Civil aerial target identification method and system based on space-time mapping knowledge domain

Abstract

The invention belongs to the technical field of aerial target recognition, and particularly discloses a civil aerial target recognition method and system based on a space-time knowledge graph, and the method comprises the steps: constructing a civil aerial target recognition space-time knowledge graph; based on the space-time knowledge graph and the multi-dimensional situation data of the plurality of sample targets, obtaining a multi-dimensional situation identification evidence matrix and an identification result vector corresponding to the plurality of sample targets, and taking the multi-dimensional situation identification evidence matrix and the identification result vector as a data set; training a target recognition model based on the data set; receiving multi-dimensional situation data of the aerial target, and obtaining a corresponding multi-dimensional situation identification evidence matrix based on the multi-dimensional situation data of the aerial target; and inputting the multi-dimensional situation recognition evidence matrix corresponding to the aerial target into the trained target recognition model to obtain a target recognition result output by the target recognition model.

Description

Technical Field

[0001] The present application belongs to the technical field of aerial target recognition, and more specifically, to a civil aerial target recognition method and system based on spatiotemporal knowledge graph. Background Art

[0002] In air transport routes, a large number of civil flights and small general aviation aircraft fly in the airspace every day to carry out various civil operations. Due to transponder failure, signal interference, and manual shutdown of self-reporting equipment, some civil equipment cannot clearly indicate their own position and identity in the Automatic Dependent Surveillance-Broadcast (ADS-B) system, becoming a good cover environment for illegal aerial targets, making it difficult to determine the target identity; at the same time, a large number of military manned / unmanned aircraft from neighboring countries are active in the airspace, and some targets are similar to civil aircraft in terms of movement ability, flight characteristics, and route selection, making it difficult to distinguish; what's more, using fabricated false ADS-B and other civil signals as a cover identity to carry out military operations makes it difficult to identify the military and civilian attributes of the target. Therefore, the existing air target recognition method has a low recognition accuracy rate for uncooperative civil aerial targets. Summary of the invention

[0003] In view of the defects of the prior art, the purpose of this application is to provide a civil aerial target recognition method and system based on spatiotemporal knowledge graph, aiming to effectively solve the problem of effective identification of civil legal aerial targets in daily duty environment, determine the military and civilian attributes, and improve the recognition accuracy of civil aerial targets.

[0004] To achieve the above objectives, in a first aspect, the present application provides a civil aerial target recognition method based on a spatiotemporal knowledge graph, comprising:

[0005] Construct a spatiotemporal knowledge graph for civil aerial target identification;

[0006] Based on the spatiotemporal knowledge graph and the multidimensional situation data of multiple sample targets, a multidimensional situation recognition evidence matrix and a recognition result vector corresponding to the multiple sample targets are obtained, and the multidimensional situation recognition evidence matrix and the recognition result vector are used as a data set;

[0007] Based on the data set, training a target recognition model;

[0008] Receiving multi-dimensional situation data of an aerial target, and obtaining a corresponding multi-dimensional situation recognition evidence matrix based on the multi-dimensional situation data of the aerial target;

[0009] The multi-dimensional situation recognition evidence matrix corresponding to the aerial target is input into the trained target recognition model to obtain the target recognition result output by the target recognition model.

[0010] This application constructs a spatiotemporal knowledge graph for civil aerial target identification and uses it to construct a multi-dimensional situation identification evidence matrix and the corresponding identification result vectors. It can combine a variety of spatiotemporal knowledge and apply it to aerial target identification to enhance the accuracy of civil aerial target identification. It can be used as a data set to train a target recognition model to enhance the recognition performance of the model and improve the recognition accuracy of civil aerial targets.

[0011] According to the civil aerial target recognition method based on spatiotemporal knowledge graph provided by the present invention, the civil aerial target recognition spatiotemporal knowledge graph includes one or more of the following:

[0012] Knowledge of civil air target identification equipment, air traffic control, airports, airspace, routes, basic flight regulations, and the spatial and temporal laws of civil air targets.

[0013] The present application constructs a knowledge graph based on one or more of the knowledge of civil air target identification equipment, air traffic control knowledge, airport knowledge, airspace knowledge, route and enroute knowledge, basic flight regulations knowledge, and knowledge of the spatiotemporal laws of civil air targets, so as to construct a multi-dimensional situation identification evidence matrix and the corresponding identification result vectors. It can combine a variety of spatiotemporal knowledge and apply it to air target identification to enhance the accuracy of civil air target identification.

