Unmanned aerial vehicle identification method and device, electronic equipment and storage medium
By collecting radio frequency signals to generate time-frequency diagrams for drone identification, the problem of limited visual monitoring in the prior art is solved, and the efficiency and accuracy of drone identification are improved.
Patent Information
- Application Number
- CN202410724110.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-05
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing drone recognition technology mainly relies on visual monitoring, resulting in low recognition efficiency and low accuracy in the case of low environmental visibility or poor camera acquisition performance.
By collecting radio frequency signals to be identified, signal discovery and drone type recognition are performed, and drone type recognition is used to identify drone type, avoiding the limitations of visual methods.
It improves the efficiency and accuracy of drone recognition, and solves the problems of low recognition efficiency and low accuracy under the limitations of visual conditions.
Smart Images

Figure CN119068404A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of drones, and in particular to a drone identification method, device, electronic device and storage medium. Background Art
[0002] With the development of society, UAV technology has developed rapidly, making UAVs widely used in various fields. Research and development of anti-UAV technology for detection and identification of UAVs is crucial.
[0003] Among the currently available relevant technical means, visual monitoring is mainly used for drone identification, such as monitoring and identifying drones in the airspace through optical recognition or camera acquisition. Although this method can play a role in identifying drones, it is limited by visual conditions. For example, when the environmental visibility and camera acquisition performance are low, there will be problems with low efficiency and accuracy in drone identification. Summary of the invention
[0004] In order to solve the above technical problems, the present application provides a drone identification method, device, electronic device and storage medium.
[0005] In a first aspect, the present application provides a method for identifying a drone, comprising:
[0006] Generate a time-frequency diagram to be identified based on the collected radio frequency signal;
[0007] Perform signal discovery based on the time-frequency graph to be identified to obtain a target time-frequency graph corresponding to the signal discovery information;
[0008] The type of UAV is identified based on the target time-frequency diagram to obtain a UAV type identification result.
[0009] Optionally, generating a time-frequency graph to be identified based on the collected radio frequency signal includes:
[0010] Get the acquisition frequency band parameters;
[0011] Determining a target frequency band corresponding to the acquisition frequency band parameter;
[0012] collecting the radio frequency signal of the target frequency band;
[0013] The time-frequency diagram to be identified is generated according to the radio frequency signal.
[0014] Optionally, generating the time-frequency graph to be identified according to the radio frequency signal includes:
[0015] Performing short-time Fourier transform on the radio frequency signal to obtain signal time-frequency information;
[0016] The signal time-frequency information is used to perform energy color mapping to generate the time-frequency diagram to be identified.
[0017] Optionally, performing signal discovery according to the time-frequency graph to be identified to obtain a target time-frequency graph corresponding to the signal discovery information includes:
[0018] Performing signal discovery on a preset target location area in the to-be-identified time-frequency graph to obtain signal discovery information;
[0019] The signal discovery information is used in combination with the time-frequency diagram to be identified to generate the target time-frequency diagram.
[0020] Optionally, the performing signal discovery on a preset target location area in the to-be-identified time-frequency graph to obtain signal discovery information includes:
[0021] Determining a preset target location area in the time-frequency graph to be identified;
[0022] Based on the preset target location area, detecting signal occurrence frequency location information and signal occurrence time location information;
[0023] The signal occurrence frequency position information and the signal occurrence time position information are determined as the signal discovery information.
[0024] Optionally, the using the signal discovery information in combination with the time-frequency graph to be identified to generate the target time-frequency graph includes:
[0025] generating a signal discovery tag according to the signal discovery information;
[0026] Combining the signal discovery label with the time-frequency graph to be identified to obtain a label time-frequency graph;
[0027] Data enhancement is performed on the label time-frequency graph to generate the target time-frequency graph.
[0028] Optionally, the performing drone type identification based on the target time-frequency graph to obtain a drone type identification result includes:
[0029] Determine energy change information corresponding to the target signal in the target time-frequency diagram;
[0030] Determine the preset drone category information matching the energy change information as the target drone category information;
[0031] Generate a drone type recognition result based on the target drone category information.
