Processing method and system for in-orbit real-time target association of multi-source satellite data and medium

Abstract

The invention provides a multi-source satellite data in-orbit real-time target association processing method and system and a medium, and the method comprises the steps: S1, receiving detection data of a multi-source satellite, and carrying out the classification caching; s2, carrying out coarse screening processing on a multi-load on-satellite processing target result; setting a coarse screening condition associated with the identity of the two load targets as follows: in a time domain, the occurrence moments of two target plots are similar; in the airspace, the working areas of the two loads are overlapped as much as possible; s3, assuming that the fuzzy factor sets are # imgabs0 # and # imgabs1 # as the number of the fuzzy factor sets, calculating the fuzzy distance between every two fuzzy factor sets, and constructing a fuzzy incidence matrix; and S4, judging the fuzzy correlation matrix to complete correlation of the multi-load data. The technical problems of possible matching errors and one-to-many results are solved, and the purpose that a single satellite data source index better meets application requirements is achieved.

Description

Technical Field

[0001] The present invention belongs to the field of on-board real-time processing of satellites, and in particular, relates to a method, system and medium for processing multi-source satellite data in real-time on-orbit target association. Background Art

[0002] During the satellite real-time target association process, matching errors and one-to-many results may occur, which have a significant impact on the accuracy of target tracking and data analysis. The specific description is as follows: Matching error types include: (1) False Match. False match means that the system mistakenly associates two different targets together. This situation is usually caused by the high similarity of target features or misjudgment of the algorithm. For example, when two targets are similar in appearance, size or movement mode, the algorithm may mistakenly think that they are the same target. This false match will lead to inaccurate target tracking path, thus affecting subsequent analysis and decision-making; (2) Missed Match. Missed match means that the system fails to correctly identify and associate the target that should be associated. Significant changes in target features or deficiencies in the algorithm may cause this to occur. For example, if the target changes speed, direction or shape in a short period of time, the algorithm may not be able to update its association relationship in time, resulting in target loss. This missed match will interrupt target tracking and make continuous monitoring and analysis impossible.

[0003] Results One-to-Many Matching, including: (1) Multiple Target Association Multiple target association means that at a point in time, the system mistakenly associates one target with multiple possible targets. This is usually due to the high similarity of the features of multiple targets, which makes the system unable to accurately distinguish them. For example, when multiple targets appear in the same area at the same time and their movement trajectories are similar, the system may think that they are all different manifestations of the same target. This situation will lead to data redundancy and analysis confusion, increasing the complexity of processing; (2) Target Overlap Target overlap means that multiple targets overlap in the image, making it difficult for the system to accurately distinguish and associate each target. This situation usually occurs when the target movement trajectories are close or the positions overlap. For example, when multiple targets gather or intersect in the same area, the system may not be able to accurately distinguish the specific location and movement path of each target. This target overlap will affect the accurate tracking and identification of the target, increasing the risk of misjudgment and missed judgment.

[0004] The occurrence of the above matching errors and one-to-many results poses a challenge to the accuracy and reliability of satellite real-time target association. These problems not only affect the continuity and accuracy of target tracking, but may also lead to deviations in data analysis and decision-making.

[0005] Specifically, since multiple loads observe the same area and generate processing results, and due to the different characteristics of multi-source data, such as the low classification accuracy of load 1 and the low positioning accuracy of load 2, it is necessary to perform target association on the results of multiple loads and fuse the results of the same target in a way of taking advantage of each other's strengths and making up for each other's weaknesses, so that the fusion results have both high positioning accuracy and classification accuracy.

[0006] At present, there are few technical solutions for the description or report of on-board application when the single satellite data source indicator does not meet the application requirements. Existing technologies at home and abroad rarely involve this. Summary of the invention

[0007] In view of the defects in the prior art, the purpose of the present invention is to provide a method, system and medium for processing multi-source satellite data in real time on-orbit target association.

[0008] A method for processing multi-source satellite data on-orbit real-time target association according to the present invention comprises: Step S1: receiving detection data from multiple satellite sources and caching them by categories; Step S2: taking a rough screening process on the multi-payload on-board processing target results; Step S3: Assume that the fuzzy factor set is , is the number of fuzzy factor sets, calculate the pairwise fuzzy distances, and construct the fuzzy correlation matrix; Step S4: judge the fuzzy correlation matrix to complete the correlation of multiple load data.