[0014] According to the civil aerial target recognition method based on the spatiotemporal knowledge graph provided by the present invention, the multidimensional situation data based on the spatiotemporal knowledge graph and the multiple sample targets are used to obtain the multidimensional situation recognition evidence matrix and the recognition result vector corresponding to the multiple sample targets, including:

[0015] Based on the spatiotemporal knowledge graph, generate multiple single-dimensional knowledge discriminant element vectors;

[0016] Obtain multi-dimensional situation data of multiple sample targets;

[0017] Based on the multidimensional situation data of the multiple sample targets and the multiple single-dimensional knowledge discriminant element vectors, obtaining a multidimensional situation recognition evidence matrix corresponding to the multiple sample targets respectively;

[0018] Based on the multi-dimensional situation recognition evidence matrix, a recognition result vector corresponding to the multi-dimensional situation recognition evidence matrix is ​​established.

[0019] When generating a data set for training a target recognition model, the present application generates multiple single-dimensional knowledge discriminant element vectors based on the knowledge graph, and then fills the multidimensional situation data of multiple samples into the vector to generate a multidimensional situation recognition evidence matrix. The recognition result vector is determined based on the matrix, and the target recognition model is trained with the data set, which can improve the recognition effect of the target recognition model.

[0020] According to the civil aerial target recognition method based on spatiotemporal knowledge graph provided by the present invention, the target recognition model is a deep learning model, and the target recognition model is trained based on the data set, including:

[0021] Randomly use 60% of the data in the data set as a training set to train the deep learning model;

[0022] The remaining 40% of the data in the data set is used as a test set to test the trained deep learning model. If the recognition accuracy is greater than the preset threshold, the training is completed. If the recognition accuracy is less than or equal to the preset threshold, retraining is performed.

[0023] This application divides the data set into a training set and a test set, uses the training set to train the deep learning model, uses the test set to test the deep learning model, and sets a preset threshold to improve the recognition effect of the target recognition model.

[0024] According to the civil aerial target recognition method based on spatiotemporal knowledge graph provided by the present invention, the multidimensional situation data based on the aerial target is used to obtain a corresponding multidimensional situation recognition evidence matrix, including:

[0025] Based on the multi-dimensional situation data of the aerial target and the multiple single-dimensional knowledge discriminant element vectors, a multi-dimensional situation recognition evidence matrix corresponding to the aerial target is obtained.

[0026] According to the civil aerial target recognition method based on spatiotemporal knowledge graph provided by the present invention, the multi-dimensional situation recognition evidence matrix corresponding to the aerial target is input into the trained target recognition model to obtain the target recognition result output by the target recognition model, including:

[0027] Input the multi-dimensional situation recognition evidence matrix corresponding to the aerial target into the trained target recognition model to obtain the recognition result vector output by the target recognition model;

[0028] Each element in the recognition result vector is traversed, and the row identifiers of the elements that are true are output as the target recognition result.

[0029] In the second aspect, the present application provides a civil aerial target recognition system based on spatiotemporal knowledge graph, comprising:

[0030] A construction module for building a spatiotemporal knowledge graph for civil aerial target identification;

[0031] An acquisition module, for obtaining a multidimensional situation recognition evidence matrix and a recognition result vector respectively corresponding to a plurality of sample targets based on the spatiotemporal knowledge graph and the multidimensional situation data of a plurality of sample targets, and using the multidimensional situation recognition evidence matrix and the recognition result vector as a data set;

[0032] A training module, used for training a target recognition model based on the data set;

[0033] A receiving module, used for receiving multi-dimensional situation data of an aerial target, and obtaining a corresponding multi-dimensional situation recognition evidence matrix based on the multi-dimensional situation data of the aerial target;

[0034] The recognition module is used to input the multi-dimensional situation recognition evidence matrix corresponding to the aerial target into the trained target recognition model to obtain the target recognition result output by the target recognition model.

[0035] In a third aspect, the present application provides an electronic device comprising: at least one memory for storing programs; and at least one processor for executing the programs stored in the memory. When the programs stored in the memory are executed, the processor is used to execute the civil aerial target recognition method based on the spatiotemporal knowledge graph described in the first aspect or any possible implementation method of the first aspect.

