Big data intelligent processing method and system based on space-air-ground integration

Abstract

The invention discloses a big data intelligent processing method and system based on space-air-ground integration. The method comprises the following steps: acquiring space-air-ground data; splitting the air-space-ground data to obtain air data, space data and ground data; respectively extracting feature values in the air data, the sky data and the ground data, and carrying out fusion processing to obtain a fused feature value; performing residual analysis on the fusion feature value and feature values in the air data, the sky data and the ground data to obtain a fusion feature difference set; if the fusion feature difference values in the fusion feature difference value set are all smaller than or equal to a preset first feature difference threshold value, the corresponding fusion feature value is normal; generating fusion data corresponding to the space-air-ground data according to the fusion feature value; and sending the space-air-ground data and the corresponding fusion data to a preset management end for display. According to the invention, fusion processing is carried out on the space-air-ground data, and the accuracy of data acquisition is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and more specifically, to a big data intelligent processing method and system based on space-air-ground integration. Background Art

[0002] With the development of technology, the data acquisition has evolved from one dimension to multiple dimensions, achieving the diversity of data sources and thus ensuring the accuracy of data. However, there are currently problems such as independent storage of space-air-ground data, lack of unified standards and coordination mechanisms, and difficulty in data integration. Summary of the Invention

[0003] To solve at least one of the above technical problems, the present invention provides a big data intelligent processing method and system based on space-air-ground integration, which can improve the accuracy of the collected data.

[0004] The first aspect of the present invention provides a big data intelligent processing method based on space-air-ground integration, including:

[0005] Obtain space-air-ground data;

[0006] Split the space-air-ground data to obtain space data, air data, and ground data;

[0007] Extract the eigenvalue in the space data, air data, and ground data respectively, and perform fusion processing to obtain a fusion eigenvalue;

[0008] Perform residual analysis on the fusion eigenvalue and the eigenvalue in the space data, air data, and ground data respectively to obtain a set of fusion feature differences;

[0009] If the fusion feature differences in the set of fusion feature differences are all less than or equal to a preset first feature difference threshold, the corresponding fusion eigenvalue is normal;

[0010] Generate fusion data corresponding to the space-air-ground data according to the fusion eigenvalue;

[0011] Send the space-air-ground data and the corresponding fusion data to a preset management terminal for display.

[0012] In this solution, after extracting the eigenvalue in the space data, air data, and ground data, it further includes:

[0013] Obtain the time series corresponding to the space-air-ground data, and construct a space data eigenvalue set, an air data eigenvalue set, and a ground data eigenvalue set with the corresponding eigenvalue;

[0014] Set the space data eigenvalue set as ;

[0015] Set the air data eigenvalue set as ;

[0016] Set the ground data feature value set as ;

[0017] Randomly extract three data feature values at the same time node;

[0018] Perform difference calculation on any two of the three data feature values at the same time node to obtain a second feature difference;

[0019] If all the second feature differences are less than or equal to the preset second feature difference threshold, perform fusion processing on the corresponding three data feature values;

[0020] If there is a second feature difference greater than the preset second feature difference threshold, calibrate the data feature values in the three data feature sets to obtain the calibrated data feature values;

[0021] Perform fusion processing according to the calibrated data feature values.

[0022] In this solution, the step of calibrating the data feature values in the three data feature sets specifically includes:

[0023] When only one second feature difference is greater than the preset second feature difference threshold, determine the corresponding time node t based on the data feature value other than the two data feature values corresponding to the second feature difference;

[0024] When there are two second feature differences greater than the preset second feature difference threshold, determine the corresponding time node t based on the common data feature value among the two data feature values corresponding to the two second feature differences;

[0025] When there are three second feature differences greater than the preset second feature difference threshold, randomly extract one data feature value as the benchmark to determine the corresponding time node t;

[0026] According to the time node t, find the corresponding time range in the other two data feature value sets, and determine the time interval ratio k according to the corresponding time range;

[0027] According to the corresponding time range, find two data feature values of the data feature value set in the corresponding time range, and set them as and ;

[0028] The calibrated data feature value .

[0029] In this solution, it also includes:

[0030] When the data feature values in the data feature set are not calibrated, based on the same time node, multiply the data feature values by the corresponding dimension weight coefficients, accumulate the products, and obtain the fusion feature values corresponding to the time node;

[0031] When there are data feature values in the data feature set that are calibrated, set the reference data feature value as the reference data feature value and the corresponding time node as the reference time node;

[0032] Multiply the reference data feature value by the corresponding dimension weight coefficient to obtain the data feature weight value of the corresponding dimension;

[0033] Multiply the calibrated data feature values of other dimensions corresponding to the reference data feature value by the corresponding other dimension weight coefficients to obtain the data feature weight values of other dimensions;

[0034] Accumulate the data feature weight value corresponding to the reference data feature value and the data feature weight values of other dimensions to obtain the fusion feature value of the reference time node.