[0009] Preferably, step S2 comprises: Step S2.1: Set the rough screening conditions for the identity association of the two load targets as follows: in the time domain, the two target traces appear at similar times; in the spatial domain, the working areas of the two loads should overlap as much as possible; Step S2.2: Rough screening in the time domain, if the target appears At this moment, only Only the targets that appear within the time period can be associated with target 1. The initial value is preset on the ground and can be modified by annotation; Step S2.3: Rough screening in the airspace requires that the working areas of the two loads meet the following requirements:

[0010] in, is the longitude and latitude coordinates of the payload data target with a larger field of view, It is the overlapped area of ​​the payload work, a square area organized by the processing results, and the vertices of the square are composed of the maximum longitude and latitude values ​​processed on the satellite; In order to roughly screen the error tolerance area, the initial value is preset on the ground and can be changed by annotation.

[0011] Preferably, step S3 comprises: calculating the position factor, the type factor and the number factor by using the position, the target type and the target number information obtained by the two load detections, that is, , calculate the pairwise fuzzy distances and construct the fuzzy association matrix.

[0012] Preferably, step S3 comprises: Step S3.1: Define the position factor as: The position factor is calculated using the set of all target coordinate measurement values ​​generated after coarse correlation of multiple loads in a time period close to time T;

[0013] in, is the position factor, is the weight parameter of the fuzzy factor associated with the target location feature; , , , are the coordinate measurements of load 1 and load 2 respectively; Step S3.2: Define the type factor as: According to DS evidence theory, the target type information is fuzzified. Assume The recognition framework for target type determination, and yes The target types of multiple loads in a time period close to each other are basically assigned, and the combined fuzzy factor is expressed as:

[0014] in, is the type factor; is the weight parameter of the fuzzy factor associated with the target type feature; Step S3.3: Define the numbering factor as: and yes The target numbers of two load data in a time period close to each other are expressed as Represents two load data to Number since time and The probability that the target meets the identity determination requirements;

[0015] in, is the numbering factor; is the weight parameter of the fuzzy factor associated with the target number feature, is the number distance, which is 1 if it is the same as the number at the previous moment, otherwise it is 0; Step S3.4: Assumptions It is the recognition framework for target type judgment. The fuzzy factor is comprehensively evaluated by weighted average method. The calculation formula of comprehensive similarity is:

[0016] express Load 1 in a similar time period Targets and payloads 2 The comprehensive similarity of the targets; is the number of factors; for , , .

[0017] Preferably, the weight values ​​of the factors described in step 3 are initially preset on the ground and can be modified online.

[0018] Preferably, the multi-load data association in step S4 is performed by the formula:

[0019] in, is the result of multi-load data correlation. Each element in the matrix ~ In step S3-4 The formula is obtained. For example, the first target of load 1 and the second target of load 2 are filled in .

[0020] Preferably, in step S4, the fuzzy incidence matrix is ​​judged by using the Hungarian algorithm.

[0021] A multi-source satellite data on-orbit real-time target association processing system, comprising: Module M1: Receives detection data from multiple satellite sources and caches them in categories; Module M2: Coarse screening of multi-payload onboard processing target results; Module M3: Assume that the fuzzy factor set is , is the number of fuzzy factor sets, calculate the pairwise fuzzy distances, and construct the fuzzy correlation matrix; Module M4: judge the fuzzy correlation matrix and complete the correlation of multiple load data.

[0022] Preferably, module M2 comprises: Module M2.1: Set the rough screening conditions for the identity association of two load targets as follows: in the time domain, the two target traces appear at similar times; in the spatial domain, the working areas of the two loads should overlap as much as possible; Module M2.2: Coarse screening in the time domain, if the target appears At this moment, only Only the targets that appear within the time period can be associated with target 1. The initial value is preset on the ground and can be modified by annotation; Module M2.3: Rough screening in airspace requires that the working areas of the two loads meet the following requirements:

