in latitude between the j-th transient optical signal and the ith transient optical signal;

[0089] Represents the longitude of the j-th transient optical signal;

[0090] Represents the longitude of the i-th transient optical signal;

[0091] = - This represents the difference in longitude between the j-th transient optical signal and the ith transient optical signal.

[0092] In the embodiments described in this specification, , The first preset window is a 3x3 grid.

[0093] Using the grid points of the constructed transient light signal spatial index grid as the center, and a 3×3 grid as the window, transient light signals that occur simultaneously (with a time distance of 0 ms), have a spatial distance of less than 16.5 km, and have different root node index numbers are clustered into a transient light signal "group", representing a ground flash return stroke or cloud flash backstream.

[0094] Step S207: Based on the aspect ratio of the transient optical signal group, the first two principal components of the transient optical signal group principal component analysis, and the energy variation coefficient of the transient optical signal group, perform pseudo-continuous false signal filtering at the transient optical signal group level to obtain the preprocessed transient optical signal group.

[0095] In the embodiments of this specification, the step of filtering out pseudo-continuities and spurious signals at the transient optical signal group level based on the aspect ratio of the transient optical signal group, the first two principal components of the transient optical signal group principal component analysis, and the energy variation coefficient of the transient optical signal group to obtain a preprocessed transient optical signal group specifically includes:

[0096] Principal component analysis is performed on transient light signal groups containing more than 5 transient light signals. The latitude and longitude of the transient light signals in the transient light signal group are used as feature variables to calculate the covariance matrix and obtain the principal axis and secondary axis of the transient light signal group. The principal axis and secondary axis of the transient light signal group are the first two principal components of the principal component analysis of the transient light signal group.

[0097] The aspect ratio of the transient optical signal group is determined based on the principal axis and the secondary axis of the transient optical signal group.

[0098] If the aspect ratio of the transient optical signal group is greater than the second preset threshold, and the principal axis of the transient optical signal group is greater than the principal axis threshold, the secondary axis of the transient optical signal group is less than the primary axis threshold, and the energy variation coefficient of the transient optical signal group is less than the second coefficient threshold, then the transient optical signal group is a pseudo-continuous signal, and pseudo-continuous false signal filtering is performed at the transient optical signal group level.

[0099] Each transient optical signal group in the transient optical signal group is sequentially subjected to transient optical signal group-level pseudo-continuity spurious signal filtering to obtain the preprocessed transient optical signal group.

[0100] In the embodiments of this specification, the transient optical signal group is calculated with reference to formulas (1)-(7), and the energy variation coefficient of the transient optical signal group is calculated using formula (7). In a specific embodiment, the principal axis threshold is 15, the first axis threshold is 5, the second preset threshold is 10, and the second coefficient threshold is 0.3.

[0101] Step S209: Construct a second sliding window using the spatial index grid of the transient light signal group determined by the preprocessed transient light signal group, and perform spatiotemporal clustering analysis on the transient light signal groups that meet the second preset conditions within the second sliding window to obtain the lightning corresponding to the transient light signal to be processed.

[0102] In this embodiment of the specification, the step of constructing a second sliding window using the spatial index grid of the transient light signal group determined by the preprocessed transient light signal group, and performing spatiotemporal clustering analysis on the transient light signal groups within the second sliding window that meet the second preset conditions to obtain the lightning corresponding to the transient light signal to be processed, specifically includes:

[0103] Centered on each grid point of the spatial index grid of the preprocessed transient optical signal group, a second preset grid is used as the second sliding window, and there are several second sliding windows;

[0104] Calculate the time interval between any two groups of signals within the second sliding window, and filter the signal groups that meet the preset time conditions;

[0105] The minimum spherical distance between any two pixels within a signal group that meets the preset time condition is taken as the spatial distance between the two signal groups.

[0106] Based on the spatial distance and temporal distance between the two sets of signals, calculate the time-space weighted Euclidean distance;

[0107] The transient light signal groups whose time-space weighted Euclidean distance satisfies the second preset condition are grouped into a single lightning bolt.

[0108] The transient light signal group within the second sliding window is traversed to obtain the lightning corresponding to the transient light signal to be processed.

[0109] In the embodiments of this specification, the spatial index grid of the transient optical signal group determined by the preprocessed transient optical signal group is constructed using the following method: the maximum and minimum latitude and longitude of the preprocessed transient optical signal group are used as the boundaries of the spatial index grid, and a second preset distance is used as the grid spacing.