[0035] In this solution, the steps for obtaining the dimension weight coefficients further include:

[0036] Obtain the historical space-air-ground dataset for a preset first time period;

[0037] Extract the historical data feature values and historical fusion feature values corresponding to the historical space-air-ground data at any time node within the preset first time period;

[0038] Calculate the difference between the historical data feature values of different dimensions and the historical fusion feature values respectively to obtain a plurality of fourth feature differences;

[0039] If all the fourth feature differences are less than or equal to the preset fourth feature difference threshold, set all the dimension weight coefficients to one-third;

[0040] If there is at least one fourth feature difference less than or equal to the preset fourth feature difference threshold and there is at least one fourth feature difference greater than the preset fourth feature difference threshold, identify the fourth feature differences greater than the preset fourth feature difference threshold to obtain the identified fourth feature differences; set the dimension weight coefficient adjustment value to , and its formula is , where represents the fourth feature difference, Indicates the historical fusion eigenvalue; when a fourth feature difference value is less than or equal to a preset fourth feature difference threshold, there are two fourth feature difference values that are greater than the preset fourth feature difference threshold, that is, there are two dimensional weight coefficient adjustment values, indicating that the dimensional weight coefficient corresponding to the fourth feature difference value is one-third minus the dimensional weight coefficient adjustment value, and the other dimensional weight coefficient is one-third plus two dimensional weight coefficient adjustment values; when two fourth feature difference values are less than or equal to the preset fourth feature difference threshold, there is one fourth feature difference value that is greater than the preset fourth feature difference threshold, that is, there is one dimensional weight coefficient adjustment value, indicating that the dimensional weight coefficient corresponding to the fourth feature difference value is one-third minus the dimensional weight coefficient adjustment value, and the other two dimensional weight coefficient values are one-third plus one-half of the dimensional weight coefficient adjustment value;

[0041] If all fourth feature difference values are greater than the preset fourth feature difference threshold, the historical space-air-ground data of the corresponding time node will be deleted, and the historical space-air-ground data of other time nodes will be re-extracted until all fourth feature difference values are less than or equal to the preset fourth feature difference threshold or all the historical space-air-ground data in the historical space-air-ground data set are deleted.

[0042] This solution also includes:

[0043] Set the data eigenvalue with the benchmark as the benchmark data eigenvalue;

[0044] Calculate the difference between the calibrated data eigenvalue and the benchmark data eigenvalue to obtain the third feature difference;

[0045] If there are two third feature difference values greater than or equal to the preset third feature difference threshold, delete the data eigenvalues corresponding to the two third feature difference values, and then calculate the average value of the remaining two data eigenvalues to obtain the fusion eigenvalue;

[0046] When there are three third feature difference values greater than or equal to the preset third feature difference threshold, a data warning message corresponding to the time node will be triggered;

[0047] If there is one third feature difference value greater than or equal to the preset third feature difference threshold, or all third feature difference values are less than or equal to the preset third feature difference threshold, the currently calibrated data eigenvalue and the benchmark data eigenvalue are normal.

[0048] This solution also includes:

[0049] Form a team with any two of the empty data, sky data, and ground data, and conduct a comparative analysis to obtain a similarity value set;

[0050] Calculate the average value of the similarity values in the similarity value set to obtain the combined similarity value of the space-air-ground data;

[0051] Determine whether the combined similarity value of the space-air-ground data is greater than or equal to a preset first similarity threshold. If so, the current space-air-ground data is normal; if not, the current space-air-ground data is abnormal and an abnormal prompt message is triggered.

[0052] Send the abnormal prompt message to a preset management terminal for display.

[0053] The second aspect of the present invention provides a big data intelligent processing system based on space-air-ground integration, including a memory and a processor. A program of a big data intelligent processing method based on space-air-ground integration is stored in the memory. When the program of the big data intelligent processing method based on space-air-ground integration is executed by the processor, the following steps are implemented:

[0054] Obtain space-air-ground data;

[0055] Split the space-air-ground data to obtain space data, air data, and ground data;

[0056] Extract the feature values from the space data, air data, and ground data respectively, and perform fusion processing to obtain fusion feature values;

[0057] Perform residual analysis on the fusion feature values respectively with the feature values in the space data, air data, and ground data to obtain a set of fusion feature differences;

[0058] If the fusion feature differences in the set of fusion feature differences are all less than or equal to a preset first feature difference threshold, the corresponding fusion feature value is normal;

[0059] Generate fusion data corresponding to the space-air-ground data according to the fusion feature values;

[0060] Send the space-air-ground data and the corresponding fusion data to a preset management terminal for display.