[0023] in, is the longitude and latitude coordinates of the payload data target with a larger field of view, It is the overlapped area of ​​the payload work, a square area organized by the processing results, and the vertices of the square are composed of the maximum longitude and latitude values ​​processed on the satellite; To roughly screen the error tolerance area, the initial value is preset on the ground and can be modified by annotation; Module M3 includes: calculating the position factor, type factor and number factor by using the position, target type and target number information detected by the two payloads, that is, , calculate the pairwise fuzzy distances and construct the fuzzy correlation matrix; Module M3, including: Module M3.1: Define the position factor as: The position factor is calculated using the set of all target coordinate measurement values ​​generated after coarse correlation of multiple loads in a time period close to time T;

[0024] in, is the position factor, is the weight parameter of the fuzzy factor associated with the target location feature; , , , are the coordinate measurements of load 1 and load 2 respectively; Module M3.2: Define the type factor as: According to DS evidence theory, the target type information is fuzzified. Assume The recognition framework for target type determination, and yes The target types of multiple loads in a time period close to each other are basically assigned, and the combined fuzzy factor is expressed as:

[0025] in, is the type factor; is the weight parameter of the fuzzy factor associated with the target type feature; Module M3.3: Define the numbering factors as: and yes The target numbers of two load data in a time period close to each other are expressed as Represents two load data to Number since time and The probability that the target meets the identity determination requirements;

[0026] in, is the numbering factor; is the weight parameter of the fuzzy factor associated with the target number feature, is the number distance, which is 1 if it is the same as the number at the previous moment, otherwise it is 0; Module M3.4: Assumptions It is the recognition framework for target type judgment. The fuzzy factor is comprehensively evaluated by weighted average method. The calculation formula of comprehensive similarity is:

[0027] express Load 1 in a similar time period Targets and payloads 2 The comprehensive similarity of the targets; is the number of factors; for , , ; The weight values ​​of each factor described in step 3 are initially preset on the ground and can be modified by annotation; The multi-load data association described in module M4 is as follows:

[0028] in, is the result of multi-load data correlation. Each element in the matrix ~ Through module M3-4 The formula is obtained. For example, the first target of load 1 and the second target of load 2 are filled in ; Module M4 judges the fuzzy correlation matrix through the Hungarian algorithm.

[0029] A computer-readable storage medium storing a computer program, characterized in that when the computer program is executed by a processor, the steps of a method for processing multi-source satellite data in orbit in real time for target association are implemented.

[0030] Compared with the prior art, the present invention has the following beneficial effects: 1. Effectively solve the possible matching errors in the satellite real-time target association process; 2. Effectively solve the one-to-many problem in the satellite real-time target association process; 3. In the process of satellite real-time target association, a single satellite data source indicator is realized to better meet application requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings: Figure 1 This is a technical flow chart of the on-orbit real-time target association processing method for multi-source satellite data; Figure 2 Schematic diagram of the on-orbit real-time target association processing method for multi-source satellite data. DETAILED DESCRIPTION

[0032] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0033] In order to solve the problem that the single satellite data source indicator does not meet the application requirements, the purpose of the present invention is to provide a method and system for processing multi-source satellite data in-orbit real-time target association.

[0034] To achieve the above purpose, Figure 1 As shown, it is achieved through the following technical solutions: Step S1: receiving detection data from multiple satellite sources and caching them by categories; Step S2: In order to greatly reduce the computational complexity of the target identity association algorithm, a rough screening process is performed on the multi-payload on-board target processing results, such as Figure 2 shown.

[0035] Step S2.1: Set the rough screening conditions for the identity association of the two load targets as follows: in the time domain, the two target points appear at similar times; in the spatial domain, the working areas of the two loads should overlap as much as possible.

[0036] Step S2.2: Rough screening in the time domain, if the target appears At this moment, only Only the targets that appear within the time period can be associated with target 1. The initial value is preset on the ground and can be modified by annotation; Step S2.3: Rough screening in the airspace requires that the working areas of the two loads meet the following requirements:

[0037] in, is the longitude and latitude coordinates of the payload data target with a larger field of view, It is the overlapped area of ​​the payload work, a square area organized by the processing results, and the vertices of the square are composed of the maximum longitude and latitude values ​​processed on the satellite; To roughly screen the error tolerance area, the initial value is preset on the ground and can be modified by annotation; Step S3: Assume that the fuzzy factor set is , is the number of fuzzy factor sets; In this embodiment, the position, target type and target number information obtained by two payload detections are mainly used to calculate the position factor, type factor and number factor, that is, , calculate the pairwise fuzzy distances and construct the fuzzy correlation matrix; Position factor, the set of all target coordinate measurement values ​​generated by coarse correlation of multiple loads in a time period close to time T Step S3.1: Define the position factor as: The position factor is calculated using the set of all target coordinate measurement values ​​generated after coarse correlation of multiple loads in a time period close to time T.