[0032] In a second aspect, the present application provides a drone identification device, comprising:
[0033] A generation module, used to generate a time-frequency diagram to be identified based on the collected radio frequency signal;
[0034] A signal discovery module, used to perform signal discovery based on the time-frequency graph to be identified, and obtain a target time-frequency graph corresponding to the signal discovery information;
[0035] The identification module is used to identify the type of drone based on the target time-frequency diagram to obtain a drone type identification result.
[0036] In a third aspect, an electronic device is provided, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;
[0037] Memory, used to store computer programs;
[0038] The processor is used to implement the drone identification method described in any one of the first aspects when executing the program stored in the memory.
[0039] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the drone identification method as described in any one of the first aspects is implemented.
[0040] The embodiment of the present application generates a time-frequency graph to be identified based on the collected radio frequency signal, and performs signal discovery based on the time-frequency graph to be identified, to obtain a target time-frequency graph corresponding to the signal discovery information, so as to perform drone type identification based on the target time-frequency graph to obtain a drone type identification result. As a result, there is no need to use a visual method to detect the drone, but instead the time-frequency graph is used to identify the drone type, which solves the problem that the existing drone identification method is limited by visual conditions, and has low drone identification efficiency and low accuracy, thereby improving drone identification efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.
[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0043] Figure 1 A schematic diagram of a flow chart of a drone identification method provided in an embodiment of the present application;
[0044] Figure 2A schematic diagram of an application scenario of a drone identification method provided in an embodiment of the present application;
[0045] Figure 3 A schematic diagram of another application scenario of a drone identification method provided in an embodiment of the present application;
[0046] Figure 4 A schematic diagram of another application scenario of a drone identification method provided in an embodiment of the present application;
[0047] Figure 5 A schematic diagram of the structure of a drone identification device provided in an embodiment of the present application;
[0048] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0049] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0050] Among the existing traditional technical means, drone identification is mainly carried out by visual detection. For example, visual data of the airspace environment, such as image data, video data, etc., is collected through cameras, optical recognition devices, etc., and then the visual collected data is analyzed to determine whether there is a drone. Since the visual detection method collects visual data of the airspace environment, it will be directly affected by the visual conditions in the airspace environment. For example, in cases where the visual conditions are poor, such as cloudy days, rainy days, haze, trees, etc., the visual data collected by cameras, optical recognition devices and other equipment will have low clarity and low recognition, which will result in low recognition efficiency and low accuracy in the subsequent identification of whether a drone is present in the visual data.
[0051] In order to solve the problems of low recognition efficiency and low accuracy in the existing technical solutions, the embodiments of the present application provide a drone recognition method, device, electronic device and storage medium, by generating a time-frequency graph to be recognized based on the collected radio frequency signal, and performing signal discovery based on the time-frequency graph to be recognized, obtaining a target time-frequency graph corresponding to the signal discovery information, and performing drone type identification based on the target time-frequency graph to obtain a drone type identification result, so that there is no need to use a visual method to detect the drone, but instead the drone type is identified by using the time-frequency graph, which solves the problem that the existing drone recognition method is limited by visual conditions and has low drone recognition efficiency and low accuracy, and improves the drone recognition efficiency and accuracy.
[0052] Figure 1 A flowchart of a drone identification method provided in an embodiment of the present application.
[0053] like Figure 1 As shown, the present application discloses an embodiment, providing a method for identifying a drone, comprising the following steps:
[0054] Step S110: generating a time-frequency diagram to be identified according to the collected radio frequency signal.
[0055] Specifically, the radio frequency signal in this embodiment represents the radio waves in the target area of collection, and the specific collection method may be to collect through spectrum detection equipment, radio frequency collection equipment and other equipment capable of collecting radio waves; thereby, a time-frequency diagram to be identified may be generated based on the collected radio frequency signal, and the time-frequency diagram to be identified may represent an image showing the signal frequency of the radio frequency signal changing over time.
[0056] Step S120: performing signal discovery based on the time-frequency graph to be identified, and obtaining a target time-frequency graph corresponding to the signal discovery information.
[0057] Specifically, after generating the time-frequency graph to be identified, signal discovery can be performed based on the time-frequency graph to be identified to obtain a target time-frequency graph corresponding to the signal discovery information, wherein signal discovery means performing drone signal detection in the time-frequency graph to be identified, and obtaining signal discovery information after performing signal discovery on the time-frequency graph to be identified. The signal discovery information can represent the information of the drone signal contained in the time-frequency graph to be identified, such as the occurrence time information, frequency range and other information of the drone signal. The target time-frequency graph can represent the time-frequency graph generated based on the signal discovery information. The specific generation method can be to generate the target time-frequency graph after adding the signal discovery information to the time-frequency graph to be identified, or to regenerate the target time-frequency graph based on the signal discovery information. Of course, other methods can also be used, and this embodiment does not make specific limitations.