[0061] In this solution, after extracting the feature values from the space data, air data, and ground data, the following is also included:

[0062] Obtain the time series corresponding to the space-air-ground data, and construct a space data feature value set, an air data feature value set, and a ground data feature value set with the corresponding feature values;

[0063] Set the space data feature value set as ;

[0064] Set the air data feature value set as ;

[0065] Set the ground data feature value set as ;

[0066] Randomly extract three data feature values at the same time node;

[0067] Calculate the difference between any two of the three data feature values at the same time node to obtain a second feature difference;

[0068] If all the second feature differences are less than or equal to a preset second feature difference threshold, perform fusion processing on the corresponding three data feature values;

[0069] If there is a second feature difference greater than the preset second feature difference threshold, calibrate the data feature values in the three data feature sets to obtain the calibrated data feature values;

[0070] Perform fusion processing according to the calibrated data feature values.

[0071] In this solution, the step of calibrating the data feature values in the three data feature sets specifically includes:

[0072] When only one second feature difference is greater than the preset second feature difference threshold, determine the corresponding time node t based on the data feature value other than the two data feature values corresponding to the second feature difference;

[0073] When there are two second feature differences greater than the preset second feature difference threshold, determine the corresponding time node t based on the common data feature value among the two data feature values corresponding to the two second feature differences;

[0074] When there are three second feature differences greater than the preset second feature difference threshold, arbitrarily extract one data feature value as the benchmark to determine the corresponding time node t;

[0075] According to the time node t, find the corresponding time range in the other two data feature value sets, and determine the time interval ratio k according to the corresponding time range;

[0076] According to the corresponding time range, find the two data feature values of the data feature value set in the corresponding time range, and set them as and ;

[0077] The calibrated data feature value .

[0078] A big data intelligent processing method and system based on space-air-ground integration disclosed by the present invention fuse space-air-ground data to improve the accuracy of the collected data. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 Shows a flowchart of a big data intelligent processing method based on space-air-ground integration of the present invention;

[0080] Figure 2The block diagram of a big data intelligent processing system based on the integration of space, air and ground of the present invention is shown. Detailed implementation manners

[0081] In order to be able to more clearly understand the above objects, features and advantages of the present invention, the present invention will be further described in detail below with reference to the drawings and specific implementation manners. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

[0082] Many specific details are set forth in the following description in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0083] Figure 1 The flowchart of a big data intelligent processing method based on the integration of space, air and ground of the present invention is shown.

[0084] S101, Obtain space-air-ground data;

[0085] S102, Split the space-air-ground data to obtain space data, air data and ground data;

[0086] S103, Respectively extract the feature values in the space data, air data and ground data, and perform fusion processing to obtain fusion feature values;

[0087] S104, Perform residual analysis on the fusion feature values and the feature values in the space data, air data and ground data respectively to obtain a set of fusion feature difference values;

[0088] S105, If the fusion feature differences in the set of fusion feature difference values are all less than or equal to a preset first feature difference threshold, the corresponding fusion feature value is normal;

[0089] S106, Generate fusion data corresponding to the space-air-ground data according to the fusion feature values;

[0090] S107, Send the space-air-ground data and the corresponding fusion data to a preset management end for display.

[0091] According to the embodiment of the present invention, the space-air-ground data are data in three different dimensions, and the space-air-ground data are composed of the original data of satellites (space), unmanned aerial vehicles (air), and ground sensors (ground). The feature values of the space-air-ground data are determined by the corresponding data attributes. For example, when monitoring the flood situation in a place, the corresponding feature value is the rainfall; set the fusion feature difference as , and its formula is , where f represents an eigenvalue among null data, sky data, and ground data, and F represents a fused eigenvalue. When f represents the eigenvalue of null data, a fused feature difference corresponding to the null data is obtained; when f represents the eigenvalue of sky data, a fused feature difference corresponding to the sky data is obtained; when f represents the eigenvalue of ground data, a fused feature difference corresponding to the ground data is obtained. After obtaining the fused eigenvalue, the fused eigenvalue is implanted into a preset standard format to obtain the fused data corresponding to the null-sky-ground data, thereby realizing the unified management of different formats of null-sky-ground data. When there is a fused feature difference in the fused feature difference set that is greater than a preset first feature difference threshold, it indicates that there are obvious differences in the null-sky-ground data at the current time node, triggering a prompt message and sending the prompt message to a preset management terminal for prompting.