[0038]

[0039] in, is the position factor, is the weight parameter of the fuzzy factor associated with the target location feature; , , , are the coordinate measurements of load 1 and load 2 respectively; Step S3.2: Define the type factor as: According to DS evidence theory, the target type information is fuzzified. Assume The recognition framework for target type determination, and yes The target types of multiple loads in a time period close to each other are basically assigned, and the combined fuzzy factor is expressed as:

[0040] in, is the type factor; It is the weight parameter of the fuzzy factor associated with the target type feature.

[0041] Step S3.3: Define the numbering factor as: and yes The target numbers of two load data in a time period close to each other are expressed as Represents two load data to Number since time and The probability that the target meets the identity determination requirements.

[0042]

[0043] in, is the numbering factor; is the weight parameter of the fuzzy factor associated with the target number feature, is the number distance (1 if the number is the same as the previous moment, otherwise 0).

[0044] Step S3.4: Assumptions It is the recognition framework for target type judgment. The fuzzy factor is comprehensively evaluated by weighted average method. The calculation formula of comprehensive similarity is:

[0045] express Load 1 in a similar time period Targets and payloads 2 The comprehensive similarity of the targets; is the number of factors; for , , .

[0046] For multiple choice, the weight values ​​of each factor in step 3 are initially preset and can be modified by annotation.

[0047] Step S4: The fuzzy correlation matrix is ​​judged by the Hungarian algorithm to complete the correlation of multiple load data.

[0048]

[0049] in, is the result of multi-load data correlation. Each element in the matrix ~ In step S3-4 The formula is obtained. For example, the first target of load 1 and the second target of load 2 are filled in .

[0050] The present invention also provides a processing system for on-orbit real-time target association of multi-source satellite data, which can be implemented by executing the process steps of the processing method for on-orbit real-time target association of multi-source satellite data, that is, those skilled in the art can understand the processing method for on-orbit real-time target association of multi-source satellite data as a preferred implementation of the processing system for on-orbit real-time target association of multi-source satellite data. It includes: Module M1: Receives detection data from multiple satellite sources and caches them in categories; Module M2: Coarse screening of multi-payload onboard processing target results; Module M3: Assume that the fuzzy factor set is , is the number of fuzzy factor sets, calculate the pairwise fuzzy distances, and construct the fuzzy correlation matrix; Module M4: judge the fuzzy correlation matrix and complete the correlation of multiple load data.

[0051] Specifically: Module M2, including: Module M2.1: Set the rough screening conditions for the identity association of two load targets as follows: in the time domain, the two target traces appear at similar times; in the spatial domain, the working areas of the two loads should overlap as much as possible; Module M2.2: Coarse screening in the time domain, if the target appears At this moment, only Only the targets that appear within the time period can be associated with target 1. The initial value is preset on the ground and can be modified by annotation; Module M2.3: Rough screening in airspace requires that the working areas of the two loads meet the following requirements:

[0052] in, is the longitude and latitude coordinates of the payload data target with a larger field of view, It is the overlapped area of ​​the payload work, a square area organized by the processing results, and the vertices of the square are composed of the maximum longitude and latitude values ​​processed on the satellite; To roughly screen the error tolerance area, the initial value is preset on the ground and can be modified by annotation; Module M3 includes: calculating the position factor, type factor and number factor by using the position, target type and target number information detected by the two payloads, that is, , calculate the pairwise fuzzy distances and construct the fuzzy correlation matrix; Module M3, including: Module M3.1: Define the position factor as: The position factor is calculated using the set of all target coordinate measurement values ​​generated after coarse correlation of multiple loads in a time period close to time T;

[0053] in, is the position factor, is the weight parameter of the fuzzy factor associated with the target location feature; , , , are the coordinate measurements of load 1 and load 2 respectively; Module M3.2: Define the type factor as: According to DS evidence theory, the target type information is fuzzified. Assume The recognition framework for target type determination, and yes The target types of multiple loads in a time period close to each other are basically assigned, and the combined fuzzy factor is expressed as:

[0054] in, is the type factor; is the weight parameter of the fuzzy factor associated with the target type feature; Module M3.3: Define the numbering factors as: and yes The target numbers of two load data in a time period close to each other are expressed as Represents two load data to Number since time and The probability that the target meets the identity determination requirements;

[0055] in, is the numbering factor; is the weight parameter of the fuzzy factor associated with the target number feature, is the number distance, which is 1 if it is the same as the number at the previous moment, otherwise it is 0; Module M3.4: Assumptions It is the recognition framework for target type judgment. The fuzzy factor is comprehensively evaluated by weighted average method. The calculation formula of comprehensive similarity is:

[0056] express Load 1 in a similar time period Targets and payloads 2 The comprehensive similarity of the targets; is the number of factors; for , , ; The weight values ​​of each factor in step 3 are initially preset on the ground and can be modified by annotation; Module M4 multi-load data association, the formula is:

[0057] in, is the result of multi-load data correlation. Each element in the matrix ~ Through module M3-4 The formula is obtained. For example, the first target of load 1 and the second target of load 2 are filled in ; Module M4 judges the fuzzy correlation matrix through the Hungarian algorithm.

[0058] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of a method for processing multi-source satellite data in-orbit real-time target association.

[0059] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0060] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A method for processing multi-source satellite data in-orbit real-time target association, characterized in that: include: Step S1: receiving detection data from multiple satellite sources and caching them by categories; Step S2: taking a rough screening process on the multi-payload on-board processing target results; Step S3: Assume that the fuzzy factor set is , is the number of fuzzy factor sets, calculate the pairwise fuzzy distances, and construct the fuzzy correlation matrix; Step S4: judge the fuzzy correlation matrix to complete the correlation of multiple load data.

2. The method for processing multi-source satellite data on-orbit real-time target association according to claim 1, characterized in that: The step S2 comprises: Step S2.1: Set the rough screening conditions for the identity association of the two load targets as follows: in the time domain, the two target traces appear at similar times; in the spatial domain, the working areas of the two loads should overlap as much as possible; Step S2.2: Rough screening in the time domain, if the target appears At this moment, only Only the targets that appear within the time period can be associated with target 1. The initial value is preset on the ground and can be modified by annotation; Step S2.3: Rough screening in the airspace requires that the working areas of the two loads meet the following requirements: in, is the longitude and latitude coordinates of the payload data target with a larger field of view, It is the overlapped area of ​​the payload work, a square area organized by the processing results, and the vertices of the square are composed of the maximum longitude and latitude values ​​processed on the satellite; In order to roughly screen the error tolerance area, the initial value is preset on the ground and can be changed by annotation.

3. The method for processing multi-source satellite data on-orbit real-time target association according to claim 1, characterized in that: The step S3 comprises: calculating the position factor, the type factor and the number factor by using the position, the target type and the target number information obtained by the two load detections, that is, , calculate the pairwise fuzzy distances and construct the fuzzy association matrix.

4. The method for processing multi-source satellite data on-orbit real-time target association according to claim 1, characterized in that: The step S3 comprises: Step S3.1: Define the position factor as: The position factor is calculated using the set of all target coordinate measurement values ​​generated after coarse correlation of multiple loads in a time period close to time T; in, is the position factor, is the weight parameter of the fuzzy factor associated with the target location feature; , , , are the coordinate measurements of load 1 and load 2 respectively; Step S3.2: Define the type factor as: According to DS evidence theory, the target type information is fuzzified. Assume The recognition framework for target type determination, and yes The target types of multiple loads in a time period close to each other are basically assigned, and the combined fuzzy factor is expressed as: in, is the type factor; is the weight parameter of the fuzzy factor associated with the target type feature; Step S3.3: Define the numbering factor as: and yes The target numbers of two load data in a time period close to each other are expressed as Represents two load data to Number since time and The probability that the target meets the identity determination requirements; in, is the numbering factor; is the weight parameter of the fuzzy factor associated with the target number feature, is the number distance, which is 1 if it is the same as the number at the previous moment, otherwise it is 0; Step S3.4: Assumptions It is the recognition framework for target type judgment. The fuzzy factor is comprehensively evaluated by weighted average method. The calculation formula of comprehensive similarity is: express Load 1 in a similar time period Targets and payloads 2 The comprehensive similarity of the targets; is the number of factors; for , , .