[0036] In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the civil aerial target recognition method based on the spatiotemporal knowledge graph described in the first aspect or any possible implementation of the first aspect.

[0037] In a fifth aspect, the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the civil aerial target recognition method based on the spatiotemporal knowledge graph described in the first aspect or any possible implementation of the first aspect.

[0038] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0039] In general, the above technical solutions conceived by this application have the following beneficial effects compared with the prior art:

[0040] By constructing a spatiotemporal knowledge graph for civil aerial target identification and using it to construct a multi-dimensional situational identification evidence matrix and the corresponding identification result vectors, a variety of spatiotemporal knowledge can be combined and applied to aerial target identification to enhance the accuracy of civil aerial target identification. Using it as a data set to train a target recognition model can enhance the recognition performance of the model and improve the recognition accuracy of civil aerial targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 This is one of the flow diagrams of the civil aerial target recognition method based on the spatiotemporal knowledge graph provided in the embodiment of the present application;

[0043] Figure 2 This is the second flow chart of the civil aerial target recognition method based on spatiotemporal knowledge graph provided in the embodiment of the present application;

[0044] Figure 3 It is a structural schematic diagram of a civil aerial target recognition system based on a spatiotemporal knowledge graph provided in an embodiment of the present application;

[0045] Figure 4 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0047] The term "and / or" in this article is a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The symbol " / " in this article indicates that the associated objects are in an or relationship, for example, A / B means A or B.

[0048] The terms "first" and "second" in the specification and claims herein are used to distinguish different objects rather than to describe a specific order of the objects. For example, a first response message and a second response message are used to distinguish different response messages rather than to describe a specific order of the response messages.

[0049] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0050] In the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more than two. For example, multiple processing units refer to two or more processing units, etc.; multiple elements refer to two or more elements, etc.

[0051] Next, combine Figure 1-Figure 2 The civil aerial target recognition method based on spatiotemporal knowledge graph provided in the embodiment of the present application is introduced.

[0052] Figure 1 is a flow chart of a civil aerial target recognition method based on a spatiotemporal knowledge graph provided in an embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps:

[0053] Step 100, constructing a spatiotemporal knowledge graph for civil aerial target recognition;

[0054] A knowledge graph is a structured semantic knowledge base that represents entities and various relationships between entities in the form of a graph. This application first obtains the spatiotemporal knowledge related to civil aerial target recognition, and then constructs a spatiotemporal knowledge graph for civil aerial target recognition.

[0055] Optionally, the spatiotemporal knowledge related to civil air target identification may include but is not limited to civil air target identification equipment knowledge, air control knowledge, etc. The relevant knowledge can be obtained through the Internet or related documents, and this application does not limit this.

[0056] Optionally, the present application does not limit the specific method of constructing the spatiotemporal knowledge graph for civil aerial target identification, and it can be constructed through a large language model or manually.

[0057] Optionally, the knowledge graph establishes connections between identified knowledge entity elements through entity-relationship-entity triples, and establishes associations between spatiotemporal data elements through multi-layer spatiotemporal grids.

[0058] Step 110, based on the spatiotemporal knowledge graph and the multidimensional situation data of the multiple sample targets, obtain a multidimensional situation recognition evidence matrix and a recognition result vector corresponding to the multiple sample targets respectively, and use the multidimensional situation recognition evidence matrix and the recognition result vector as a data set;

[0059] After constructing the knowledge graph, a multi-dimensional situation recognition evidence matrix can be constructed based on the connections between the knowledge entity elements and the correlation between the spatiotemporal data elements in the knowledge graph. The matrix contains multiple discriminant factors for determining whether the aerial target is civilian. For different aerial targets, the discriminant factors may be the same or different. Based on multiple discriminant factors, it can be comprehensively determined whether the aerial target is civilian.

[0060] Optionally, the multi-dimensional situation data may be various operational data of the aerial target, such as longitude and latitude position, flight speed, heading, flight altitude, etc.

[0061] The recognition result vector corresponds to the multi-dimensional situation recognition evidence matrix, which contains a variety of possible recognition results, such as civilian targets, military targets, etc., and is used to display the recognition results corresponding to the multi-dimensional situation recognition evidence matrix.