[0092] According to an embodiment of the present invention, after extracting the eigenvalues of the null data, sky data, and ground data, it further includes:

[0093] Obtain the time series corresponding to the null-sky-ground data, and construct a null data eigenvalue set, a sky data eigenvalue set, and a ground data eigenvalue set with the corresponding eigenvalues;

[0094] Set the null data eigenvalue set as ;

[0095] Set the sky data eigenvalue set as ;

[0096] Set the ground data eigenvalue set as ;

[0097] Randomly extract three data eigenvalues at the same time node;

[0098] Perform a difference calculation on any two of the three data eigenvalues at the same time node to obtain a second feature difference;

[0099] If the second feature differences are all less than or equal to a preset second feature difference threshold, perform a fusion process on the corresponding three data eigenvalues;

[0100] If there is a second feature difference greater than the preset second feature difference threshold, calibrate the data eigenvalues in the three data eigenvalue sets to obtain the calibrated data eigenvalues;

[0101] Perform a fusion process according to the calibrated data eigenvalues.

[0102] It should be noted that when the second feature differences at the same time node are all less than or equal to the preset second feature difference threshold, it indicates that the time errors of the data in the corresponding different dimensions are all within a reasonable range; when there are second feature differences greater than the preset second feature difference threshold, it indicates that the time errors of the data in the corresponding different dimensions cause obvious errors to the data eigenvalues in different dimensions.

[0103] According to an embodiment of the present invention, the step of calibrating the data feature values in the three data feature sets specifically includes:

[0104] When only one second feature difference is greater than a preset second feature difference threshold, based on the data feature value other than the two data feature values corresponding to the second feature difference, determine the corresponding time node t;

[0105] When there are two second feature differences greater than the preset second feature difference threshold, then based on the common data feature value among the two data feature values corresponding to the two second feature differences, determine the corresponding time node t;

[0106] When there are three second feature differences greater than the preset second feature difference threshold, then arbitrarily extract one data feature value as a reference to determine the corresponding time node t;

[0107] According to the time node t, find the corresponding time range in the other two data feature sets, and according to the corresponding time range, determine the time interval ratio k;

[0108] According to the corresponding time range, find the two data feature values of the data feature set in the corresponding time range, and set them respectively according to the time sequence as and ;

[0109] The calibrated data feature value .

[0110] It should be noted that the time node is divided into multiple time intervals according to a preset time reference value. For example, if the preset time reference value is 1 second, then there is seconds. For example, for the time node , then corresponding to . For example, taking the null data feature as a reference and the sky data feature as the data feature to be calibrated, then , and the corresponding and are respectively and , and the corresponding calibrated sky data feature value .

[0111] According to an embodiment of the present invention, it further includes:

[0112] When the data feature values in the data feature set are not calibrated, based on the same time node, multiply the data feature value by the corresponding dimension weight coefficient, accumulate the products, and obtain the fusion feature value corresponding to the time node;

[0113] When calibrating the data feature values in the data feature set, the data feature value of the benchmark is set as the benchmark data feature value, and the corresponding time node is set as the benchmark time node;

[0114] Multiply the benchmark data feature value by the corresponding dimension weight coefficient to obtain the data feature weight value of the corresponding dimension;

[0115] Multiply the calibrated data feature values of other dimensions corresponding to the benchmark data feature value by the corresponding other dimension weight coefficients to obtain the data feature weight values of other dimensions;

[0116] Accumulate the data feature weight value corresponding to the benchmark data feature value and the data feature weight values of other dimensions to obtain the fusion feature value of the benchmark time node.

[0117] According to the embodiments of the present invention, the step of obtaining the dimension weight coefficient further includes:

[0118] Obtain the historical space-air-ground dataset of the preset first time period;

[0119] Extract the historical data feature values and historical fusion feature values corresponding to the historical space-air-ground data at any time node within the preset first time period;

[0120] Calculate the difference between the historical data feature values of different dimensions and the historical fusion feature values respectively to obtain a plurality of fourth feature differences;

[0121] If all the fourth feature differences are less than or equal to the preset fourth feature difference threshold, set the dimension weight coefficients to one-third;