[0110] In the embodiments of this specification, the latitude and longitude of the preprocessed transient optical signal group are weighted by the ratio of the light radiation energy of a single transient optical signal in the preprocessed transient optical signal group to the total light radiation energy of the preprocessed transient optical signal group. The latitude and longitude of a single transient optical signal are calculated by weighted averaging of the latitude and longitude of the transient optical signals in the preprocessed transient optical signal group. The set of latitude and longitude of the single transient optical signals constitutes the latitude and longitude of the preprocessed transient optical signal group.

[0111] In the embodiments described in this specification, the second preset distance is twice the first preset distance. In one embodiment, the second preset distance is 33 km.

[0112] In the embodiments of this specification, the spatial distance between the two sets of signals is the minimum value of the spherical distance between the two sets of signals;

[0113] The time distance between the two sets of signals is the absolute value of the difference between the signal occurrence times of the two sets of signals;

[0114] The formula for calculating the time-space weighted Euclidean distance is as follows:

[0115] Formula (9);

[0116] in,

[0117] This represents the time-space weighted Euclidean distance between two transient optical signal groups;

[0118] Indicates time weight;

[0119] Indicates distance weight;

[0120] This represents the time combination of transient optical signals in transient optical signal group A;

[0121] This represents the transient optical signal time combination of transient optical signal group B;

[0122] A represents transient optical signal group A;

[0123] B represents transient optical signal group B;

[0124] This represents the minimum time distance between the transient optical signals contained in transient optical signal group A and transient optical signal group B;

[0125] Formula (10);

[0126] This indicates the time corresponding to transient optical signal group A;

[0127] This indicates the time corresponding to transient optical signal group B;

[0128] This represents the minimum spherical distance between any two signals in transient optical signal group A and transient optical signal group B;

[0129] In the embodiments described in this specification,

[0130] Formula (11);

[0131] 'a' represents the transient optical signal in transient optical signal group A;

[0132] b represents the transient optical signal in transient optical signal group B;

[0133] haversine represents the semi-sine algorithm;

[0134] The second preset condition is: time-space weighted Euclidean distance ≤ 1.0.

[0135] In the embodiments described in this specification, the second preset grid is a 3×3 window; the preset time condition is 330ms; .

[0136] Continuing from the previous example, using the grid points of the transient light signal group spatial index grid as the center, the transient light signal groups within a 3×3 window are traversed. First, the time interval between any two signal groups is calculated, and combinations less than 330ms are selected. For signal groups that meet the time condition, the pairwise spherical distances between their internal pixels are calculated, and the minimum value is taken as the spatial distance between the two signal groups.

[0137] Step S211: Perform lightning-level pseudo-continuous false signal filtering and lightning discrete false signal filtering sequentially on the lightning corresponding to the transient light signal to be processed.

[0138] In the embodiments of this specification, the step of sequentially filtering out pseudo-continuous spurious signals and discrete spurious signals of lightning corresponding to the transient optical signal to be processed specifically includes:

[0139] Principal component analysis is performed on lightning that includes two or more transient light signals corresponding to the transient light signals to be processed, to obtain the aspect ratio, secondary axis, energy variation coefficient, and time variation coefficient of the lightning.

[0140] Based on the aspect ratio, secondary axis, energy variation coefficient, and time variation coefficient of the lightning, false signals are identified, and pseudo-continuous false signals of the lightning level are filtered out to obtain lightning with the false signals filtered out.

[0141] Discrete random noise in the lightning that is used to filter out false signals is identified, and discrete false signals of lightning are filtered out.

[0142] The aspect ratio of lightning, the secondary axis of lightning, and the coefficient of variation of lightning energy are all calculated using formulas (1) to (7).

[0143] Considering the uniformity of noise time intervals and small time variance, a time uniformity test is added to the filtering of lightning pseudo-continuous spurious signals. Time uniformity is assessed using the lightning time variation coefficient (CV). t The time variation coefficient (CV) of lightning is used to characterize this. t The calculation of ) is performed using the following formula:

[0144] Formula (12);

[0145] in, This represents the time offset of the i-th transient light signal contained in the lightning relative to the start time. This represents the standard deviation of the time offset of the transient light signal contained in lightning.