[0062] Based on the spatiotemporal knowledge graph and the multi-dimensional situation data of multiple sample targets, this application obtains the multi-dimensional situation recognition evidence matrix and recognition result vector corresponding to the multiple sample targets respectively, as a data set to prepare for training the target recognition model.

[0063] Step 120, training a target recognition model based on the data set;

[0064] Optionally, the target recognition model can be a deep learning model, which, after being trained with a data set, can establish an association between a multi-dimensional situation recognition evidence matrix and a recognition result vector, that is, input the multi-dimensional situation recognition evidence matrix into it and output the corresponding recognition result vector.

[0065] Step 130, receiving multi-dimensional situation data of the aerial target, and obtaining a corresponding multi-dimensional situation recognition evidence matrix based on the multi-dimensional situation data of the aerial target;

[0066] Optionally, the present application does not limit the method for receiving the multi-dimensional situation data of the aerial target, which may be received by wireless communication or wired reception, etc.

[0067] Step 140, input the multi-dimensional situation recognition evidence matrix corresponding to the aerial target into the trained target recognition model to obtain the target recognition result output by the target recognition model.

[0068] After the model training is completed, the multi-dimensional situation recognition evidence matrix of the aerial target is input into it, the corresponding recognition result vector is output and converted into the target recognition result.

[0069] The civil aerial target recognition method based on the spatiotemporal knowledge graph provided by the present invention can combine a variety of spatiotemporal knowledge and apply them to aerial target recognition to enhance the accuracy of civil aerial target recognition by constructing a spatiotemporal knowledge graph for civil aerial target recognition and using it to construct a multi-dimensional situation recognition evidence matrix and the corresponding recognition result vectors. It can also be used as a data set to train a target recognition model, which can enhance the recognition performance of the model and improve the recognition accuracy of civil aerial targets.

[0070] In some embodiments, the civil aerial target identification spatiotemporal knowledge graph includes one or more of the following:

[0071] Knowledge of civil air target identification equipment, air traffic control, airports, airspace, routes, basic flight regulations, and the spatial and temporal laws of civil air targets.

[0072] Optionally, the knowledge of civil aerial target identification equipment is mainly used to describe the flight performance knowledge of various common types of civil aircraft.

[0073] Optionally, air traffic control knowledge is used to describe air traffic control information such as flight control area divisions and flight control regulations.

[0074] Optionally, airport knowledge is used to describe airport location, take-off and landing capabilities, runways, take-off, approach requirements and other airport take-off and landing information.

[0075] Optionally, airspace knowledge is used to describe the reported divisions and opening and closing conditions of various airspace types.

[0076] Optionally, route and enroute knowledge is used to describe the reviewed flight plan, route, air corridor and other knowledge.

[0077] Optionally, the basic flight regulations knowledge is used to describe my country's basic flight requirements for civil aircraft, including flight interval regulations, altitude layer occupancy regulations, flight speed regulations and other knowledge.

[0078] Optionally, the knowledge of the spatiotemporal regularities of civil aerial targets is mainly used to mine and generate the air trajectory regularities, flight profile regularities, flight speed range, turning radius, turning slope and other spatiotemporal regularity characteristics of confirmed civil aerial targets during daily flight.

[0079] By constructing a knowledge graph based on the above knowledge, a multi-dimensional situation recognition evidence matrix and the corresponding recognition result vectors can be constructed. Various spatiotemporal knowledge can be combined and applied to aerial target recognition to enhance the accuracy of civil aerial target recognition.

[0080] In some embodiments, step 110 specifically includes:

[0081] Step 1101, generating multiple single-dimensional knowledge discriminant element vectors based on the spatiotemporal knowledge graph;

[0082] Step 1102, obtaining multi-dimensional situation data of multiple sample targets;

[0083] Step 1103, based on the multi-dimensional situation data of the multiple sample targets and the multiple single-dimensional knowledge discriminant element vectors, obtain a multi-dimensional situation recognition evidence matrix corresponding to the multiple sample targets respectively;

[0084] Step 1104: Based on the multi-dimensional situation recognition evidence matrix, establish a recognition result vector corresponding to the multi-dimensional situation recognition evidence matrix.

[0085] Optionally, a single-dimensional knowledge discriminant factor vector can be generated based on the knowledge in the spatiotemporal knowledge graph. For example, through the knowledge of air routes, single-dimensional knowledge discriminant factor vectors such as "whether the track is consistent with the plan" and "whether the use of the route is compliant" can be generated.