[0122] If there is at least one fourth feature difference less than or equal to the preset fourth feature difference threshold and there is at least one fourth feature difference greater than the preset fourth feature difference threshold, identify the fourth feature differences greater than the preset fourth feature difference threshold to obtain the identified fourth feature differences; set the dimension weight coefficient adjustment value to , and its formula is , where represents the fourth feature difference, Represents the historical fusion eigenvalue; when a fourth feature difference is less than or equal to the preset fourth feature difference threshold, there are two fourth feature differences greater than the preset fourth feature difference threshold, that is, there are two dimension weight coefficient adjustment values, indicating that the dimension weight coefficient corresponding to the fourth feature difference is one-third minus the dimension weight coefficient adjustment value, and the other dimension weight coefficient is one-third plus two dimension weight coefficient adjustment values; when two fourth feature differences are less than or equal to the preset fourth feature difference threshold, there is one fourth feature difference greater than the preset fourth feature difference threshold, that is, there is one dimension weight coefficient adjustment value, indicating that the dimension weight coefficient corresponding to the fourth feature difference is one-third minus the dimension weight coefficient adjustment value, and the other two dimension weight coefficients are one-third plus one-half of the dimension weight coefficient adjustment value;

[0123] If all fourth feature differences are greater than the preset fourth feature difference threshold, the historical space-air-ground data of the corresponding time node will be deleted, and the historical space-air-ground data of other time nodes will be re-extracted until all fourth feature differences are less than or equal to the preset fourth feature difference threshold or all the historical space-air-ground data in the historical space-air-ground dataset are deleted.

[0124] It should be noted that after all the historical space-air-ground data in the historical space-air-ground dataset are deleted, if all the fourth feature differences are still greater than the preset fourth feature difference threshold, the corresponding three dimension weight coefficients will be set to one-third.

[0125] According to an embodiment of the present invention, it further includes:

[0126] Set the data eigenvalue with a reference as the reference data eigenvalue;

[0127] Calculate the difference between the calibrated data eigenvalue and the reference data eigenvalue to obtain the third feature difference;

[0128] If there are two third feature differences greater than or equal to the preset third feature difference threshold, delete the common data eigenvalues corresponding to the two third feature differences, and then calculate the average value of the remaining two data eigenvalues to obtain the fusion eigenvalue;

[0129] When there are three third feature differences greater than or equal to the preset third feature difference threshold, a data warning message corresponding to the time node will be triggered;

[0130] If there is one third feature difference greater than or equal to the preset third feature difference threshold, or all third feature differences are less than or equal to the preset third feature difference threshold, the currently calibrated data eigenvalue and the reference data eigenvalue are normal.

[0131] According to an embodiment of the present invention, when there is a third feature difference greater than or equal to a preset third feature difference threshold, or all third feature differences are less than or equal to the preset third feature difference threshold, a fusion feature value is constructed by multiplying three data feature values by corresponding dimension weight coefficients respectively. When there are two third feature differences greater than or equal to the preset third feature difference threshold, a fusion feature value is constructed by multiplying two data feature values by corresponding dimension weight coefficients. When three third feature differences are greater than or equal to the preset third feature difference threshold, it indicates that there are errors in the data of the current three dimensions. A warning message of the data at the current time node is used to remind a preset management end, so as to verify and detect the dimension data of the common data feature value of the two third feature differences greater than or equal to the preset third feature difference threshold.

[0132] According to an embodiment of the present invention, it further includes:

[0133] Any two of the empty data, sky data, and ground data are grouped and compared and analyzed to obtain a similarity value set;

[0134] The similarity values in the similarity value set are averaged to obtain a combined similarity value of the empty-sky-ground data;

[0135] It is judged whether the combined similarity value of the empty-sky-ground data is greater than or equal to a preset first similarity threshold. If so, the current empty-sky-ground data is normal; if not, the current empty-sky-ground data is abnormal and an abnormal prompt message is triggered;

[0136] The abnormal prompt message is sent to a preset management end for display.

[0137] It should be noted that the correlation between the corresponding three-dimensional data is determined through the associated similarity value. When the combined similarity value of the empty-sky-ground data is less than the preset first similarity threshold, the corresponding empty-sky-ground data is marked as abnormal, and the abnormal empty-sky-ground data is no longer subjected to fusion processing, thereby improving the effectiveness of data processing.

[0138] Figure 2 The block diagram of a big data intelligent processing system based on the integration of space, sky and ground of the present invention is shown.