[0058] In one example, signal discovery based on the time-frequency graph to be identified may be signal detection of a preset target position area in the time-frequency graph to be identified, or detection of whether a preset target signal exists in the time-frequency graph to be identified. Of course, other signal discovery methods may also be used, which are not specifically limited in this embodiment. In the case where signal discovery is signal detection of a preset target position area in the time-frequency graph to be identified, the signal discovery information may be information of a drone signal detected in the preset target position area. In the case where signal discovery is detection of whether a preset target signal exists in the time-frequency graph to be identified, the signal discovery information is target information. Of course, other methods may also be used, which are not specifically limited in this embodiment.
[0059] Step S130: Perform drone type identification based on the target time-frequency graph to obtain a drone type identification result.
[0060] Specifically, after obtaining the target time-frequency graph, the drone type can be identified based on the target time-frequency graph to obtain a drone type identification result, wherein the drone type identification represents the identification process of determining the type of drone to which the signal contained in the target time-frequency graph belongs. The drone type identification result can be used to represent the result obtained after performing drone type identification on the target time-frequency graph, such as not belonging to the drone type, belonging to drone type A, belonging to drone type B, etc. Of course, the drone type identification result can also include information such as the identification time and the frequency band in which the signal appears, which is not specifically limited in this embodiment.
[0061] In an optional embodiment of the present application, step S110 generates a time-frequency diagram to be identified based on the collected RF signal, and may specifically include the following sub-steps: obtaining collection frequency band parameters; determining a target frequency band corresponding to the collection frequency band parameters; collecting RF signals of the target frequency band; and generating a time-frequency diagram to be identified based on the RF signal.
[0062] In the process of collecting radio frequency signals in this embodiment, collection frequency band parameters can be obtained, and the collection frequency band parameters represent the parameters of the frequency band that needs to be collected. The collection frequency band parameters can be determined based on the user's input operation, and can also be determined based on pre-configured parameter data. Of course, other methods can also be used to obtain them, and this embodiment does not specifically limit this; thereby, the target frequency band corresponding to the collection frequency band parameters can be determined, and the target frequency band can represent the frequency band where the radio frequency signal needs to be collected. The target collection frequency band can be one frequency band or multiple frequency bands. At this time, the radio frequency signal in the target frequency band can be collected, and a time-frequency diagram to be identified is generated based on the radio frequency signal, thereby playing a role in targeted collection of radio frequency signals in the target frequency band.
[0063] In an optional embodiment of the present application, a time-frequency diagram to be identified is generated based on a radio frequency signal, which may specifically include the following sub-steps: performing a short-time Fourier transform on the radio frequency signal to obtain signal time-frequency information; performing energy color mapping using the signal time-frequency information to generate a time-frequency diagram to be identified.
[0064] In this embodiment, after obtaining the RF signal of the target frequency band, the RF signal can be short-time Fourier transformed to obtain the signal time-frequency information, and the signal time-frequency information can be energy-colored mapped to generate a time-frequency diagram to be identified, thereby converting the RF signal into a time-frequency diagram to be identified, so that subsequent signal processing can be performed based on the time-frequency diagram to be identified, thereby improving the efficiency and accuracy of RF signal processing; it should be noted that in this embodiment, when the target frequency band contains multiple frequency bands, the RF signals corresponding to each target frequency band can be short-time Fourier transformed and energy-colored mapped respectively to generate time-frequency diagrams to be identified corresponding to each RF signal.
