A multi-source heterogeneous data fusion sky-earth-intelligent monitoring system

Abstract

The present invention relates to the technical field of data processing, and provides a sky-ground-intelligent monitoring system for multi-source heterogeneous data fusion. It acquires ground sensor network data and satellite remote sensing data, and performs spatio-temporal calibration on the obtained ground sensor network data and satellite remote sensing data; constructs a multi-dimensional spatio-temporal association model to further analyze the data calibrated by the spatio-temporal dynamic calibration module, and optimize the fusion weights of the spatial and temporal association matrices; implements collaborative correction of multi-source data to optimize the spatio-temporal joint association matrix and eliminate spatial and temporal deviations; finally fuses and calibrates the processed multi-source heterogeneous data to provide data output. The present invention ensures that multi-source data can be fused with the same accuracy, and also improves the system's dynamic response ability to the credibility of different data sources, effectively reducing the errors caused by data deviations and inconsistencies, and improving the accuracy and reliability of monitoring results.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular, to a sky-earth intelligent monitoring system for multi-source heterogeneous data fusion. Background Art

[0002] With the development of remote sensing technology and ground sensor technology, the sky-earth integrated intelligent monitoring system has gradually become an important application means in the fields of environmental monitoring, meteorological prediction, disaster warning, etc. Especially in large-scale and multi-scale environmental monitoring tasks, the combination of ground sensors and satellite remote sensing data can provide more comprehensive and accurate information. Currently, the ground sensor network is mainly used to provide high-precision and real-time local environmental data, covering various environmental parameters such as temperature, humidity, and air pressure; while satellite remote sensing technology can cover a large range of areas and obtain important environmental information such as atmospheric composition, cloud cover, and surface temperature. With the continuous progress of sensor data acquisition technology and satellite remote sensing technology, multi-source heterogeneous data fusion technology has gradually become a key technology to improve monitoring accuracy and timeliness. This technology unifies the spatio-temporal characteristics of different data sources for modeling and fusion, thereby making up for the limitations of a single data source and providing a more comprehensive and efficient monitoring means.

[0003] However, the existing technology still faces many challenges in practical applications. For example, the spatio-temporal inconsistency between ground sensors and satellite remote sensing data is an issue that cannot be ignored. The spatial distribution of ground sensors is limited, and the data update frequency is relatively high, while satellite remote sensing data is restricted by satellite orbits, observation angles, and imaging frequencies, resulting in large differences in time and space resolutions between the two. Existing technologies mostly use static correction or linear interpolation methods to correct spatio-temporal deviations, but this method often ignores the complex non-linear relationships between data, resulting in insufficient correction effects; existing multi-source data fusion methods usually assume that each data source has the same credibility or fusion weight, and do not fully consider the reliability differences of different data sources. Especially in abnormal weather conditions or high-density monitoring areas, large fusion errors may occur, etc., affecting the accuracy of the final data fusion. Summary of the Invention

[0004] Embodiments of the present invention provide a sky-earth intelligent monitoring system for multi-source heterogeneous data fusion, which can at least to some extent reduce errors caused by data deviation and inconsistency, and improve the accuracy and reliability of monitoring results.

[0005] Other features and advantages of the present invention will become apparent through the following detailed description, or be learned in part through the practice of the present invention.

[0006] According to one aspect of the present invention, a sky-ground-intelligent monitoring system for multi-source heterogeneous data fusion is provided, including: a data acquisition module for acquiring ground sensor network data and satellite remote sensing data; a spatio-temporal dynamic calibration module for performing spatio-temporal calibration on the obtained ground sensor network data and satellite remote sensing data to ensure the consistency of the ground sensor network data and satellite remote sensing data in space and time; a spatio-temporal correlation matrix analysis module for constructing a multi-dimensional spatio-temporal correlation model, further analyzing the data calibrated by the spatio-temporal dynamic calibration module, optimizing the fusion weights of the spatial and temporal correlation matrices, and generating a spatio-temporal joint correlation matrix; an intelligent collaborative correction module for implementing collaborative correction of multi-source data, optimizing the spatio-temporal joint correlation matrix, and eliminating spatial and temporal deviations; a data fusion and calibration output module for finally fusing and calibrating the processed multi-source heterogeneous data and providing data output.

[0007] In the present invention, based on the foregoing solution, according to the basic information, the data acquisition module includes: constructing the network data collected by distributed sensors into a ground sensing matrix, where the spatial dimension maps the geographical coordinates of the sensors, and the time dimension records the sequential changes of parameters; decomposing the satellite remote sensing data into an atmospheric component matrix, a cloud cover matrix, and a surface temperature matrix.

[0008] In the present invention, based on the foregoing solution, the spatio-temporal dynamic calibration module includes: matching the spatial coordinates of the ground sensor network data with the spatial coordinates of the satellite remote sensing data, and aligning the data at the corresponding positions in the ground sensing matrix with the data in the atmospheric component matrix, the cloud cover matrix, and the surface temperature matrix; performing mutual verification on the ground sensor network data and the satellite remote sensing data, and when it is detected that the spatial coordinate deviation exceeds a preset threshold, automatically triggering a coordinate correction strategy to adjust the coordinate information of the ground or satellite data to ensure spatial alignment between the two; when the time step difference between the ground sensor network data and the satellite remote sensing data exceeds the set tolerance range, the interpolation compensation algorithm will be activated, and interpolation or extrapolation methods will be used to supplement the missing data.

[0009] In the present invention, based on the foregoing solution, the spatio-temporal correlation matrix analysis module includes: a spatial correlation degree calculation unit for analyzing the spatial distribution correlation between the ground sensing matrix and the satellite data matrix to generate a spatial correlation matrix; a time dynamic coupling unit for extracting the sequential characteristics of each data source through sliding window Fourier transform to construct a time dynamic coupling matrix; a correlation matrix optimization unit for iteratively optimizing the fusion weights of the spatial correlation matrix and the time coupling matrix to generate a spatio-temporal joint correlation matrix.

[0010] In the present invention, based on the foregoing solution, the generation of the spatial correlation matrix includes the following steps: converting the ground sensor network data into a spatio-temporal feature matrix according to the geographic coordinate mapping algorithm, where the rows of the matrix represent sensor nodes, the columns represent time steps, and each element in the matrix corresponds to the environmental parameters at a certain moment; decomposing the satellite remote sensing data into a grid matrix that is spatially aligned with the ground sensor network data, where each grid cell represents a fixed geographic area and the time dimension is synchronized with the ground sensor network data; calculating the mean of the time series data of each sensor node to obtain the spatial distribution characteristics of the ground data; calculating the mean of the time series data of each satellite grid cell to reflect the spatial distribution characteristics of the satellite observations; for the ground node i and the satellite grid j, the covariance value is the average of the product of the time series data of the two deviating from their respective means; normalizing the covariance matrix to obtain the spatial correlation matrix.