[0146] In the embodiments of this specification, the step of determining false signals based on the aspect ratio of the lightning, the secondary axis of the lightning, the energy variation coefficient of the lightning, and the time variation coefficient of the lightning, performing lightning-level pseudo-continuous false signal filtering, and obtaining lightning with filtered false signals specifically includes:

[0147] If the aspect ratio of the lightning is greater than the third preset threshold, the secondary axis of the lightning is less than the second axis threshold, the energy variation coefficient of the lightning is less than the third coefficient threshold, and the time variation coefficient of the lightning is less than the fourth coefficient threshold, then the lightning is a false signal and a false signal filtering of the lightning level is performed.

[0148] For each lightning bolt in the lightning corresponding to the transient light signal to be processed, lightning-level pseudo-continuous spurious signal filtering is performed sequentially to obtain the lightning with the spurious signal filtered out.

[0149] The process of determining discrete random noise in the lightning strikes from which false signals are filtered, and performing discrete false signal filtering for lightning strikes, specifically includes:

[0150] If the duration of a lightning strike in the lightning strikes for which false signals are filtered out is less than the observation time of a single frame, and it contains only one transient light signal group, then the lightning strike is discrete random noise, and discrete false signals of lightning strikes are filtered out.

[0151] The discrete spurious signal filtering operation is performed sequentially on each lightning bolt in the lightning bolts for which spurious signals are filtered out.

[0152] In the embodiments described in this specification, the third preset threshold is 10, the second axis threshold is 5, the third coefficient threshold is 0.2, and the fourth coefficient threshold is 0.3.

[0153] The duration of lightning is the time distance between the earliest and latest transient light signals that occur during lightning.

[0154] Step S213: Repeat steps S203-S211 above for the transient light signal to be processed until the number of transient light signals before and after filtering is the same, thus completing the filtering of false lightning signals and outputting the real lightning signal.

[0155] To ensure that the output real lightning signal can be used for lightning location, the process of constructing a second sliding window using the spatial index grid of the preprocessed transient light signal group, performing spatiotemporal clustering analysis on the transient light signal groups within the second sliding window that meet the second preset conditions, and obtaining the lightning corresponding to the transient light signal to be processed, further includes:

[0156] The latitude and longitude of the lightning corresponding to the transient light signal to be processed, the spatial lightning density of the lightning corresponding to the transient light signal to be processed, and the spatiotemporal lightning density of the lightning corresponding to the transient light signal to be processed are determined, and the lightning corresponding to the transient light signal to be processed is located.

[0157] In the embodiments of this specification, the latitude and longitude of the lightning corresponding to the transient light signal to be processed are the weighted average of the latitude and longitude of the transient light signal group contained in the lightning;

[0158] The spatial lightning density of the lightning corresponding to the transient light signal to be processed is the number of times the lightning corresponding to the transient light signal to be processed occurs per unit area;

[0159] The temporal lightning density corresponding to the transient light signal to be processed is the number of times the lightning corresponding to the data to be processed occurs per unit time and per unit area.

[0160] After obtaining lightning location information, the number of lightning strikes per unit area is counted based on the spatial grid size, i.e., spatial lightning density. Then, the number of lightning strikes per unit time and per unit area is counted based on the spatial grid size and time span, i.e., spatiotemporal lightning density. Lightning density can be used as an indicator to measure the intensity of lightning activity in a given area.

[0161] To further understand the stepwise filtering method for lightning spurious signals based on principal component analysis provided in the embodiments of this specification, the following will describe it in conjunction with specific embodiments. Figure 3 This is a schematic diagram of an embodiment of a method for filtering out false lightning signals based on principal component analysis, provided in this specification. In this embodiment, the transient light signal, transient light signal group, and lightning are named as follows: event represents transient light signal, group represents transient light signal group, and flash represents lightning.

[0162] Step S301: Read the L1 data and positioning information from the lightning imager of the geostationary meteorological satellite.