[0139] As Figure 2 shown, a second aspect of the present invention provides a big data intelligent processing system 2 based on the integration of space, sky and ground, including a memory 21 and a processor 22. A program of a big data intelligent processing method based on the integration of space, sky and ground is stored in the memory. When the program of the big data intelligent processing method based on the integration of space, sky and ground is executed by the processor, the following steps are implemented:

[0140] Obtain empty-sky-ground data;

[0141] Split the space-air-ground data to obtain space data, air data, and ground data;

[0142] Extract the eigenvalue from the space data, air data, and ground data respectively, and perform fusion processing to obtain the fusion eigenvalue;

[0143] Perform residual analysis on the fusion eigenvalue and the eigenvalues in the space data, air data, and ground data respectively to obtain the fusion feature difference set;

[0144] If the fusion feature differences in the fusion feature difference set are all less than or equal to the preset first feature difference threshold, the corresponding fusion eigenvalue is normal;

[0145] Generate the fusion data corresponding to the space-air-ground data according to the fusion eigenvalue;

[0146] Send the space-air-ground data and the corresponding fusion data to the preset management terminal for display.

[0147] In this solution, after extracting the eigenvalues from the space data, air data, and ground data, it further includes:

[0148] Obtain the time series corresponding to the space-air-ground data, and construct the space data eigenvalue set, air data eigenvalue set, and ground data eigenvalue set with the corresponding eigenvalues;

[0149] Set the space data eigenvalue set as ;

[0150] Set the air data eigenvalue set as ;

[0151] Set the ground data eigenvalue set as ;

[0152] Randomly extract three data eigenvalues at the same time node;

[0153] Calculate the difference between any two of the three data eigenvalues at the same time node to obtain the second feature difference;

[0154] If the second feature differences are all less than or equal to the preset second feature difference threshold, perform fusion processing on the corresponding three data eigenvalues;

[0155] If there is a second feature difference greater than the preset second feature difference threshold, calibrate the data eigenvalues in the three data eigenvalue sets to obtain the calibrated data eigenvalues;

[0156] Perform fusion processing according to the calibrated data eigenvalues.

[0157] In this solution, the step of calibrating the data eigenvalues in the three data eigenvalue sets specifically includes:

[0158] When there is only one second feature difference greater than the preset second feature difference threshold, based on the data feature value other than the two data feature values corresponding to the second feature difference, determine the corresponding time node t;

[0159] When there are two second feature differences greater than the preset second feature difference threshold, then based on the common data feature value among the two data feature values corresponding to the two second feature differences, determine the corresponding time node t;

[0160] When there are three second feature differences greater than the preset second feature difference threshold, then arbitrarily extract one data feature value as the basis to determine the corresponding time node t;

[0161] According to the time node t, find the corresponding time range in the other two data feature value sets, and according to the corresponding time range, determine the time interval ratio k;

[0162] According to the corresponding time range, find the two data feature values of the data feature value set in the corresponding time range, and set them in chronological order as and ;

[0163] The calibrated data feature value .

[0164] A big data intelligent processing method and system based on space-air-ground integration disclosed by the present invention fuse space-air-ground data to improve the accuracy of the collected data.

[0165] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed with each other can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical, or other forms.

[0166] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units; they can be located in one place or distributed to multiple network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0167] In addition, each functional unit in the embodiments of the present invention may all be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware, or in the form of a combination of hardware and software functional units.

[0168] Those of ordinary skill in the art can understand that all or part of the steps for implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps including the above method embodiments; and the foregoing storage medium includes: removable storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks and other various media that can store program codes.

[0169] Alternatively, if the above-mentioned integrated units of the present invention are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the embodiments of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the embodiments of the present invention. And the foregoing storage medium includes: removable storage devices, ROM, RAM, magnetic disks, or optical disks and other various media that can store program codes.

Claims

1. A big data intelligent processing method based on air-ground integration, characterized in that: include: Get air, space and ground data; Split the air-space-ground-land data into air data, sky data and ground data; The characteristic values ​​of the air data, sky data and ground data are extracted respectively, and fused to obtain the fused characteristic values; Perform residual analysis on the fused feature values ​​and the feature values ​​in the air data, sky data and ground data to obtain a fused feature difference set; If the fused feature difference values ​​in the fused feature difference value set are all less than or equal to the preset first feature difference threshold, the corresponding fused feature value is normal; Generate fusion data corresponding to the air-space-ground data according to the fusion feature value; The air-space-ground data and the corresponding fused data are sent to the preset management terminal for display.