[0011] In the present invention, based on the foregoing solution, the correlation matrix optimization unit specifically includes: defining the spatial correlation matrix and the time coupling tensor , where is the number of sensor nodes, is the number of satellite observation grids, the spatial correlation matrix is the spatial correlation coefficient matrix of the ground sensor and satellite data, with the dimension of the number of sensors × the number of satellite grids; the time coupling tensor contains a set of dynamic coupling matrixes with frequency components, and the dimension of each matrix is the number of sensors × the number of satellite grids; calculating the error to obtain the total optimization objective value, and using the momentum-accelerated adaptive gradient descent algorithm to accelerate convergence; the adaptive gradient descent algorithm includes: marking the current iteration number as k, and respectively represent the current spatial weight matrix and time weight tensor; and are the learning rates of the spatial and time weights, and the initial values are set according to experience; for each sensor-grid pair (i,j), its gradient value is the product of the spatial correlation coefficient R(i,j) at the corresponding position and the element of the error matrix (i,j), and then multiplied by 2; calculating the partial derivative of the objective function with respect to : for each frequency component w and sensor-grid pair (i,j), its gradient value is the product of the time coupling coefficient D(i,j,w) and the element of the error matrix (i,j), and then multiplied by 2; performing spatial weight update; fusing the updated weight and the original correlation feature to generate a spatio-temporal joint matrix.

[0012] In the present invention, based on the foregoing solution, the error calculation includes: multiplying the spatial weight matrix α element by element with the spatial correlation matrix R, that is, multiplying the elements at the same positions of the two matrices, to obtain a weighted spatial correlation contribution matrix; for each frequency component w, multiplying the w-th layer slice of the temporal weight tensor β element by element with the w-th layer slice of the temporal coupling tensor D, and then adding up the results of all frequency components to obtain a weighted temporal coupling contribution matrix; adding the spatial correlation contribution matrix and the temporal coupling contribution matrix, and subtracting the ground truth matrix to obtain an error matrix.

[0013] In the present invention, based on the foregoing solution, the intelligent collaborative correction module includes: a deviation propagation analysis unit for calculating the propagation path of the spatial deviation through the Jacobian matrix; a dynamic weight matrix generation unit for constructing a dynamic credibility weight matrix based on the spatio-temporal joint correlation matrix, wherein the weight values of each data source are dynamically updated according to the real-time verification results; a collaborative correction execution unit for projecting high-dimensional data into a unified feature space by using matrix decomposition technology and realizing the collaborative correction of multi-source data through weighted least squares method.

[0014] In the present invention, based on the foregoing solution, the deviation propagation analysis unit includes: constructing a two-way deviation propagation equation; initializing the Jacobian matrix from the ground to the satellite based on historical data statistics and the Jacobian matrix from the satellite to the ground for the normalization of the spatial correlation matrix; using the least squares method for iterative optimization to minimize the mean square error between the model prediction deviation and the actual deviation, to obtain an optimized Jacobian matrix; obtaining the deviation propagation path according to the optimized Jacobian matrix.

[0015] In the present invention, based on the foregoing solution, the collaborative correction execution unit includes: performing low-rank matrix decomposition to obtain a low-dimensional representation; after obtaining the low-dimensional representation, applying the weighted least squares method for data correction; the applying the weighted least squares method for data correction includes: giving different weights to each data point according to the dynamic credibility weight matrix.

[0016] According to one aspect of the present invention, there is provided an electronic device, including: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, enabling the one or more processors to implement the sky-ground intelligent monitoring system for multi-source heterogeneous data fusion as described in the above embodiments.

[0017] According to one aspect of the present invention, there is provided a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the sky-ground intelligent monitoring system for multi-source heterogeneous data fusion provided in the above various optional implementation manners.

[0018] In the technical solution of the present invention, by comparing and aligning the coordinate and time information of different data sources, the errors caused by spatio-temporal deviations can be eliminated, ensuring that the data can be compared and fused under the same reference framework, improving the accuracy and reliability of the integrated data, avoiding monitoring errors caused by data deviations, and ensuring the basic accuracy of data fusion; effectively coupling the spatial distribution and time dynamics of the data sources. In this stage, by calculating the fusion of the spatial correlation degree and time dynamic characteristics, quantitative support can be provided for the association between different data sources. The process of optimizing the spatio-temporal association matrix enables the association between different data sources to be accurately quantified, thus avoiding the assumption of overly simple linear relationships and enhancing the complexity and adaptability of data fusion.

[0019] By calculating the deviation propagation path and dynamically adjusting the weights of the data sources, the system can accurately propagate and eliminate deviations between multi-source data. This not only ensures that multi-source data can be fused with the same accuracy, but also enhances the system's dynamic response ability to the credibility of different data sources. This correction process effectively reduces the errors caused by data deviations and inconsistencies, and improves the accuracy and reliability of the monitoring results.

[0020] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present invention, and are used together with the specification to explain the principles of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0022] Figure 1 Schematically shows a flowchart of a sky-ground intelligent monitoring system for multi-source heterogeneous data fusion in an embodiment of the present invention.

[0023] Figure 2 Schematically shows a flowchart of a sky-ground intelligent monitoring solution for multi-source heterogeneous data fusion in an embodiment of the present invention.

[0024] Figure 3 Schematically shows a block diagram of a spatio-temporal correlation matrix analysis module of a sky-earth-air intelligent monitoring system for multi-source heterogeneous data fusion in an embodiment of the present invention.

[0025] Figure 4 Schematically shows a block diagram of an intelligent collaborative correction module of a sky-earth-air intelligent monitoring system for multi-source heterogeneous data fusion in an embodiment of the present invention. Detailed implementation manners

[0026] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this invention will be more complete and comprehensive, and will fully convey the concept of the example embodiments to those skilled in the art.

[0027] In addition, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of the embodiments of the present invention. However, those skilled in the art will realize that the technical solutions of the present invention can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, well-known methods, devices, implementations, or operations are not shown or described in detail to avoid obscuring aspects of the present invention.

[0028] The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0029] The flowcharts shown in the drawings are only illustrative and do not necessarily include all the contents and operations / steps, nor do they necessarily have to be executed in the described order. For example, some operations / steps can be decomposed, while some operations / steps can be combined or partially combined, so the actual execution order may be changed according to the actual situation.