[0163] The data is in HDF format and contains lightning signals spanning one minute, with a spatial resolution of 7.8 km. The nadir point is located at (0°N, 104.7°E). Observations are conducted using a semi-disk method, covering China and its surrounding waters from March 23rd to September 23rd annually, and shifting to Australia and its surrounding waters from September 23rd to March 23rd of the following year. The L1 level data of FY-4A / LMI contains the latitude and longitude data, energy (event radiance), and observation start time information of the transient light signals (events) after on-board filtering, required by the algorithm. Based on the frame number read from the one-minute data, the time offset (time_offsets, in ms) of the transient light signal (event) relative to the observation start time in each frame is calculated at a time resolution of 2 ms, and this is used as the event's time information.

[0164] Step S303: Filter out false signals of event continuity.

[0165] All events within the field of view are sorted by their occurrence time. Then, on the event time series, with a sliding time window of 10ms, PCA (Principal Component Analysis) is performed using the latitude and longitude of the events within the time window as feature variables. The aspect ratio of the events within the window is obtained.

[0166] Step S305: Construct the event spatial index grid.

[0167] The maximum and minimum latitude and longitude of all events within the field of view are statistically analyzed. Using these maximum and minimum latitude and longitude as the boundaries of the spatial index grid, and with a grid spacing of 16.5 km, an event spatial index grid is constructed. The number of events contained in each grid point is then counted.

[0168] Step S307: Events spatial clustering analysis.

[0169] Centered on each grid point of the event spatial index grid, and using a 3×3 grid as a window, events that occur simultaneously (with a time distance of 0 ms), have a spatial distance of less than 16.5 km, and have different root node index numbers are clustered into a group. The spatial distance is calculated using the Haversine distance method.

[0170] Step S309: Locating Groups.

[0171] Based on the groups obtained from spatial clustering analysis, the latitude and longitude of a group are calculated by weighting the latitude and longitude of the events within the group, using the proportion of the light radiation energy (in joules) of a single event in the group relative to the total light radiation energy of the group. The calculated range of latitude and longitude values ​​for the groups. , .

[0172] Step S311: Groups pseudo-continuous spurious signal filtering.

[0173] Principal component analysis (PCA) is performed on groups containing more than five events. Using the latitude and longitude of the events within the group as characteristic variables, the covariance matrix is ​​calculated to obtain the first two principal components. The aspect ratio of the group is then calculated based on these two principal components, followed by an energy uniformity test, and the energy variation coefficient (CV) of the group is calculated. e When the aspect ratio of the group More than 10, spindle More than 15, secondary axis When it is less than 5, and the coefficient of variation of energy CV e If the value is less than 0.3, the group and all events it contains are considered false signals and are filtered out.

[0174] Step S313: Construct a spatial index grid for groups.

[0175] The maximum and minimum latitude and longitude of all groups within the field of view are statistically analyzed. Using the maximum and minimum latitude and longitude as the boundaries of the spatial index grid, and with a grid spacing of 33 km, a spatial index grid for the groups is constructed. The number of groups contained in each grid point is then counted.

[0176] Step S315: Spatiotemporal clustering analysis of Groups.

[0177] Centered on the spatial index grid points generated in step S311, the groups within the 3×3 window are traversed. First, the time interval between any two sets of signals is calculated, and combinations less than 330ms are selected. For groups that meet the time condition, the pairwise spherical distances between pixels within the group are calculated, and the minimum value is taken as the spatial distance between the two groups. Finally, the time-space weighted Euclidean distance (ED) is calculated. If ED ≤ 1.0, the two groups are merged into one flash.

[0178] Step S317: Location of Flashes.

[0179] Based on the flash obtained from spatiotemporal clustering analysis, the ratio of the light radiation energy of a single group within the flash to the total light radiation energy of the flash is used as the weight, and the latitude and longitude of the groups contained in the flash are analyzed. By performing a weighted average, the latitude and longitude of a flash can be calculated. .in Indicates latitude, Indicates longitude.

[0180] Step S319: Filter out false and continuous flashes.

[0181] Based on step S317, PCA principal component analysis is performed on flashes containing two or more events. Using the latitude and longitude of the events within the flash as characteristic variables, the covariance matrix is ​​calculated to obtain the first two principal components. The aspect ratio of the flash is calculated based on the first two principal components, and an energy uniformity test is performed. The energy variation coefficient (CV) of the flash is then calculated. e Considering the uniformity of noise time intervals and the small time variance, a time uniformity test is added to the filtering of spurious signals from flash memory. Time uniformity is expressed as the flash time coefficient of variation (CV). t ) is used to characterize it.