5. The method for processing multi-source satellite data on-orbit real-time target association according to claim 1, characterized in that: The initial value of the weight of each factor described in step 3 is preset on the ground and can be modified by annotation.

6. The method for processing multi-source satellite data on-orbit real-time target association according to claim 1, characterized in that: The multi-load data association in step S4 is as follows: in, is the result of multi-load data correlation. Each element in the matrix ~ In step S3-4 The formula is obtained. For example, the first target of load 1 and the second target of load 2 are filled in .

7. The method for processing multi-source satellite data on-orbit real-time target association according to claim 1, characterized in that: In step S4, the fuzzy incidence matrix is ​​judged by the Hungarian algorithm.

8. A multi-source satellite data on-orbit real-time target association processing system, characterized in that: include: Module M1: Receives detection data from multiple satellite sources and caches them in categories; Module M2: Coarse screening of multi-payload onboard processing target results; Module M3: Assume that the fuzzy factor set is , is the number of fuzzy factor sets, calculate the pairwise fuzzy distances, and construct the fuzzy correlation matrix; Module M4: judge the fuzzy correlation matrix and complete the correlation of multiple load data.

9. The system for processing multi-source satellite data on-orbit real-time target association according to claim 8, characterized in that: Module M2, including: Module M2.1: Set the rough screening conditions for the identity association of two load targets as follows: in the time domain, the two target traces appear at similar times; in the spatial domain, the working areas of the two loads should overlap as much as possible; Module M2.2: Coarse screening in the time domain, if the target appears At this moment, only Only the targets that appear within the time period can be associated with target 1. The initial value is preset on the ground and can be modified by annotation; Module M2.3: Rough screening in airspace requires that the working areas of the two loads meet the following requirements: in, is the longitude and latitude coordinates of the payload data target with a larger field of view, It is the overlapped area of ​​the payload work, a square area organized by the processing results, and the vertices of the square are composed of the maximum longitude and latitude values ​​processed on the satellite; To roughly screen the error tolerance area, the initial value is preset on the ground and can be modified by annotation; Module M3 includes: calculating the position factor, type factor and number factor by using the position, target type and target number information detected by the two payloads, that is, , calculate the pairwise fuzzy distances and construct the fuzzy correlation matrix; Module M3, including: Module M3.1: Define the position factor as: The position factor is calculated using the set of all target coordinate measurement values ​​generated after coarse correlation of multiple loads in a time period close to time T; in, is the position factor, is the weight parameter of the fuzzy factor associated with the target location feature; , , , are the coordinate measurements of load 1 and load 2 respectively; Module M3.2: Define the type factor as: According to DS evidence theory, the target type information is fuzzified. Assume The recognition framework for target type determination, and yes The target types of multiple loads in a time period close to each other are basically assigned, and the combined fuzzy factor is expressed as: in, is the type factor; is the weight parameter of the fuzzy factor associated with the target type feature; Module M3.3: Define the numbering factors as: and yes The target numbers of two load data in a time period close to each other are expressed as Represents two load data to Number since time and The probability that the target meets the identity determination requirements; in, is the numbering factor; is the weight parameter of the fuzzy factor associated with the target number feature, is the number distance, which is 1 if it is the same as the number at the previous moment, otherwise it is 0; Module M3.4: Assumptions It is the recognition framework for target type judgment. The fuzzy factor is comprehensively evaluated by weighted average method. The calculation formula of comprehensive similarity is: express Load 1 in a similar time period Targets and payloads 2 The comprehensive similarity of the targets; is the number of factors; for , , ; The weight values ​​of each factor described in step 3 are initially preset on the ground and can be modified by annotation; The multi-load data association described in module M4 is as follows: in, is the result of multi-load data correlation. Each element in the matrix ~ Through module M3-4 The formula is obtained. For example, the first target of load 1 and the second target of load 2 are filled in ; Module M4 judges the fuzzy correlation matrix through the Hungarian algorithm.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for processing on-orbit real-time target association of multi-source satellite data according to any one of claims 1 to 7 are implemented.

Images