[0030] With the development of remote sensing technology and ground sensor technology, the sky-ground-integrated intelligent monitoring system has gradually become an important application means in the fields of environmental monitoring, meteorological prediction, disaster warning, etc. Especially in large-scale and multi-scale environmental monitoring tasks, the combination of ground sensors and satellite remote sensing data can provide more comprehensive and accurate information. Currently, ground sensor networks are mainly used to provide high-precision and real-time local environmental data, covering various environmental parameters such as temperature, humidity, and air pressure; while satellite remote sensing technology can cover large areas and obtain important environmental information such as atmospheric composition, cloud cover, and surface temperature. With the continuous progress of sensor data acquisition technology and satellite remote sensing technology, multi-source heterogeneous data fusion technology has gradually become a key technology for improving monitoring accuracy and timeliness. This technology unifies the modeling and fusion of the spatio-temporal characteristics of different data sources, thereby making up for the limitations of a single data source and providing a more comprehensive and efficient monitoring method.

[0031] However, existing technologies still face multiple challenges in practical applications. For example, the spatio-temporal inconsistency between ground sensors and satellite remote sensing data is an issue that cannot be ignored. The spatial distribution of ground sensors is limited, and the data update frequency is high, while satellite remote sensing data is restricted by satellite orbits, observation angles, and imaging frequencies, resulting in significant differences in the time and space resolutions of the two. Existing technologies mostly use static correction or linear interpolation methods to correct spatio-temporal deviations, but this method often ignores the complex non-linear relationships between data, resulting in insufficient correction effects; existing multi-source data fusion methods usually assume that each data source has the same credibility or fusion weight, and do not fully consider the reliability differences of different data sources. Especially in abnormal weather conditions or high-density monitoring areas, large fusion errors may occur; furthermore, most of the spatio-temporal correlation analyses in existing technologies rely on simple spatial distance calculations or time synchronization strategies, making it difficult to effectively extract the deep spatio-temporal characteristics between data, resulting in the construction of spatio-temporal correlation models being insufficient to fully reflect the dynamic coupling relationship between data sources and affecting the accuracy of the final data fusion.

[0032] Therefore, to address the above issues, the technical concept of the present invention is to achieve precise spatio-temporal alignment of ground sensor network data and satellite remote sensing data through a spatio-temporal dynamic calibration module, using geographic coordinate mapping and timestamp interpolation algorithms, thereby eliminating spatio-temporal deviations between different data sources. In particular, the two-way verification control unit further improves the accuracy of spatio-temporal calibration through the mutual verification mechanism between ground sensors and satellite remote sensing data; through the spatio-temporal correlation matrix analysis module, methods such as covariance matrix analysis and sliding window Fourier transform are used to comprehensively extract the spatial correlation and time dynamic coupling characteristics between data sources, and the fusion weights of the correlation matrix are optimized through an adaptive learning algorithm to improve the accuracy and reliability of data fusion; in addition, the intelligent collaborative calibration module generates a dynamic weight matrix based on the spatio-temporal joint correlation matrix, and combines matrix decomposition technology and weighted least squares method to achieve collaborative calibration of multi-source data, thus ensuring the accuracy and credibility of the final fusion result.

[0033] Figure 1 FIG. is a block diagram of a sky-ground intelligent monitoring system for multi-source heterogeneous data fusion according to an embodiment of the present invention. Figure 2 FIG. is a schematic diagram of data flow of a sky-ground intelligent monitoring system for multi-source heterogeneous data fusion according to an embodiment of the present invention. As Figure 1 and Figure 2 shown, a sky-ground intelligent monitoring system for multi-source heterogeneous data fusion according to an embodiment of the present invention includes:

[0034] A data acquisition module 100 for obtaining ground sensor network data and satellite remote sensing data; a spatio-temporal dynamic calibration module 200 for establishing a spatio-temporal reference framework for multi-source data and performing spatio-temporal calibration on the ground sensor network data and satellite remote sensing data obtained by the data acquisition module to ensure the spatial and temporal consistency of the ground sensor network data and satellite remote sensing data; a spatio-temporal correlation matrix analysis module 300 for constructing a multi-dimensional spatio-temporal correlation model, further analyzing the data calibrated by the spatio-temporal dynamic calibration module, optimizing the fusion weights of the spatial and temporal correlation matrices, and improving the spatio-temporal consistency of the data; an intelligent collaborative calibration module 400 for implementing collaborative calibration of multi-source data, optimizing the spatio-temporal joint correlation matrix generated by the spatio-temporal correlation matrix analysis module, eliminating spatial and temporal deviations, and ensuring the accuracy of the data fusion result; a data fusion and calibration output module 500 for finally fusing and calibrating multi-source heterogeneous data and providing data output.

[0035] In the embodiment of the present invention, the data acquisition module is used to obtain ground sensor network data and satellite remote sensing data, ensuring the synchronous acquisition and spatio-temporal alignment of multi-source heterogeneous data for subsequent analysis and optimization.

[0036] Specifically, the network data such as temperature, humidity, and air pressure collected by the distributed sensors are constructed into a ground sensing matrix, where the spatial dimension maps the geographical coordinates of the sensors, and the time dimension records the temporal variation of the parameters. By organizing the sensor network data in the manner of geographical coordinates and time series, the data of environmental parameters not only have spatial correlation but also can reflect their changing trends over time. This method provides an effective data structure for spatio-temporal correlation analysis and subsequent processing; the satellite remote sensing data is decomposed into an atmospheric component matrix, a cloud cover matrix, and a land surface temperature matrix, and the dimensions of each matrix are spatially aligned with the ground sensing matrix. By parsing the satellite remote sensing data into different matrices, each data source can independently provide specific environmental information that matches the ground sensing data. In addition, maintaining spatial alignment is the key to ensuring that the ground and satellite data are processed in the same spatial coordinate system, which helps to improve the data fusion accuracy.

[0037] It should be understood that the ground sensor network and satellite remote sensing data in the data acquisition module 100 are organized and parsed through the spatio-temporal feature matrix, ensuring the spatial and temporal consistency between multi-source data. First, through the construction of the ground sensing matrix, the ground sensor network data can be spatially mapped according to geographical coordinates, visualizing environmental parameters such as temperature, humidity, and air pressure in the spatial dimension, laying a foundation for subsequent data analysis. By recording the time dimension, the changing trends of environmental parameters over time can be clearly presented.

[0038] In terms of remote sensing data parsing, the satellite remote sensing data is deconstructed into multiple matrices, including an atmospheric component matrix, a cloud cover matrix, and a land surface temperature matrix. These matrices provide independent data dimensions for analyzing the changes in different atmospheric layers and land surface temperature. In this way, the mutual relationship and influence between the ground and air environments can be more intuitively reflected, providing an accurate data basis for spatio-temporal correlation analysis and subsequent data fusion.