[0182] Finally, based on the aspect ratio, energy variation coefficient, and time variation coefficient, the aspect ratio is... More than 10, secondary axis Less than 5, and the coefficient of variation over time (CV) t and energy variation coefficient CV e Flash signals with values ​​less than 0.2 and 0.3, respectively, along with all events they contain, are identified as spurious signals and filtered out.

[0183] Step S321: Flashes discrete spurious signal filtering.

[0184] The duration of all flashes within the field of view after filtering out spurious signals in step S319 is recorded. If the duration of a flash is less than 2ms and contains only one group, it indicates that the flash is discrete random noise and is therefore filtered out.

[0185] Step S323: Repeat steps S301-S321 until the number of transient light signals before and after filtering is the same, output the actual lightning signal location and calculate the lightning spatial density.

[0186] Count the number of events within the field of view after filtering out spurious signals in step S321. ), and the number of events before false signals were filtered out ( Compare them. If Then repeat steps S301-S321. Once If no false signal is detected, the loop ends. Output the flash position generated in step S317, and based on this, take 0.1°×0.1° as the unit grid, count the number of flashes in the grid, and obtain the spatial density of the flash at the current time.

[0187] To verify the step-by-step filtering method for spurious lightning signals based on principal component analysis provided in the embodiments of this specification, the lightning prediction results at the same time were further verified, such as... Figure 4-7 As shown. From Figure 4-7 As can be seen, the principal component analysis-based stepwise lightning false signal filtering method provided in this application can effectively remove 85.6% of the false signals. After filtering out the false signals, more than 94% of the lightning events are concentrated in cloud regions with cloud top temperatures below 0°C, and these clouds subsequently develop into severe thunderstorms. This result indicates that effective lightning signals can keenly capture the initial development process of thunderstorm clouds, thus providing crucial information for early warning of severe convective weather.

[0188] Furthermore, compared with global ground-based lightning observations (WWLLN), using a matching radius of 50 km and a time threshold of 1 minute, the average matching rate between satellite lightning and ground-based lightning obtained by the principal component analysis-based method for progressively filtering out false lightning signals was 64.2%. In contrast, using the same matching strategy, the matching rate between official products and ground-based lightning was only 24.43%. Compared to official products, the matching rate between satellite lightning and ground-based lightning obtained by the principal component analysis-based method for progressively filtering out false lightning signals improved by nearly 40%.

[0189] The principal component analysis-based stepwise filtering method for lightning spurious signals provided in this specification not only achieves efficient filtering of spurious signals from geostationary meteorological satellites, but also effectively overcomes the limitation of traditional algorithms that are prone to mistakenly deleting real signals, significantly improving the accuracy of lightning detection and the signal-to-noise ratio. This improvement optimizes the operational availability of satellite lightning products and lays a data foundation for accurate monitoring and early warning of severe convective weather.

[0190] The above describes in detail a step-by-step filtering method for lightning spurious signals based on principal component analysis. Correspondingly, this specification also provides a step-by-step filtering system for lightning spurious signals based on principal component analysis, such as... Figure 8 As shown. Figure 8 This specification provides a schematic diagram of a lightning spurious signal filtering system based on principal component analysis, which is applied to lightning observation data from geostationary meteorological satellites and includes:

[0191] Acquisition module 801 acquires the transient optical signal to be processed;

[0192] The transient optical signal preprocessing module 803 performs pseudo-continuous false signal filtering at the transient optical signal level based on the aspect ratio and energy variation coefficient of the transient optical signal within the sliding time window of the transient optical signal to be processed, thereby obtaining the preprocessed transient optical signal.

[0193] The transient optical signal group generation module 805 determines the clustering core based on the spatial index grid of the transient optical signal determined by the preprocessed transient optical signal, constructs a first sliding window, and clusters the transient optical signals that meet the first preset conditions within the first sliding window to generate a transient optical signal group.

[0194] The transient optical signal group preprocessing module 807 performs pseudo-continuous false signal filtering at the transient optical signal group level based on the aspect ratio of the transient optical signal group, the first two principal components of the transient optical signal group principal component analysis, and the energy variation coefficient of the transient optical signal group, to obtain the preprocessed transient optical signal group.

[0195] The lightning generation module 809 constructs a second sliding window using the spatial index grid of the transient light signal group determined by the preprocessed transient light signal group, and performs spatiotemporal clustering analysis on the transient light signal groups that meet the second preset conditions within the second sliding window to obtain the lightning corresponding to the transient light signal to be processed.