[0065] In one example, when the acquired acquisition frequency band parameters are 985MHZ, 2.4GHZ, and 5.8GHZ, the target frequency bands corresponding to 985MHZ, 2.4GHZ, and 5.8GHZ can be determined as target frequency band 985MHZ, target frequency band 2.4GHZ, and target frequency band 5.8GHZ. Then, the RF signals of the three target frequency bands, namely, 985MHZ RF signal, 2.4GHZ RF signal, and 5.8GHZ RF signal, are collected, so that the three RF signals can be respectively analyzed. Short-time Fourier transform and energy color mapping are used to generate a time-frequency diagram to be identified corresponding to a 985MHZ radio frequency signal, a time-frequency diagram to be identified corresponding to a 2.4GHZ radio frequency signal, and a time-frequency diagram to be identified corresponding to a 5.8GHZ radio frequency signal, that is, three time-frequency diagrams to be identified corresponding to radio frequency signals of three target frequency bands are generated; of course, the target frequency band can also be a frequency band range, such as 2.2GHz-2.5GHz, 5.0GHz—5.4GHz, 5.6GHz—6GHz, etc., and this embodiment does not make specific limitations on this.
[0066] In an optional embodiment of the present application, step S120 performs signal discovery based on the time-frequency graph to be identified to obtain a target time-frequency graph corresponding to the signal discovery information, which may specifically include the following sub-steps:
[0067] Perform signal discovery on a preset target location area in the time-frequency diagram to be identified, and obtain signal discovery information;
[0068] The signal discovery information is used in combination with the time-frequency graph to be identified to generate the target time-frequency graph.
[0069] In this embodiment, after determining the time-frequency graph to be identified, the method for performing signal discovery on the time-frequency graph to be identified may be to perform signal discovery on a preset target position area in the time-frequency graph to be identified to obtain signal discovery information, wherein the preset target position area indicates the position area corresponding to the drone signal in the pre-configured time-frequency graph. Since the radio frequency signal of the drone is usually in a fixed area in the time-frequency graph, the drone signal can be accurately discovered in the time-frequency graph to be identified through the pre-configured preset target position area to obtain signal discovery information, and the signal discovery information indicates the signal identified from the preset target position area of the time-frequency graph to be identified; then, the signal discovery information can be used to generate a target time-frequency graph in combination with the time-frequency graph to be identified, wherein the target time-frequency graph indicates the time-frequency graph corresponding to the signal discovery information. The specific combination method of the signal discovery information and the time-frequency graph to be identified may be to add the signal discovery information to the time-frequency graph to be identified to generate the target time-frequency graph, or to extract the time-frequency area corresponding to the signal discovery information in the time-frequency graph to be identified and regenerate the target time-frequency graph. Of course, other methods may also be used, and this embodiment does not specifically limit this.
[0070] In an optional embodiment of the present application, signal discovery is performed on a preset target position area in a time-frequency graph to be identified to obtain signal discovery information, which may specifically include the following sub-steps: determining a preset target position area in the time-frequency graph to be identified; based on the preset target position area, detecting signal occurrence frequency position information and signal occurrence time position information; and determining the signal occurrence frequency position information and the signal occurrence time position information as signal discovery information.
[0071] In this embodiment, in the process of signal discovery for the preset target position area in the time-frequency graph to be identified, the preset target position area in the time-frequency graph to be identified may be determined. The specific determination method may be to obtain preset position parameters, which may include frequency range parameters, time parameters, etc. The position area in the time-frequency graph to be identified that matches the preset position parameters is determined as the target position area, so that based on the preset target position area, the signal occurrence frequency position information and the signal occurrence time position information can be detected, that is, the occurrence frequency position and the signal occurrence time position of the signal in the preset target position area in the time-frequency graph to be identified are detected, the signal occurrence frequency position information indicates the frequency position of the signal in the preset target position area, the signal occurrence time position information indicates the time position of the signal in the preset target position area, and the signal occurrence frequency position information and the signal occurrence time position information are determined as signal discovery information; thereby accurately identifying the signal discovery information corresponding to the drone signal in the preset target position area.