[0039] Specifically, remote sensing data preprocessing techniques (such as principal component analysis techniques like PCA, ICA, SVD, etc.) are used to decompose and reduce the dimension of the remote sensing data, and then the corresponding matrices are obtained.

[0040] Exemplarily, assume that the original remote sensing data has multiple bands, such as infrared, visible light, near-infrared, etc. Each pixel has an observed value corresponding to different bands. These observed values are organized into a matrix, for example , where is the original data of each band, is the number of pixels; calculate the covariance matrix C of this matrix to describe the relationship between different bands; perform eigenvalue decomposition on the covariance matrix C to obtain eigenvalues and eigenvectors:

[0041] ,

[0042] Among them, is the eigenvalue, is the eigenvector; sort the eigenvalues, select the eigenvectors corresponding to the largest several eigenvalues, and these eigenvectors constitute the principal components.

[0043] Project the original data onto the selected principal components to obtain the matrix Z after dimensionality reduction. Among them, the matrix Z after principal component analysis is a low-dimensional matrix, which represents the result of dimensionality reduction of the information in different bands. Each element represents the principal component value of a pixel, and these data must be mapped back to the geographic coordinate system so that they can be docked with the actual geographic data. Remote sensing data usually uses geographic references (such as longitude and latitude coordinates or projection coordinate systems) for registration. It can be seen that this technical solution for organizing and analyzing multi-source data in the form of a spatio-temporal matrix avoids problems such as data inconsistency that may occur in traditional methods, enabling subsequent steps such as spatio-temporal dynamic calibration and correlation analysis to be carried out under a unified data framework, thereby improving the accuracy and efficiency of the entire system.

[0044] The present invention solves the problem of spatio-temporal alignment of multi-source data by proposing a method for constructing a spatio-temporal feature matrix. Specifically, the data of the ground sensor network enables the environmental parameters at different positions and times to be effectively mapped to a unified coordinate system through the construction of the spatio-temporal feature matrix, avoiding spatio-temporal mismatches between data. At the same time, the decomposition and spatio-temporal alignment of satellite remote sensing data further enhance the comparability and fusion of different data sources. Through these optimization measures, problems such as low data fusion accuracy in the prior art are solved, laying a solid foundation for subsequent data analysis and optimization.

[0045] In the embodiment of the present invention, the spatio-temporal dynamic calibration module 200 is mainly used to effectively perform spatio-temporal calibration on the ground sensing matrix, atmospheric composition matrix, cloud cover matrix, and surface temperature matrix generated by the data acquisition module. This module ensures the consistency of the ground sensor network data and satellite remote sensing data in the spatial and temporal dimensions, providing a basis for subsequent data analysis, correction, and optimization. The specific implementation steps of the spatio-temporal calibration module 200 are as follows:

[0046] The spatio-temporal dynamic calibration module 200 includes:

[0047] To ensure the spatial alignment of ground data and satellite data, the spatio-temporal calibration module matches the spatial coordinates (i.e., geographical locations) of the ground sensor network data with those of the satellite remote sensing data. By using technologies such as Geographic Information System (GIS), ground and satellite data with different resolutions can be mapped into the same spatial coordinate system. In this way, the data at each position in the ground sensing matrix will be aligned with the corresponding data in the atmospheric composition matrix, cloud cover matrix, and surface temperature matrix.

[0048] Specifically, the spatial coordinates of the ground sensor network data are aligned with those of the satellite remote sensing data through a geographic coordinate mapping algorithm.

[0049] Exemplarily, by adopting Geographic Information System (GIS) technology, accurate mapping between ground and satellite data with different spatial resolutions can be achieved.

[0050] Meanwhile, since the time acquisitions of the ground sensor network data and satellite remote sensing data are usually inconsistent, a timestamp interpolation algorithm is needed to perform alignment. The timestamp interpolation algorithm ensures the synchronization of the timestamps of the ground sensor network data and satellite remote sensing data. If the acquisition times of the ground sensors and satellite remote sensing data are different, the interpolation algorithm can supplement or interpolate the inconsistent time data according to the sampled time points, thus eliminating the influence caused by time differences. This operation can ensure the consistency of all data sources on the same time scale and avoid analysis errors caused by time asynchronization.

[0051] The spatio-temporal dynamic calibration module 200 also introduces a two-way verification control mechanism for cross-verifying the ground sensor network data and satellite remote sensing data. When the system detects that the spatial coordinate deviation exceeds a preset threshold, it will automatically trigger a coordinate correction strategy to adjust the coordinate information of the ground or satellite data to ensure their spatial alignment.

[0052] Specifically, the coordinate correction strategy calculates the error and compares the position differences between the ground sensors and satellite data in space. When the detected coordinate error exceeds the set tolerance range, error correction is performed based on the known geographical information. This process can eliminate spatial mismatches caused by factors such as insufficient accuracy of the ground sensor positions and deviations in satellite data, thereby improving the accuracy of data fusion.

[0053] Among them, error calculation includes: calculating the error by means of the squared difference of geographical coordinate differences, etc.; error correction includes: correcting the error through optimization algorithms (such as the least squares method or interpolation method).

[0054] In addition to spatial alignment, an interpolation compensation algorithm for time step differences is also applied in this module. When the time step difference between the ground sensor network data and the satellite remote sensing data exceeds the set tolerance range, the interpolation compensation algorithm will be activated, and interpolation or extrapolation methods will be used to supplement the missing data to ensure that all data has the same time resolution. Among them, the interpolation compensation algorithm includes linear interpolation, spline interpolation, Lagrange interpolation, etc. The present invention uses linear interpolation to process time differences. This measure can eliminate the impact of different time steps on data accuracy, enabling all data to be analyzed and fused on a unified time scale.

[0055] It should be noted that traditional spatio-temporal calibration techniques usually only focus on the alignment of spatial coordinates or simple time interpolation methods, but the present invention realizes precise synchronization in space and time by introducing a spatial-temporal dual alignment and correction mechanism, thereby improving the accuracy of data fusion and spatio-temporal analysis.

[0056] In addition, the bidirectional verification control mechanism not only ensures the spatial consistency of the ground data and the satellite remote sensing data, but also takes into account the data deviation that may be caused by external environmental factors (such as satellite position, sensor error, etc.). By continuously performing bidirectional verification and dynamic adjustment, the system can correct the data deviation in real time according to the actual situation, avoiding the problem of error accumulation not considered in traditional methods.

[0057] In the embodiment of the present invention, the spatio-temporal correlation matrix analysis module 300 is used to further optimize the data generated by the spatio-temporal dynamic calibration module 200 and analyze the calibrated data through a multi-dimensional spatio-temporal correlation model.