[0196] The lightning false signal filtering module 811 sequentially performs lightning-level pseudo-continuous false signal filtering and lightning discrete false signal filtering on the lightning corresponding to the transient light signal to be processed.

[0197] The output module 813 repeatedly executes the above-mentioned transient light signal preprocessing module - lightning false signal filtering module steps on the transient light signal to be processed until the number of transient light signals before and after filtering is the same, thus completing the lightning false signal filtering and outputting the real lightning signal.

[0198] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0199] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments for apparatus, electronic devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0200] The apparatus, electronic device, and non-volatile computer storage medium and method provided in the embodiments of this specification are corresponding. Therefore, the apparatus, electronic device, and non-volatile computer storage medium also have similar beneficial technical effects as the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the corresponding apparatus, electronic device, and non-volatile computer storage medium will not be repeated here.

[0201] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must also be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0202] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0203] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0204] For ease of description, the above apparatus is described by dividing it into various functional units. Of course, when implementing one or more embodiments of this specification, the functions of each unit can be implemented in one or more software and / or hardware.

[0205] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product 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.

[0206] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will 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 apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0207] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function 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.

[0208] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable 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.

[0209] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0210] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0211] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0212] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus 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 apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0213] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside on local and remote computer storage media, including storage devices.

[0214] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0215] The above description is merely an embodiment of this specification and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the scope of the claims of this application.

Claims

1. A method for stepwise filtering of spurious lightning signals based on principal component analysis, characterized in that, The step-by-step filtering method for false lightning signals is applied to lightning observation data from geostationary meteorological satellites. The step-by-step filtering method for false lightning signals includes: S1: Acquire the transient optical signal to be processed; S2: Based on the aspect ratio and energy variation coefficient of the transient light signal within the sliding time window of the transient light signal to be processed, perform pseudo-continuous false signal filtering at the transient light signal level to obtain the preprocessed transient light signal. S3: After determining the clustering core based on the spatial index grid of the transient light signal determined by the preprocessed transient light signal, a first sliding window is constructed, and the transient light signals that meet the first preset conditions within the first sliding window are clustered to generate a transient light signal group. S4: Based on the aspect ratio of the transient optical signal group, the first two principal components of the transient optical signal group principal component analysis, and the energy variation coefficient of the transient optical signal group, pseudo-continuous false signals are filtered out at the transient optical signal group level to obtain the preprocessed transient optical signal group. S5: Using the spatial index grid of the transient light signal group determined by the preprocessed transient light signal group, construct a second sliding window, and perform spatiotemporal clustering analysis on the transient light signal groups that meet the second preset conditions within the second sliding window to obtain the lightning corresponding to the transient light signal to be processed; S6: Perform lightning-level pseudo-continuous false signal filtering and lightning discrete false signal filtering sequentially on the lightning corresponding to the transient light signal to be processed; Repeat steps S2-S6 for the transient light signal to be processed until the number of transient light signals before and after filtering is the same, thus completing the filtering of false lightning signals and outputting the real lightning signal.

2. The method for step-by-step filtering of false lightning signals as described in claim 1, characterized in that, The process involves filtering out pseudo-continuities and spurious signals at the transient light signal level based on the aspect ratio and energy variation coefficient of the transient light signal within a sliding time window to obtain a preprocessed transient light signal. Specifically, this includes: The transient optical signals to be processed are sorted according to their occurrence time to obtain a transient optical signal time series; On the transient optical signal time series, several sliding time windows are obtained with a preset time as the sliding time window; Principal component analysis is performed on the latitude and longitude of the transient light signal within any sliding time window as a feature variable to obtain the aspect ratio and energy variation coefficient of the transient light signal within any sliding time window. If the aspect ratio of the transient optical signal within any sliding time window is greater than the first preset threshold, and the energy variation coefficient of the transient optical signal within any sliding time window is less than the first coefficient threshold, then all transient optical signals within any sliding time window are pseudo-continuous signals, and pseudo-continuous false signals are filtered out at the transient optical signal level. The transient light signal is obtained by sequentially filtering out pseudo-continuous false signals at the transient light signal level within the several sliding time windows.