[0086] Optionally, the present application does not limit the method for obtaining multi-dimensional situation data of multiple sample targets.

[0087] In one embodiment of the present application, multidimensional situation data of n sample targets are obtained, and a multidimensional situation recognition evidence matrix as shown in Table 1 is constructed in combination with the spatiotemporal knowledge graph, which includes multiple single-dimensional knowledge discriminant element vectors and the discrimination results of the multidimensional situation data of the sample targets in each knowledge discriminant element vector.

[0088] Table 1 Multidimensional situation recognition evidence matrix

[0089]

[0090]

[0091] In one embodiment of the present application, an expert establishes a recognition result vector as shown in Table 2 based on a multi-dimensional situation recognition evidence matrix, and each column of the recognition result ensures that only one element is true and the rest are false.

[0092] Table 2 Recognition result vector

[0093] Identification results Sample Target 1 …… Sample target n Civilian Targets yes no no Military Targets no yes no Suspected civilian target no no yes Suspected military target no no no Unrecognizable no no no

[0094] For each sample target, the above-mentioned multi-dimensional situation recognition evidence matrix and recognition result vector are constructed to serve as a data set for training the target recognition model.

[0095] In some embodiments, the target recognition model is a deep learning model, and step 120 specifically includes:

[0096] Step 1201, randomly select 60% of the data in the data set as a training set to train the deep learning model;

[0097] Step 1202, use the remaining 40% of the data in the data set as a test set to test the trained deep learning model. If the recognition accuracy is greater than the preset threshold, the training is completed. If the recognition accuracy is less than or equal to the preset threshold, retraining is performed.

[0098] Optionally, the present application does not limit the specific structure of the deep learning model used, and a deep neural network model (Deep Neural Networks, DNN) etc. may be selected.

[0099] 60% of the data is randomly selected from the data set as the training sample set, and the remaining 40% of the data is used as the test sample set. The deep learning model is trained using the training sample set to obtain the model parameters. The deep learning model can select commonly used deep learning models, such as a fully connected neural network model, etc., to screen models that meet the convergence conditions.

[0100] The screened model is tested using the test sample set to determine whether the recognition accuracy is greater than the preset threshold. If not, retraining is performed until the model meets the preset threshold.

[0101] In one embodiment of the present application, the preset threshold is set to 90%, which can be set according to specific application requirements, but it is recommended not to exceed 95%.

[0102] In some embodiments, step 130 specifically includes:

[0103] Step 1301, based on the multi-dimensional situation data of the aerial target and a plurality of single-dimensional knowledge discriminant element vectors, a multi-dimensional situation recognition evidence matrix corresponding to the aerial target is obtained.

[0104] After receiving the multi-dimensional situation data of aerial targets, the target data elements are parsed, and each knowledge sub-graph is traversed using the spatiotemporal knowledge graph to fill the multi-dimensional situation recognition evidence matrix.

[0105] In some embodiments, step 140 specifically includes:

[0106] Step 1401, inputting the multi-dimensional situation recognition evidence matrix corresponding to the aerial target into the trained target recognition model to obtain a recognition result vector output by the target recognition model;

[0107] Step 1402, traverse each element in the recognition result vector, and output the row identifier of the element that is true as the target recognition result.

[0108] The multi-dimensional situation recognition evidence matrix is ​​input into the trained target recognition model, and the target recognition result vector is obtained by using the model. Then, the target result recognition vector is converted into the recognition result output. Specifically, each element in the recognition result vector is traversed, and the row identifier whose element is true is output as the target recognition result.

[0109] Figure 2 This is the second flow chart of the civil aerial target recognition method based on spatiotemporal knowledge graph provided in the embodiment of the present application, such as Figure 2 As shown, in one embodiment of the present application, the civil aerial target recognition method based on spatiotemporal knowledge graph provided by the present application includes the following steps:

[0110] 1. Construct a spatiotemporal knowledge graph for civil air target identification, including knowledge of civil air target identification equipment, air traffic control, airports, airspace, air routes, basic flight regulations, and spatiotemporal laws of civil air targets;

[0111] 2. Construct a multi-dimensional situation recognition evidence matrix and the corresponding recognition result vector as training samples based on the spatiotemporal knowledge graph of civil aerial target recognition;

[0112] 3. When there are sufficient training samples, randomly select 60% of the data from the training sample data set as the training sample set, and the remaining 40% of the data as the test sample set. Use the training sample set to train the deep learning model, and select the model with an accuracy greater than 90% as the target recognition model;

[0113] 4. Receive and analyze multi-dimensional situation data of aerial targets, and build a multi-dimensional situation identification evidence matrix;

[0114] 5. Input the multi-dimensional situation recognition evidence matrix into the trained target recognition model to obtain its output recognition result vector;

[0115] 6. Convert the recognition result vector into recognition result output.