[0072] In one example, since the signals generated by the drone are divided into frequency hopping signals for flight control communication and fixed-frequency broadband signals for transmitting images, these two signals will have obvious energy differences from noise and other signals. This difference can be mapped to the time-frequency diagram using short-time Fourier transform and energy color mapping. That is, the location area where the drone's frequency hopping signal and fixed-frequency broadband signal appear in the time-frequency diagram can be pre-determined as the preset target location area, so that the signal occurrence frequency location information and signal occurrence time location information can be detected in the preset target location area of the time-frequency diagram to be identified, that is, check whether the drone's frequency hopping signal and fixed-frequency broadband signal appear in the preset target location area. If so, the signal occurrence frequency location information and signal occurrence time location information can be determined as signal discovery information, such as Figure 2 After determining the preset target location area in the time-frequency diagram to be identified, the following is obtained: Figure 3 The preset target location area shown in the figure is Figure 3 The preset target position area shown in the figure is used to detect the signal occurrence frequency position information and the signal occurrence time position information to obtain the signal discovery information, and then the signal discovery information is used to generate a target time-frequency diagram in combination with the time-frequency diagram to be identified. The target time-frequency diagram thus generated represents the time-frequency diagram corresponding to the frequency-hopping signal and the fixed-frequency broadband signal of the drone. Therefore, there is no need to use a visual method to detect the drone, but instead the time-frequency diagram is used to identify the type of drone, which solves the problem that the existing drone identification method is limited by visual conditions, and has low drone identification efficiency and low accuracy, thereby improving the drone identification efficiency and accuracy.
[0073] It should be noted that in this embodiment, a preset target position area in the time-frequency graph to be identified is determined, and based on the preset target position area, the signal occurrence frequency position information and the signal occurrence time position information are detected, and the signal occurrence frequency position information and the signal occurrence time position information are determined as signal discovery information. A neural network model can be used for implementation, such as a pre-trained YOLOV8 network. The YOLOV8 network can play the role of a real-time target detection and image segmentation model. Therefore, the YOLOV8 network is used to perform real-time target detection and image segmentation on the time-frequency graph to be identified, and signal discovery information can be obtained. The training method of the YOLOV8 network can be as follows:
[0074] Step 1: Use spectrum detection equipment to collect RF signals of more than 12 drones with different characteristics in the spectrum, use short-time Fourier transform and energy intensity color mapping algorithm to draw the drone time-frequency diagram, and use labeling tools to create a target detection type data set.
[0075] Step 2: Combine the prepared data set with the configuration file and pre-trained model, and use the SGD algorithm to iteratively update the parameters in the YOLOV8 model. Set the hyperparameters suitable for the signal discovery task before training. After the training, save the results in the form of a model weight file, so that a mapping relationship is formed between the parameters in the model and the possible drone signal features. This mapping relationship that can be learned includes but is not limited to: the difference in the duration of a single signal between drone signals and noise signals, the single-hop bandwidth of drone signals, the relative position of drone frequency-hopping signals in the spectrum, the relative position of drone image transmission signals in the spectrum, and the morphological difference between drone image transmission signals and frequency modulation signals. These mapping relationships will provide a basis for judgment in the prediction stage.
[0076] In an optional embodiment of the present application, the signal discovery information is used in combination with the time-frequency graph to be identified to generate a target time-frequency graph, which may specifically include the following sub-steps: generating a signal discovery label based on the signal discovery information; combining the signal discovery label and the time-frequency graph to be identified to obtain a label time-frequency graph; performing data enhancement on the label time-frequency graph to generate a target time-frequency graph.
[0077] In this embodiment, after obtaining the signal discovery information, a signal discovery label can be generated based on the signal discovery information. The signal discovery label represents a label used to identify the signal discovery information. The signal discovery label and the time-frequency graph to be identified are combined to obtain a labeled time-frequency graph. The signal discovery label can be inserted into the time-frequency graph to be identified, or the signal discovery label can be generally stored as image information of the time-frequency graph to be identified to obtain a labeled time-frequency graph. The labeled time-frequency graph is then data enhanced to generate a target time-frequency graph. In addition, when the target time-frequency graph is multiple time-frequency graphs, the target time-frequency graphs can be stored in the same database.
[0078] In an optional embodiment of the present application, step S130 performs drone type identification based on the target time-frequency diagram to obtain a drone type identification result, which may specifically include the following sub-steps: determining energy change information corresponding to the target signal in the target time-frequency diagram; determining preset drone category information that matches the energy change information as the target drone category information; and generating a drone type identification result based on the target drone category information.