3. The method for step-by-step filtering of false lightning signals as described in claim 2, characterized in that, The aspect ratio of the transient optical signal within any sliding time window is: ; in, This represents the aspect ratio of the transient optical signal within any of the sliding time windows; Represents the main axis within any sliding time window; Represents the secondary axis within any sliding time window; The coefficient of variation of the energy of the transient optical signal within any sliding time window is the ratio of the standard deviation of the energy of all transient optical signals within the arbitrary sliding time window to the mean energy of all transient optical signals within the arbitrary sliding time window.

4. The method for step-by-step filtering of false lightning signals as described in claim 1, characterized in that, After determining the clustering core based on the spatial index grid of the transient optical signals determined by the preprocessed transient optical signals, a first sliding window is constructed. The transient optical signals within the first sliding window that meet the first preset condition are clustered to generate transient optical signal groups, specifically including: Using the maximum latitude and longitude and the minimum latitude and longitude of the preprocessed transient light signal as the boundaries of the spatial index grid, and using the first preset distance as the grid spacing, the spatial index grid of the transient light signal is constructed. Using each grid point in the spatial index grid of the preprocessed transient optical signal as a cluster core, and based on each cluster core, using a first preset window as the first sliding window, the transient optical signals that meet the first preset conditions within the sliding window are clustered into a transient optical signal group, thereby generating the transient optical signal group. in, The first preset condition is that the time distance between transient optical signals is 0ms, the spatial distance between the transient optical signals is less than the first preset distance, and the root node index numbers corresponding to the transient optical signals are different.

5. The method for step-by-step filtering of false lightning signals as described in claim 1, characterized in that, The process involves filtering out pseudo-continuities and spurious signals at the transient optical signal group level based on the aspect ratio, the first two principal components of the transient optical signal group principal component analysis, and the energy variation coefficient of the transient optical signal group, to obtain a preprocessed transient optical signal group. Specifically, this includes: Principal component analysis is performed on transient light signal groups containing more than 5 transient light signals. The latitude and longitude of the transient light signals in the transient light signal group are used as feature variables to calculate the covariance matrix and obtain the principal axis and secondary axis of the transient light signal group. The principal axis and secondary axis of the transient light signal group are the first two principal components of the principal component analysis of the transient light signal group. The aspect ratio of the transient optical signal group is determined based on the principal axis and the secondary axis of the transient optical signal group. If the aspect ratio of the transient optical signal group is greater than the second preset threshold, and the principal axis of the transient optical signal group is greater than the principal axis threshold, the secondary axis of the transient optical signal group is less than the primary axis threshold, and the energy variation coefficient of the transient optical signal group is less than the second coefficient threshold, then the transient optical signal group is a pseudo-continuous signal, and pseudo-continuous false signal filtering is performed at the transient optical signal group level. Each transient optical signal group in the transient optical signal group is sequentially subjected to transient optical signal group-level pseudo-continuity spurious signal filtering to obtain the preprocessed transient optical signal group.

6. The method for step-by-step filtering of false lightning signals as described in claim 1, characterized in that, The process involves constructing a second sliding window using the spatial index grid of the transient light signal group determined by the preprocessed transient light signal group, and performing spatiotemporal clustering analysis on the transient light signal groups within the second sliding window that meet the second preset conditions to obtain the lightning corresponding to the transient light signal to be processed. Specifically, this includes: Centered on each grid point of the spatial index grid of the preprocessed transient optical signal group, a second preset grid is used as the second sliding window, and there are several second sliding windows; Calculate the time interval between any two groups of signals within the second sliding window, and filter the signal groups that meet the preset time conditions; The minimum spherical distance between any two pixels within a signal group that meets the preset time condition is taken as the spatial distance between the two signal groups. Based on the spatial distance and temporal distance between the two sets of signals, calculate the time-space weighted Euclidean distance; The transient light signal groups whose time-space weighted Euclidean distance satisfies the second preset condition are grouped into a single lightning bolt. The transient light signal group within the second sliding window is traversed to obtain the lightning corresponding to the transient light signal to be processed.