[0116] Combine the following Figure 3 The civil aerial target recognition system based on the spatiotemporal knowledge graph provided by the present invention is described. The civil aerial target recognition system based on the spatiotemporal knowledge graph described below and the civil aerial target recognition method based on the spatiotemporal knowledge graph described above can be referenced to each other.

[0117] Figure 3 is a schematic diagram of the structure of a civil aerial target recognition system based on a spatiotemporal knowledge graph provided in an embodiment of the present application, such as Figure 3 As shown, the system includes a construction module 310, an acquisition module 320, a training module 330, a receiving module 340 and a recognition module 350, wherein:

[0118] A construction module 310 is used to construct a spatiotemporal knowledge graph for civil aerial target identification;

[0119] An acquisition module 320 is used to construct multiple multi-dimensional situation recognition evidence matrices and corresponding recognition result vectors based on the spatiotemporal knowledge graph, and use the multiple multi-dimensional situation recognition evidence matrices and corresponding recognition result vectors as data sets;

[0120] A training module 330 is used to train a target recognition model based on a data set;

[0121] A receiving module 340 is used to receive multi-dimensional situation data of an aerial target and obtain a corresponding multi-dimensional situation recognition evidence matrix based on the multi-dimensional situation data of the aerial target;

[0122] The recognition module 350 is used to input the multi-dimensional situation recognition evidence matrix corresponding to the aerial target into the trained target recognition model to obtain the target recognition result output by the target recognition model.

[0123] Optionally, in some embodiments, the civil aerial target identification system based on the spatiotemporal knowledge graph also includes a log management module for recording log information such as the graph construction process, training data generation process, model training process, data reception process, target identification process, and issuing alarms for abnormal situations.

[0124] It should be understood that the above-mentioned device is used to execute the method in the above-mentioned embodiment. The implementation principle and technical effect of the corresponding program module in the device are similar to those described in the above-mentioned method. The working process of the device can refer to the corresponding process in the above-mentioned method, which will not be repeated here.

[0125] Based on the method in the above embodiment, Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4 As shown, an embodiment of the present application provides an electronic device, which may include: a processor (processor) 410, a communication interface (CommunicationsInterface) 420, a memory (memory) 430 and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other through the communication bus 440. The processor 410 can call the logic instructions in the memory 430 to execute the civil aerial target recognition method based on the spatiotemporal knowledge graph in the above embodiment.

[0126] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the civil aerial target recognition method based on the spatiotemporal knowledge graph described in each embodiment of the present application.

[0127] Based on the method in the above embodiment, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the civil aerial target identification method based on the spatiotemporal knowledge graph in the above embodiment.

[0128] Based on the method in the above embodiment, an embodiment of the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the civil aerial target recognition method based on the spatiotemporal knowledge graph in the above embodiment.

[0129] It is understandable that the processor in the embodiment of the present application can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor can be a microprocessor or any conventional processor.

[0130] The method steps in the embodiments of the present application can be implemented by hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, mobile hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC.

[0131] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions may be transmitted from a website site, a computer, a server or a data center to another website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.

[0132] It should be understood that the various numerical numbers involved in the embodiments of the present application are only used for the convenience of description and are not used to limit the scope of the embodiments of the present application.

[0133] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A civil aerial target recognition method based on spatiotemporal knowledge graph, characterized in that: include: Construct a spatiotemporal knowledge graph for civil aerial target identification; Based on the spatiotemporal knowledge graph and the multidimensional situation data of the multiple sample targets, a multidimensional situation recognition evidence matrix and a recognition result vector corresponding to the multiple sample targets are obtained, and the multidimensional situation recognition evidence matrix and the recognition result vector are used as a data set; Based on the data set, training a target recognition model; Receiving multi-dimensional situation data of an aerial target, and obtaining a corresponding multi-dimensional situation recognition evidence matrix based on the multi-dimensional situation data of the aerial target; The multi-dimensional situation recognition evidence matrix corresponding to the aerial target is input into the trained target recognition model to obtain the target recognition result output by the target recognition model.