[0058] As Figure 3 shown, Figure 3 is a block diagram of the spatio-temporal correlation matrix analysis module in a sky-ground intelligent monitoring system for multi-source heterogeneous data fusion according to an embodiment of the present invention. The following is the detailed content of the spatio-temporal correlation matrix analysis module 300:

[0059] The spatio-temporal correlation matrix analysis module 300 includes a spatial correlation degree calculation unit 301, which analyzes the spatial distribution correlation between the ground sensing matrix and the satellite data matrix based on the covariance matrix to generate a spatial correlation matrix; a time dynamic coupling unit 302, which is used to extract the time series characteristics of each data source through sliding window Fourier transform to construct a time dynamic coupling matrix; and a correlation matrix optimization unit 303, which is used to iteratively optimize the fusion weight of the spatial correlation matrix and the time coupling matrix to generate a spatio-temporal joint correlation matrix.

[0060] Specifically, the spatial correlation degree calculation unit 301 is used to analyze the spatial distribution correlation between the ground sensing matrix and the satellite remote sensing data matrix by calculating the covariance matrix, and then generate a spatial correlation matrix. The spatial correlation matrix evaluates the similarity and correlation of these two types of data in space, and the specific steps are as follows:

[0061] First, the ground sensor network data is transformed into a spatio-temporal feature matrix according to the geographic coordinate mapping algorithm. The rows of this matrix represent sensor nodes (spatial dimension), the columns represent time steps, and each element in the matrix corresponds to an environmental parameter (such as temperature, humidity, etc.) at a certain moment. In this way, each row represents the change of a sensor node in the time dimension, constituting the spatio-temporal representation of the ground sensor network data.

[0062] The satellite remote sensing data (such as surface temperature, cloud cover, etc.) is decomposed into a grid matrix that is spatially aligned with the ground sensor network data. Each grid cell represents a fixed geographic area (for example, 1 km × 1 km), and the time dimension is synchronized with the ground sensing data. This grid representation ensures that the spatial information of the satellite remote sensing data can be precisely matched with the ground sensor network data, facilitating subsequent calculation of the spatial correlation degree.

[0063] To generate the spatial correlation matrix, for each time window (such as 24 hours or other appropriate time periods), calculate the covariance value between the ground sensor nodes and the satellite grid cells. The specific steps are as follows:

[0064] Calculate the mean value of the time series data of each sensor node to obtain the spatial distribution characteristics of the ground data; calculate the mean value of the time series data of each satellite grid cell to reflect the spatial distribution characteristics of the satellite observations; for the ground node i and the satellite grid j, the covariance value is the average of the product of the time series data of the two deviating from their respective means; normalize the covariance matrix to obtain the spatial correlation matrix, and the numerical range is mapped to [0,1]. The larger the value, the stronger the spatial correlation.

[0065] Through the above covariance calculation and normalization process, the finally obtained spatial correlation matrix can effectively describe the correlation between the ground sensor network data and the satellite remote sensing data in the spatial dimension. Each element of this matrix represents the spatial correlation degree between the corresponding sensor node and the satellite grid cell.

[0066] Specifically, the time dynamic coupling unit 302 is used to analyze the dynamic association mode between the ground and satellite data in the time dimension, and construct a coupling matrix that reflects the law of temporal evolution. The specific implementation steps are as follows:

[0067] First, segment the time series of ground and satellite data into sliding windows of fixed length (e.g., a 7-day window with a step size of 1 day); apply the Fourier transform to the data within each window to extract the main frequency components (e.g., diurnal variation, seasonal variation); the frequency characteristics of the ground data are represented by the spectral energy distribution of each sensor node, and the satellite data is the spectral energy distribution of each grid cell.

[0068] For each frequency component (e.g., diurnal variation frequency), calculate the spectral similarity between the ground and satellite data:

[0069] The similarity is defined as the cosine similarity of the spectral energy distributions of the two sequences, with a numerical range of [-1, 1]. A positive value indicates a consistent change trend; slide the window along the time axis to generate a dynamic coupling strength matrix, where the rows and columns of the matrix correspond to the ground nodes and satellite grids respectively, and the element value is the similarity of the current time window.

[0070] Assign higher weights to the low-frequency components (e.g., seasonal variation) as they reflect long-term trends; perform noise reduction filtering on the high-frequency components (e.g., hourly fluctuations) to avoid short-term interference; through the time-sliding mechanism, retain the memory of the historical coupling patterns to form a coupling tensor of temporal evolution.

[0071] Exemplarily, in the diurnal variation frequency component, the coupling strength between the ground sensors in urban areas and satellite data can reach above 0.9, while in the suburbs, it may drop to 0.6 due to the influence of local microclimates, indicating that the fusion weight in the time dimension needs to be reduced in the suburbs. Before a rainstorm event, the coupling strength between the ground humidity sensors and satellite cloud data will increase significantly (e.g., from 0.4 to 0.8), and the system can trigger the abnormal weather monitoring mode in advance.

[0072] Specifically, the correlation matrix optimization unit is used to dynamically optimize the fusion weights of spatial correlation and temporal coupling to generate a spatio-temporal joint correlation matrix, providing high-precision input for data correction and compensation. The specific implementation steps are as follows:

[0073] Define the spatial correlation matrix and the temporal coupling tensor , where is the number of sensor nodes, is the number of satellite observation grids. The spatial correlation matrix is the spatial correlation coefficient matrix between the ground sensors and satellite data, with a dimension of the number of sensors × the number of satellite grids; the temporal coupling tensor contains a set of dynamic coupling matrices for frequency components, and each matrix has a dimension of the number of sensors × the number of satellite grids. The initial values are set based on prior knowledge (such as sensor accuracy, satellite resolution).

[0074] Exemplarily, the spatial correlation matrix sets initial values based on sensor accuracy (such as calibration certificates) and satellite resolution (such as grid size). The weight of the grid where the high-precision sensor is located is set to 0.8, and the low-precision area is set to 0.5; the time coupling tensor is dynamically initialized according to the data update frequency. The weight of real-time satellite data (such as updated hourly) is set to 0.7, and the ground sensor (updated every 10 minutes) is set to 0.9.