7. The method for step-by-step filtering of false lightning signals as described in claim 1, characterized in that, The process of sequentially filtering out pseudo-continuous spurious signals and discrete spurious signals at the lightning level corresponding to the transient optical signal to be processed specifically includes: Principal component analysis is performed on lightning that includes two or more transient light signals corresponding to the transient light signals to be processed, to obtain the aspect ratio, secondary axis, energy variation coefficient, and time variation coefficient of the lightning. Based on the aspect ratio of the lightning, the secondary axis of the lightning, the energy variation coefficient of the lightning, and the time variation coefficient of the lightning, false signals are identified, and false signals of lightning level pseudo-continuous false signals are filtered out to obtain lightning with the false signals filtered out. Discrete random noise in the lightning that is used to filter out false signals is identified, and discrete false signals of lightning are filtered out.

8. The method for filtering out false lightning signals step by step as described in claim 7, characterized in that, The process of identifying false signals based on the aspect ratio, secondary axis, energy variation coefficient, and time variation coefficient of the lightning, performing lightning-level pseudo-continuous false signal filtering, and obtaining lightning without false signals specifically includes: If the aspect ratio of the lightning is greater than the third preset threshold, the secondary axis of the lightning is less than the second axis threshold, the energy variation coefficient of the lightning is less than the third coefficient threshold, and the time variation coefficient of the lightning is less than the fourth coefficient threshold, then the lightning is a false signal and a false signal filtering of the lightning level is performed. For each lightning bolt in the lightning corresponding to the transient light signal to be processed, lightning-level pseudo-continuous spurious signal filtering is performed sequentially to obtain the lightning with the spurious signal filtered out. The process of determining discrete random noise in the lightning strikes from which false signals are filtered, and performing discrete false signal filtering for lightning strikes, specifically includes: If the duration of a lightning strike in the lightning strikes for which false signals are filtered out is less than the observation time of a single frame, and it contains only one transient light signal group, then the lightning strike is discrete random noise, and discrete false signals of lightning strikes are filtered out. The discrete spurious signal filtering operation is performed sequentially on each lightning bolt in the lightning bolts for which spurious signals are filtered out.

9. The method for step-by-step filtering of false lightning signals as described in claim 1, characterized in that, The step of constructing a second sliding window using the spatial index grid of the transient light signal group determined by the preprocessed transient light signal group, and performing spatiotemporal clustering analysis on the transient light signal groups within the second sliding window that meet the second preset conditions to obtain the lightning corresponding to the transient light signal to be processed, further includes: The latitude and longitude of the lightning corresponding to the transient light signal to be processed, the spatial lightning density of the lightning corresponding to the transient light signal to be processed, and the spatiotemporal lightning density of the lightning corresponding to the transient light signal to be processed are determined, and the lightning corresponding to the transient light signal to be processed is located.

10. A step-by-step filtering system for spurious lightning signals based on principal component analysis, characterized in that, The lightning false signal filtering system is applied to lightning observation data from geostationary meteorological satellites. The lightning false signal filtering system includes: The acquisition module acquires the transient optical signal to be processed. The transient optical signal preprocessing module performs pseudo-continuous false signal filtering at the transient optical signal level based on the aspect ratio and energy variation coefficient of the transient optical signal within the sliding time window of the transient optical signal to be processed, thereby obtaining the preprocessed transient optical signal. The transient optical signal group generation module determines the clustering core based on the spatial index grid of the transient optical signal determined by the preprocessed transient optical signal, constructs a first sliding window, and clusters the transient optical signals that meet the first preset conditions within the first sliding window to generate transient optical signal groups. The transient optical signal group preprocessing module performs pseudo-continuous false signal filtering at the transient optical signal group level based on the aspect ratio of the transient optical signal group, the first two principal components of the transient optical signal group principal component analysis, and the energy variation coefficient of the transient optical signal group, to obtain the preprocessed transient optical signal group. The lightning generation module constructs a second sliding window using the spatial index grid of the transient light signal group determined by the preprocessed transient light signal group. It then performs spatiotemporal clustering analysis on the transient light signal groups that meet the second preset conditions within the second sliding window to obtain the lightning corresponding to the transient light signal to be processed. The lightning false signal filtering module sequentially performs lightning-level pseudo-continuous false signal filtering and lightning discrete false signal filtering on the lightning corresponding to the transient light signal to be processed; The output module repeatedly executes the steps corresponding to the transient light signal preprocessing module - lightning false signal filtering module on the transient light signal to be processed until the number of transient light signals before and after filtering is the same, thus completing the lightning false signal filtering and outputting the real lightning signal.

Images