2. The big data intelligent processing method based on air-ground integration according to claim 1 is characterized in that: After extracting the characteristic values ​​from the air data, sky data and ground data, the method further includes: Obtain the time series corresponding to the space, sky, and ground data, and construct the space data feature value set, sky data feature value set, and ground data feature value set with the corresponding feature values; Set the null data feature value set to ; Set the daily data feature value set to ; Set the ground data feature value set to ; Randomly extract three data feature values ​​at the same time node; Calculate the difference between any two of the three data feature values ​​at the same time node to obtain a second feature difference; If the second characteristic difference values ​​are all less than or equal to the preset second characteristic difference threshold, the corresponding three data characteristic values ​​are fused; If there is a second characteristic difference value greater than a preset second characteristic difference threshold, calibrating the data characteristic values ​​in the three data characteristic sets to obtain the calibrated data characteristic values; Fusion processing is performed based on the calibrated data feature values.

3. The big data intelligent processing method based on air-ground integration according to claim 2 is characterized in that: The step of calibrating the data feature values ​​in the three data feature sets specifically includes: When only one second characteristic difference value exists that is greater than the preset second characteristic difference threshold, the corresponding time node t is determined based on the data characteristic values ​​other than the two data characteristic values ​​corresponding to the second characteristic difference value; When there are two second feature difference values ​​greater than the preset second feature difference threshold, the corresponding time node t is determined based on the common data feature value of the two data feature values ​​corresponding to the two second feature difference values; When there are three second characteristic difference values ​​greater than the preset second characteristic difference threshold, a data characteristic value is randomly extracted as a reference to determine the corresponding time node t; According to the time node t, find the corresponding time range in the corresponding two data feature value sets, and determine the time interval ratio k according to the corresponding time range; According to the corresponding time range, find the two data feature values ​​of the data feature value set in the corresponding time range, and set them as and ; Data characteristic values ​​after calibration .

4. The big data intelligent processing method based on air-ground integration according to claim 3 is characterized in that: Also includes: When the data feature values ​​in the data feature set are not calibrated, based on the same time node, the data feature values ​​are multiplied by the corresponding dimension weight coefficient, the product is accumulated, and the fusion feature value of the corresponding time node is obtained; When there is a data feature value in the data feature set for calibration, the reference data feature value is set as the reference data feature value, and the corresponding time node is set as the reference time node; Multiply the benchmark data feature value by the corresponding dimension weight coefficient to obtain the data feature weight value of the corresponding dimension; Multiply the calibrated data feature values ​​of other dimensions corresponding to the benchmark data feature values ​​by the corresponding weight coefficients of other dimensions to obtain the data feature weight values ​​of other dimensions; The data feature weight value corresponding to the benchmark data feature value and the data feature weight values ​​of other dimensions are accumulated to obtain the fused feature value of the benchmark time node.

5. The big data intelligent processing method based on air-ground integration according to claim 4 is characterized in that: The step of obtaining the dimension weight coefficient also includes: Obtain a historical space-air-ground data set for a preset first time period; Extract historical data feature values ​​and historical fusion feature values ​​corresponding to historical air-space-ground-land data at any time node within a preset first time period; Perform difference calculations on the historical data feature values ​​of different dimensions and the historical fusion feature values ​​respectively to obtain multiple fourth feature difference values; If the fourth characteristic difference values ​​are all less than or equal to the preset fourth characteristic difference threshold, the dimension weight coefficients are all set to one third; If there is at least one fourth characteristic difference value less than or equal to the preset fourth characteristic difference threshold, and there is at least one fourth characteristic difference value greater than the preset fourth characteristic difference threshold, the fourth characteristic difference value greater than the preset fourth characteristic difference threshold is identified to obtain the identified fourth characteristic difference value; the dimension weight coefficient adjustment value is set to , whose formula is ,in represents the fourth characteristic difference, Represents the historical fusion feature value; when a fourth feature difference value is less than or equal to the preset fourth feature difference threshold, there are two fourth feature difference values ​​that are greater than the preset fourth feature difference threshold, that is, there are two dimensional weight coefficient adjustment values, indicating that the dimensional weight coefficient corresponding to the fourth feature difference value is one-third minus the dimensional weight coefficient adjustment value, and the other dimensional weight coefficient is one-third plus the two dimensional weight coefficient adjustment values; when two fourth feature difference values ​​are less than or equal to the preset fourth feature difference threshold, there is a fourth feature difference value that is greater than the preset fourth feature difference threshold, that is, there is a dimensional weight coefficient adjustment value, indicating that the dimensional weight coefficient corresponding to the fourth feature difference value is one-third minus the dimensional weight coefficient adjustment value, and the other two dimensional weight coefficients are one-third plus one-half of the dimensional weight coefficient adjustment value; If the fourth characteristic difference values ​​are all greater than the preset fourth characteristic difference threshold, the historical space-space-ground data of the corresponding time node will be deleted, and the historical space-space-ground data of other time nodes will be re-extracted until the fourth characteristic difference values ​​are all less than or equal to the preset fourth characteristic difference threshold or all the historical space-space-ground data in the historical space-space-ground data set are deleted.