[0075] Considering the complexity of spatio-temporal data and the high requirements for accuracy, in order to solve the existence of errors and inaccuracies, error calculation needs to be carried out first. This step is essential because through the quantification of errors, the deviation can be effectively measured. The specific operations are as follows:

[0076] Multiply the spatial weight matrix α and the spatial correlation matrix R element by element, that is, multiply the elements in the same position of the two matrices, to obtain the weighted spatial correlation contribution matrix; for each frequency component w, multiply the w-th layer slice of the time weight tensor β and the w-th layer slice of the time coupling tensor D element by element, and then sum the results of all frequency components to obtain the weighted time coupling contribution matrix; add the above-mentioned spatial correlation contribution matrix and the time coupling contribution matrix, and then subtract the ground truth matrix to obtain the error matrix. Calculate the square of the Frobenius norm of this error matrix (that is, the sum of the squares of all elements of the matrix), and sum the errors of all time steps to obtain the total optimization objective value.

[0077] Among them, the ground truth matrix is the data of high-precision weather stations.

[0078] It should be noted that through step-by-step weighting and optimization, the optimal spatio-temporal weights are found to minimize the prediction error as much as possible. By calculating the total optimization objective value, the error situation of the entire spatio-temporal data model can be centrally reflected.

[0079] Once the total optimization objective value is calculated, the parameters can be adjusted through an adaptive optimization algorithm. Specifically, the adaptive gradient descent algorithm with momentum acceleration is used to accelerate convergence. This algorithm adjusts the learning rates of spatial and temporal weights and uses the momentum factor to ensure the effectiveness of the historical update direction, thus avoiding the overfitting or underfitting problems caused by single updates. The specific steps are as follows: Use the adaptive gradient descent algorithm with momentum acceleration to adjust the weights. The specific process is as follows:

[0080] The current iteration number is marked as k, and respectively represent the current spatial weight matrix and time weight tensor; and are the learning rates of spatial and temporal weights, and the initial values are set according to experience (such as = 0.01, (where \(\alpha = 0.005\)), \(\gamma\) is the momentum factor (usually set to \(0.9\)) and is used to retain the historical update direction.

[0081] Calculate the partial derivative of the objective function with respect to as follows: For each sensor-grid pair \((i, j)\), its gradient value is the product of the spatial correlation coefficient \(R(i, j)\) at the corresponding position and the element of the error matrix \((i, j)\), multiplied by 2 (from the derivative of the squared error).

[0082] Calculate the partial derivative of the objective function with respect to as follows: For each frequency component \(w\) and sensor-grid pair \((i, j)\), its gradient value is the product of the time coupling coefficient \(D(i, j, w)\) and the element of the error matrix \((i, j)\), multiplied by 2.

[0083] The spatial weight update is expressed as:

[0084] ,

[0085] where and are dynamically adjusted learning rates, and is the momentum factor.

[0086] Fuse the updated weights with the original correlation features to generate a spatio-temporal joint matrix , where \(\tau\) is the time step, and the operation is as follows:

[0087] Perform weighted fusion on the optimized spatial weight matrix and the temporal weight tensor: Multiply the spatial correlation coefficient matrix element-wise with the spatial weight matrix to highlight high-confidence regions; multiply the coupling matrix for each frequency component with the corresponding temporal weight to strengthen the contribution of long-term trends; stack the spatial and temporal fusion results along the time axis to generate a spatio-temporal joint correlation matrix, and each element of the matrix comprehensively reflects the data reliability at a specific location at a specific time.

[0088] As Figure 4 shown, Figure 4 is a block diagram of the intelligent collaborative correction module in a sky-earth intelligent monitoring system for multi-source heterogeneous data fusion according to an embodiment of the present invention. The following is the detailed content of the intelligent collaborative correction module 400:

[0089] In an embodiment of the present invention, the intelligent collaborative correction module 400 includes a deviation propagation analysis unit 401 for calculating the propagation path of spatial deviation through a Jacobian matrix; a dynamic weight matrix generation unit 402 for constructing a dynamic credibility weight matrix based on the spatio-temporal joint correlation matrix, wherein the weight values of each data source are dynamically updated according to real-time verification results; and a collaborative correction execution unit 403 for projecting high-dimensional data into a unified feature space using matrix factorization technology and achieving collaborative correction of multi-source data through weighted least squares method.

[0090] Specifically, the deviation propagation analysis unit 401 specifically includes:

[0091] Assume that the ground sensor deviation will affect satellite data through spatial correlation, and vice versa, and construct a two-way deviation propagation equation:

[0092] ,

[0093] Wherein, is the satellite data deviation matrix, is the ground sensor deviation vector, is the Jacobian matrix from ground to satellite, is the Jacobian matrix from satellite to ground, and are noise terms respectively. In order to obtain the optimized Jacobian matrix, the least squares method is used for the iterative optimization process to minimize the mean square error between the predicted deviation and the actual deviation. Specifically: Based on historical data statistics, initialize and as the normalization of the spatial correlation matrix; use the least squares method for iterative optimization to minimize the mean square error between the model prediction deviation and the actual deviation to obtain the optimized Jacobian matrix, and finally achieve the prediction result with the minimum mean square error.

[0094] Similar to the Jacobian matrix from ground to satellite, the Jacobian matrix from satellite to ground can also be optimized by the least squares method to minimize the mean square error between the predicted deviation and the actual deviation of the model.

[0095] Through the propagation formula, the Jacobian matrix not only provides the deviation propagation path from one domain to another, but also reveals the propagation mode and direction of deviation in space.

[0096] By observing the data at multiple moments, how the Jacobian matrix is continuously updated at different time steps and thus affects the deviation propagation; by the Jacobian matrices at different positions (such as the relative positions between ground sensors and satellites), show how the deviation propagates from one spatial position to another.

[0097] In an embodiment of the present invention, the dynamic weight matrix generation unit 402 includes: generating a dynamic credibility weight matrix based on the spatio-temporal joint correlation matrix and the real-time verification result, and quantifying the reliability of each data source in the spatial and temporal dimensions. The specific operations are as follows:

[0098] Based on the spatial and temporal correlations in the spatio-temporal joint correlation matrix, a credibility model for each data source is established. Specifically, by analyzing the weights in the matrix, the system evaluates the influence of each data source at a specific spatial location and time step.

[0099] For the ground sensor network data, the model considers the spatial distribution characteristics and time variability of the data; for the satellite remote sensing data, the model mainly considers the data resolution, coverage range, and the correlation strength with the ground sensor network data.

[0100] The real-time verification result comes from the system's error analysis of the current moment, such as the comparison result of the ground sensor network data and the satellite remote sensing data. If the error of a certain data source is large, or the verification result shows its low reliability, the system will reduce the weight of this data source in the credibility weight matrix.

[0101] Conversely, when the actual error of a certain data source is small or its spatial and temporal matching degree is high, the system will correspondingly increase the weight of this data source.

[0102] With the change of real-time data, the dynamic credibility weight matrix will be dynamically updated according to the real-time verification result. Each time it is updated, the system will re-evaluate the reliability of all data sources and adjust their weights according to the performance of the current data source in space and time.