[0079] In this embodiment, after obtaining the target time-frequency diagram, the energy change information corresponding to the target signal in the target time-frequency diagram can be determined, wherein the target signal represents the signal corresponding to the signal discovery information in the target time-frequency diagram, that is, the signal corresponding to the signal discovery label, and the energy change information represents the change information of the target signal in the target time-frequency diagram. Since there are some subtle differences in the protocols for communicating between different drones and their flight control systems, these subtle differences can be mapped to differences in energy changes with time and frequency, further causing different drones to exhibit their own unique energy change laws within a certain frequency range, such as Figure 4 As shown, the energy change information can be matched and judged with the preset UAV category information, and the preset UAV category information that matches the energy change information can be determined as the target UAV category information, that is, the target UAV category information represents the UAV category corresponding to the collected RF signal.
[0080] It should be noted that, in this embodiment, the preset drone category information matching the energy change information is determined as the target drone category information. According to the target drone category information, the drone type recognition result can be generated by using a pre-trained convolutional neural network (CNN) classification model for matching and recognition, and the training method of the CNN classification model can be referred to as follows:
[0081] Step 1: Use spectrum detection equipment to collect RF signals from more than 12 different drones, use short-time Fourier transform and energy intensity color mapping algorithm to draw the drone time-frequency map, and use labeling tools to create a target detection type data set.
[0082] Step 2: Use the drone classification dataset and configuration file together with the pre-trained model and drone classification dataset to send them into the CNN network to start training, and set the hyperparameters suitable for drone classification before training. After the training, the results are saved in the form of a model weight file; and use the Adam algorithm to train the model. During the training process, the parameters in the model will adaptively learn the feature mapping relationship corresponding to different drones and establish a sample library.
[0083] In addition, after the drone type identification result is obtained, the drone type identification result can be saved, and the spectrum detection device can be turned on again to repeat step S110, and a continuously refreshed waterfall chart can be drawn using the collected RF signal; then steps S120-130 are repeated to use the trained weight file as the initial configuration of the model to deploy the two-stage model on the continuously refreshed waterfall chart at the same time to monitor the waterfall, and save and record each drone type identification result generated by the identification.
[0084] like Figure 5As shown, the present application also discloses an embodiment, providing a drone identification device, including:
[0085] A generating module 510, used to generate a time-frequency diagram to be identified based on the collected radio frequency signal;
[0086] A signal discovery module 520 is used to perform signal discovery based on the time-frequency graph to be identified, and obtain a target time-frequency graph corresponding to the signal discovery information;
[0087] The identification module 530 is used to identify the type of drone based on the target time-frequency diagram to obtain a drone type identification result.
[0088] In an optional embodiment of the present application, the generating module 510 may include:
[0089] A first acquisition unit, used to acquire acquisition frequency band parameters;
[0090] A first determining unit, used to determine a target frequency band corresponding to an acquisition frequency band parameter;
[0091] A first acquisition unit, used for acquiring radio frequency signals in a target frequency band;
[0092] The first generating unit is used to generate a time-frequency diagram to be identified according to the radio frequency signal.
[0093] In an optional embodiment of the present application, the first generating unit may include:
[0094] A transform subunit, used for performing short-time Fourier transform on the radio frequency signal to obtain the signal time-frequency information;
[0095] The mapping subunit is used to perform energy color mapping using the signal time-frequency information to generate a time-frequency diagram to be identified.
[0096] In an optional embodiment of the present application, the signal discovery module 520 may include:
[0097] A signal discovery unit, used to perform signal discovery on a preset target location area in the time-frequency diagram to be identified, and obtain signal discovery information;
[0098] The second generating unit is used to generate a target time-frequency diagram by using the signal discovery information in combination with the time-frequency diagram to be identified.
[0099] In an optional embodiment of the present application, the signal discovery unit may include:
[0100] A first determination subunit is used to determine a preset target location area in the time-frequency graph to be identified;
[0101] A detection subunit, used to detect signal occurrence frequency location information and signal occurrence time location information based on a preset target location area;
[0102] The second determining subunit is used to determine the signal occurrence frequency position information and the signal occurrence time position information as signal discovery information.
[0103] In an optional embodiment of the present application, the second generating unit may include:
[0104] A generating subunit, used for generating a signal discovery tag according to the signal discovery information;
[0105] A combining subunit, used to combine the signal discovery label and the time-frequency graph to be identified to obtain a label time-frequency graph;
[0106] The data enhancement subunit is used to perform data enhancement on the label time-frequency graph to generate the target time-frequency graph.