6. The big data intelligent processing method based on air-ground integration according to claim 1 is characterized in that: Also includes: The data characteristic value based on the benchmark is set as the benchmark data characteristic value; Calculate the difference between the calibrated data characteristic value and the reference data characteristic value to obtain a third characteristic difference value; If there are two third feature difference values ​​that are greater than or equal to the preset third feature difference threshold, the common data feature value corresponding to the two third feature difference values ​​is deleted, and then the average value of the remaining two data feature values ​​is calculated to obtain the fusion feature value; When there are three third characteristic difference values ​​greater than or equal to the preset third characteristic difference threshold, the data warning information of the corresponding time node will be triggered; If there is a third characteristic difference value greater than or equal to the preset third characteristic difference threshold, or the third characteristic difference values ​​are all less than or equal to the preset third characteristic difference threshold, the characteristic value of the data after the current calibration and the characteristic value of the reference data are normal.

7. The big data intelligent processing method based on air-ground integration according to claim 1 is characterized in that: Also includes: Group any two data from the air data, sky data, and ground data, and perform comparative analysis to obtain a similar value set; The similarity values ​​in the similarity value set are averaged to obtain the joint similarity value of the air-space-ground data; Determine whether the joint similarity value of the air-space-ground-land data is greater than or equal to a preset first similarity threshold, if so, the current air-space-ground-land data is normal; If not, the current air-space-ground data is abnormal and an abnormal prompt message is triggered; Send the abnormal prompt information to the preset management terminal for display.

8. A big data intelligent processing system based on air-ground integration, characterized in that: The invention comprises a memory and a processor, wherein the memory stores a program of a big data intelligent processing method based on the integration of air, space and land, and when the program of the big data intelligent processing method based on the integration of air, space and land is executed by the processor, the following steps are implemented: Get air, space and ground data; Split the air-space-ground-land data into air data, sky data and ground data; The characteristic values ​​of the air data, sky data and ground data are extracted respectively, and fused to obtain the fused characteristic values; Perform residual analysis on the fused feature values ​​and the feature values ​​in the air data, sky data and ground data to obtain a fused feature difference set; If the fused feature difference values ​​in the fused feature difference value set are all less than or equal to the preset first feature difference threshold, the corresponding fused feature value is normal; Generate fusion data corresponding to the air-space-ground data according to the fusion feature value; The air-space-ground data and the corresponding fused data are sent to the preset management terminal for display.

9. The big data intelligent processing system based on air-ground integration according to claim 8 is characterized in that: After extracting the characteristic values ​​from the air data, sky data and ground data, the method further includes: Obtain the time series corresponding to the space, sky, and ground data, and construct the space data feature value set, sky data feature value set, and ground data feature value set with the corresponding feature values; Set the null data feature value set to ; Set the daily data feature value set to ; Set the ground data feature value set to ; Randomly extract three data feature values ​​at the same time node; Calculate the difference between any two of the three data feature values ​​at the same time node to obtain a second feature difference; If the second characteristic difference values ​​are all less than or equal to the preset second characteristic difference threshold, the corresponding three data characteristic values ​​are fused; If there is a second characteristic difference value greater than a preset second characteristic difference threshold, calibrating the data characteristic values ​​in the three data characteristic sets to obtain the calibrated data characteristic values; Fusion processing is performed based on the calibrated data feature values.

10. The big data intelligent processing system based on air-ground integration according to claim 9 is characterized in that: The step of calibrating the data feature values ​​in the three data feature sets specifically includes: When only one second characteristic difference value exists that is greater than the preset second characteristic difference threshold, the corresponding time node t is determined based on the data characteristic values ​​other than the two data characteristic values ​​corresponding to the second characteristic difference value; When there are two second feature difference values ​​greater than the preset second feature difference threshold, the corresponding time node t is determined based on the common data feature value of the two data feature values ​​corresponding to the two second feature difference values; When there are three second characteristic difference values ​​greater than the preset second characteristic difference threshold, a data characteristic value is randomly extracted as a reference to determine the corresponding time node t; According to the time node t, find the corresponding time range in the corresponding two data feature value sets, and determine the time interval ratio k according to the corresponding time range; According to the corresponding time range, find the two data feature values ​​of the data feature value set in the corresponding time range, and set them as and ; Data characteristic values ​​after calibration .

Images