[0103] Through an iterative method, it is ensured that the weights of the data sources can accurately reflect their true reliability in the spatio-temporal dimension during the correction process, thereby affecting the subsequent correction process.

[0104] Through the above dynamic adjustment steps, a new dynamic credibility weight matrix is finally generated. Each element in the matrix represents the reliability of a certain data source at a specific spatio-temporal point, and the numerical range is usually between [0, 1]. The higher the value, the stronger the reliability of the data source.

[0105] In an embodiment of the present invention, the collaborative correction execution unit 403 includes:

[0106] The collaborative correction execution unit first receives the dynamic credibility weight matrix from the dynamic weight matrix generation unit 402, and this dynamic credibility weight matrix provides weight values for the subsequent weighted least squares method.

[0107] It should be noted that matrix factorization technology generally involves decomposing a high-dimensional data matrix into multiple low-dimensional matrix forms to facilitate the extraction of important information from a more concise feature space. In the present invention, the goal of decomposition is to project the ground sensor network data matrix and the satellite remote sensing data matrix into a common feature space so that comparison and calibration can be performed under a unified coordinate system.

[0108] Specifically, through low-rank matrix factorization, the goal is to find a low-dimensional representation that contains the most important features of the ground and satellite data, facilitating subsequent calibration.

[0109] After obtaining the low-dimensional representation, the weighted least squares method is applied for data calibration. This method minimizes the weighted error function, making the calibrated data as close as possible to the actual values.

[0110] It can be seen that the basic idea of the weighted least squares method is to assign different weights to each data point according to the dynamic credibility weight matrix, and data points with higher weight values will have a greater impact on the final result during the calibration process.

[0111] Through the weighted least squares method, the calibration result can maximize the preservation of the accuracy of high-credibility data sources while appropriately correcting low-credibility data sources, ultimately achieving the best effect of data fusion.

[0112] Through the optimization process of the weighted least squares method, the finally output calibrated data will be a result with relatively high consistency after multi-source data fusion. These results not only integrate the advantages of ground sensors and satellite remote sensing data but also ensure that the influence of each data source is reasonably controlled through dynamic weight adjustment.

[0113] The data fusion and calibration output module 500 is used to perform the final fusion and calibration of the multi-source heterogeneous data processed by the foregoing modules to provide consistent and accurate data output.

[0114] Specifically, the fused data is output in a unified format, providing a standardized data interface for subsequent applications.

[0115] This interface can be docked with external application systems to achieve seamless data flow.

[0116] In addition, the fusion result can be real-time fed back to the monitoring module of the system. If there are problems with the data quality, the system will automatically trigger the calibration and data update mechanism to ensure the long-term stability and accuracy of the data.

[0117] By simultaneously obtaining environmental data from different sources through a ground sensor network and a satellite remote sensing system, it lays the foundation for subsequent processing. The integration of this multi-source data not only improves the comprehensiveness of monitoring but also enables the monitoring system to adapt to different environmental monitoring requirements by providing cross-time and cross-space data perspectives.

[0118] Next, through the spatio-temporal dynamic calibration step, the consistency of ground sensor data and satellite remote sensing data in space and time is ensured. Comparing and aligning the coordinate and time information of different data sources can eliminate the errors caused by spatio-temporal deviations, ensuring that the data can be compared and fused under the same reference framework. The beneficial effect of this step is to improve the accuracy and reliability of the integrated data, avoid monitoring errors caused by data deviations, and ensure the basic accuracy of data fusion.

[0119] On this basis, the technical solution further effectively couples the spatial distribution and time dynamics of data sources through the construction and optimization of a spatio-temporal correlation matrix. At this stage, by calculating the fusion of spatial correlation degrees and time dynamic characteristics, it can provide quantitative support for the correlation between different data sources. The process of optimizing the spatio-temporal correlation matrix enables the correlation between different data sources to be accurately quantified, thus avoiding the assumption of overly simple linear relationships and enhancing the complexity and adaptability of data fusion.

[0120] The introduction of an intelligent collaborative correction module further enhances the system's ability to correct data deviations. By calculating the deviation propagation path and dynamically adjusting the weights of data sources, the system can accurately propagate and eliminate deviations among multi-source data. This not only ensures that multi-source data can be fused at the same accuracy but also enhances the system's dynamic response ability to the credibility of different data sources. This correction process effectively reduces the errors caused by data deviations and inconsistencies, improving the accuracy and reliability of monitoring results.

[0121] Finally, the data processed by the output module after data fusion and calibration can provide unified and accurate monitoring results. These results not only have high accuracy but also can adapt to environmental changes in real time, providing more accurate monitoring information. With the cooperation of each step, the entire technical solution significantly improves the intelligent level and accuracy of data processing, ensures the seamless fusion of multi-source heterogeneous data in the spatial and temporal dimensions, and greatly enhances the comprehensive performance of the sky-ground-intelligent monitoring system.

[0122] It should be noted that the computer-readable medium shown in the embodiments of the present invention can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. In the present invention, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable computer program. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and this computer-readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted by any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0123] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. Among them, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code, and the above module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings. For example, two consecutive blocks shown can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0124] The units involved in the embodiments of the present invention can be implemented in software or in hardware, and the described units can also be provided in a processor. Among them, the names of these units do not, in some cases, constitute a limitation on the units themselves.

[0125] According to one aspect of the present invention, there is provided a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in the above various optional implementation manners.

[0126] As another aspect, the present invention further provides a computer-readable medium, which may be included in the electronic device described in the above embodiments; or may exist alone without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements a multi-source heterogeneous data fusion sky-ground intelligent monitoring system described in the above embodiments.

[0127] It should be noted that although several modules or units of a device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present invention, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0128] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described here can be implemented by software or by a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (such as a personal computer, a server, a touch terminal, or a network device, etc.) to execute the methods according to the embodiments of the present invention.

[0129] After considering the specification and practicing the disclosed embodiments here, those skilled in the art will readily conceive of other embodiments of the present invention. The present invention is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include known common knowledge or conventional technical means in the technical field not disclosed by the present invention.