2. The civil aerial target recognition method based on spatiotemporal knowledge graph according to claim 1 is characterized in that: The civil aerial target identification spatiotemporal knowledge graph includes one or more of the following: Knowledge of civil air target identification equipment, air traffic control, airports, airspace, routes, basic flight regulations, and the spatial and temporal laws of civil air targets.

3. The civil aerial target recognition method based on spatiotemporal knowledge graph according to claim 1 or 2, characterized in that: The method of obtaining a multidimensional situation recognition evidence matrix and a recognition result vector respectively corresponding to the multiple sample targets based on the spatiotemporal knowledge graph and the multidimensional situation data of the multiple sample targets includes: Based on the spatiotemporal knowledge graph, generate multiple single-dimensional knowledge discriminant element vectors; Obtain multi-dimensional situation data of multiple sample targets; Based on the multidimensional situation data of the multiple sample targets and the multiple single-dimensional knowledge discriminant element vectors, obtaining a multidimensional situation recognition evidence matrix corresponding to the multiple sample targets respectively; Based on the multi-dimensional situation recognition evidence matrix, a recognition result vector corresponding to the multi-dimensional situation recognition evidence matrix is ​​established.

4. The civil aerial target recognition method based on spatiotemporal knowledge graph according to claim 1 is characterized in that: The target recognition model is a deep learning model, and the target recognition model is trained based on the data set, including: Randomly use 60% of the data in the data set as a training set to train the deep learning model; The remaining 40% of the data in the data set is used as a test set to test the trained deep learning model. If the recognition accuracy is greater than the preset threshold, the training is completed. If the recognition accuracy is less than or equal to the preset threshold, retraining is performed.

5. The civil aerial target recognition method based on spatiotemporal knowledge graph according to claim 3 is characterized in that: The method of obtaining a corresponding multi-dimensional situation recognition evidence matrix based on the multi-dimensional situation data of the aerial target includes: Based on the multi-dimensional situation data of the aerial target and the multiple single-dimensional knowledge discriminant element vectors, a multi-dimensional situation recognition evidence matrix corresponding to the aerial target is obtained.

6. The civil aerial target recognition method based on spatiotemporal knowledge graph according to claim 1 is characterized in that: The step of inputting the multi-dimensional situation recognition evidence matrix corresponding to the aerial target into the trained target recognition model to obtain the target recognition result output by the target recognition model comprises: Input the multi-dimensional situation recognition evidence matrix corresponding to the aerial target into the trained target recognition model to obtain the recognition result vector output by the target recognition model; Each element in the recognition result vector is traversed, and the row identifiers of the elements that are true are output as the target recognition result.

7. A civil aerial target recognition system based on spatiotemporal knowledge graph, characterized in that: include: A construction module for building a spatiotemporal knowledge graph for civil aerial target identification; An acquisition module, for obtaining a multidimensional situation recognition evidence matrix and a recognition result vector respectively corresponding to a plurality of sample targets based on the spatiotemporal knowledge graph and the multidimensional situation data of a plurality of sample targets, and using the multidimensional situation recognition evidence matrix and the recognition result vector as a data set; A training module, used for training a target recognition model based on the data set; A receiving module, used for receiving multi-dimensional situation data of an aerial target, and obtaining a corresponding multi-dimensional situation recognition evidence matrix based on the multi-dimensional situation data of the aerial target; The recognition module is used to input the multi-dimensional situation recognition evidence matrix corresponding to the aerial target into the trained target recognition model to obtain the target recognition result output by the target recognition model.

8. An electronic device, characterized in that: include: at least one memory for storing a computer program; At least one processor is used to execute the program stored in the memory. When the program stored in the memory is executed, the processor is used to execute the civil aerial target recognition method based on the spatiotemporal knowledge graph as described in any one of claims 1-6.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program runs on a processor, the processor executes the civil aerial target recognition method based on spatiotemporal knowledge graph as described in any one of claims 1 to 6.

10. A computer program product, characterized in that When the computer program product runs on a processor, the processor executes the civil aerial target recognition method based on spatiotemporal knowledge graph as described in any one of claims 1 to 6.

Images