[0130] It should be understood that the present invention is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A sky-earth intelligent monitoring system for multi-source heterogeneous data fusion, characterized in that, Including: A data acquisition module for obtaining ground sensor network data and satellite remote sensing data; A spatio-temporal dynamic calibration module for performing spatio-temporal calibration on the obtained ground sensor network data and satellite remote sensing data to ensure the consistency of the ground sensor network data and satellite remote sensing data in space and time; A spatio-temporal correlation matrix analysis module for constructing a multi-dimensional spatio-temporal correlation model, further analyzing the data calibrated by the spatio-temporal dynamic calibration module, optimizing the fusion weights of the spatial and temporal correlation matrices, and generating a spatio-temporal joint correlation matrix; An intelligent collaborative correction module for implementing collaborative correction of multi-source data, optimizing the spatio-temporal joint correlation matrix, and eliminating spatial and temporal biases; A data fusion and calibration output module for finally fusing and calibrating the processed multi-source heterogeneous data and providing data output; The data acquisition module includes: Constructing the network data collected by distributed sensors into a ground sensing matrix, where the spatial dimension maps the geographical coordinates of the sensors, and the time dimension records the temporal variation of the parameters; Decomposing the satellite remote sensing data into an atmospheric component matrix, a cloud cover matrix, and a surface temperature matrix; The spatio-temporal dynamic calibration module includes: Matching the spatial coordinates of the ground sensor network data with the spatial coordinates of the satellite remote sensing data, and aligning the data at the corresponding positions in the ground sensing matrix with the data in the atmospheric component matrix, cloud cover matrix, and surface temperature matrix; Mutually verifying the ground sensor network data and satellite remote sensing data. When it is detected that the spatial coordinate deviation exceeds a preset threshold, a coordinate correction strategy is automatically triggered to adjust the coordinate information of the ground or satellite data to ensure spatial alignment between the two; When the time step difference between the ground sensor network data and the satellite remote sensing data exceeds the set tolerance range, the interpolation compensation algorithm will be activated, and interpolation or extrapolation methods will be used to supplement the missing data.

2. The multi-source heterogeneous data fusion sky-earth-intelligent monitoring system according to claim 1, characterized in that, The spatio-temporal correlation matrix analysis module includes: A spatial correlation degree calculation unit for analyzing the spatial distribution correlation between the ground sensing matrix and the satellite data matrix and generating a spatial correlation matrix; A time dynamic coupling unit for extracting the temporal sequence features of each data source through sliding window Fourier transform and constructing a time dynamic coupling matrix; A correlation matrix optimization unit for iteratively optimizing the fusion weights of the spatial correlation matrix and the time coupling matrix to generate a spatio-temporal joint correlation matrix.

3. The multi-source heterogeneous data fusion sky-earth-intelligent monitoring system according to claim 2, characterized in that, The generation of the spatial correlation matrix includes the following steps: Converting the ground sensor network data into a spatio-temporal feature matrix according to the geographical coordinate mapping algorithm. The rows of the matrix represent sensor nodes, the columns represent time steps, and each element in the matrix corresponds to the environmental parameters at a certain moment; Decomposing the satellite remote sensing data into a grid matrix that is spatially aligned with the ground sensor network data. Each grid cell represents a fixed geographical area, and the time dimension is synchronized with the ground sensor network data; Calculating the mean value of the time series data of each sensor node to obtain the spatial distribution characteristics of the ground data; calculating the mean value of the time series data of each satellite grid cell to reflect the spatial distribution characteristics of satellite observations; For the ground node i and the satellite grid j, the covariance value is the average of the products of the time series data of the two deviating from their respective means; Normalize the covariance matrix to obtain the spatial correlation matrix.

4. The multi-source heterogeneous data fusion sky-earth-intelligent monitoring system according to claim 3, characterized in that, The correlation matrix optimization unit specifically includes: Define the spatial correlation matrix and the time coupling tensor , where is the number of sensor nodes, is the number of satellite observation grids, and the spatial correlation matrix is the spatial correlation coefficient matrix between ground sensors and satellite data, with dimensions of the number of sensors × the number of satellite grids; Time coupling tensor including a set of dynamic coupling matrices containing Perform error calculation to obtain the total optimization objective value, and use the momentum-accelerated adaptive gradient descent algorithm to accelerate convergence; The adaptive gradient descent algorithm includes: The current iteration number is marked as k, and represent the current spatial weight matrix and temporal weight tensor, respectively; and are the learning rates for spatial and temporal weights, and their initial values are set according to experience; For each sensor-grid pair (i, j), its gradient value is the product of the spatial correlation coefficient R(i, j) at the corresponding position and the element of the error matrix (i, j), multiplied by 2; Calculate the partial derivative of the objective function with respect to : For each frequency component w and sensor-grid pair (i, j), the gradient value is the product of the time coupling coefficient D(i, j, w) and the element of the error matrix (i, j), multiplied by 2; ​ Perform spatial weight update; Fuse the updated weight and the original correlation feature to generate a spatio-temporal joint matrix.

5. The multi-source heterogeneous data fusion sky-earth-intelligent monitoring system according to claim 4, wherein, The error calculation includes: Multiply the spatial weight matrix α and the spatial correlation matrix R element by element, that is, multiply the elements at the same positions of the two matrices, to obtain the weighted spatial correlation contribution matrix; For each frequency component w, multiply the w-th slice of the time weight tensor β and the w-th slice of the time coupling tensor D element by element, and then sum the results of all frequency components to obtain the weighted time coupling contribution matrix; Add the spatial correlation contribution matrix and the time coupling contribution matrix, and subtract the ground truth matrix to obtain the error matrix.

6. The multi-source heterogeneous data fusion sky-earth-intelligent monitoring system according to claim 1, characterized in that, The intelligent collaborative correction module includes: A deviation propagation analysis unit for calculating the propagation path of the spatial deviation through the Jacobian matrix; A dynamic weight matrix generation unit for constructing a dynamic credibility weight matrix based on the spatio-temporal joint correlation matrix, where the weight values of each data source are dynamically updated according to the real-time verification results; A collaborative correction execution unit for projecting high-dimensional data into a unified feature space using matrix factorization technology and realizing collaborative correction of multi-source data through weighted least squares.

7. The multi-source heterogeneous data fusion sky-earth-intelligent monitoring system according to claim 6, characterized in that, The deviation propagation analysis unit includes: Construct a two-way deviation propagation equation; Initialize the Jacobian matrix from the ground to the satellite based on historical data statistics and the Jacobian matrix from the satellite to the ground for the normalization of the spatial correlation matrix; Use the least squares method for iterative optimization to minimize the mean square error between the predicted deviation and the actual deviation, and obtain the optimized Jacobian matrix; Obtain the deviation propagation path according to the optimized Jacobian matrix.

8. The multi-source heterogeneous data fusion sky-earth-intelligent monitoring system according to claim 7, characterized in that, The collaborative correction execution unit includes: Perform low-rank matrix factorization to obtain a low-dimensional representation; After obtaining the low-dimensional representation, apply the weighted least squares method for data correction; The application of the weighted least squares method for data correction includes: Give different weights to each data point according to the dynamic credibility weight matrix.

Images