[0107] In an optional embodiment of the present application, the identification module 530 may include:
[0108] A second determining unit is used to determine energy change information corresponding to the target signal in the target time-frequency graph;
[0109] A third determining unit is used to determine the preset UAV category information matching the energy change information as the target UAV category information;
[0110] The third generating unit is used to generate a drone type identification result according to the target drone category information.
[0111] The implementation process of the functions and effects of each module in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, which will not be repeated here.
[0112] like Figure 6 As shown, an embodiment of the present application provides an electronic device, including a processor 610, a communication interface 620, a memory 630 and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other through the communication bus 640;
[0113] Memory 630, for storing computer programs;
[0114] In one embodiment of the present application, the processor 610 is used to execute the program stored in the memory 630 to implement the drone identification method provided by any of the aforementioned method embodiments, by generating a time-frequency graph to be identified based on the collected radio frequency signal, and performing signal discovery based on the time-frequency graph to be identified, obtaining a target time-frequency graph corresponding to the signal discovery information, and performing drone type identification based on the target time-frequency graph to obtain a drone type identification result, so that there is no need to use a visual method to detect the drone, but instead the drone type is identified using the time-frequency graph, which solves the problem that the existing drone identification method is limited by visual conditions, has low drone identification efficiency and low accuracy, and improves drone identification efficiency and accuracy.
[0115] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for identifying a drone as provided in any of the aforementioned method embodiments is implemented, by generating a time-frequency graph to be identified based on the collected radio frequency signal, and performing signal discovery based on the time-frequency graph to be identified, obtaining a target time-frequency graph corresponding to the signal discovery information, and performing drone type identification based on the target time-frequency graph to obtain a drone type identification result. This eliminates the need to use a visual method to detect the drone, and instead uses a time-frequency graph to identify the drone type. This solves the problem that the existing drone identification method is limited by visual conditions, has low drone identification efficiency and low accuracy, and improves drone identification efficiency and accuracy.
[0116] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0117] The above describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0118] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.
Claims
1. A method for identifying a drone, characterized in that: include: Generate a time-frequency diagram to be identified based on the collected radio frequency signal; Perform signal discovery based on the time-frequency graph to be identified to obtain a target time-frequency graph corresponding to the signal discovery information; The type of UAV is identified based on the target time-frequency diagram to obtain a UAV type identification result.
2. The drone identification method according to claim 1, characterized in that: The step of generating a time-frequency graph to be identified based on the collected radio frequency signal includes: Get the acquisition frequency band parameters; Determining a target frequency band corresponding to the acquisition frequency band parameter; collecting the radio frequency signal of the target frequency band; The time-frequency diagram to be identified is generated according to the radio frequency signal.
3. The drone identification method according to claim 2, characterized in that: The step of generating the time-frequency graph to be identified according to the radio frequency signal includes: Performing short-time Fourier transform on the radio frequency signal to obtain signal time-frequency information; The signal time-frequency information is used to perform energy color mapping to generate the time-frequency diagram to be identified.
4. The drone identification method according to claim 1, characterized in that: The signal discovery is performed according to the time-frequency graph to be identified to obtain a target time-frequency graph corresponding to the signal discovery information, including: Performing signal discovery on a preset target location area in the to-be-identified time-frequency graph to obtain signal discovery information; The signal discovery information is used in combination with the time-frequency diagram to be identified to generate the target time-frequency diagram.
5. The drone identification method according to claim 4, characterized in that: The performing signal discovery on the preset target location area in the to-be-identified time-frequency graph to obtain signal discovery information includes: Determining a preset target location area in the time-frequency graph to be identified; Based on the preset target location area, detecting signal occurrence frequency location information and signal occurrence time location information; The signal occurrence frequency position information and the signal occurrence time position information are determined as the signal discovery information.
6. The drone identification method according to claim 4, characterized in that: The step of using the signal discovery information and combining the time-frequency graph to be identified to generate the target time-frequency graph includes: generating a signal discovery tag according to the signal discovery information; Combining the signal discovery label with the time-frequency graph to be identified to obtain a label time-frequency graph; Data enhancement is performed on the label time-frequency graph to generate the target time-frequency graph.
7. The drone identification method according to any one of claims 1 to 6, characterized in that: The performing of drone type identification based on the target time-frequency graph to obtain a drone type identification result includes: Determine energy change information corresponding to the target signal in the target time-frequency diagram; Determine the preset drone category information matching the energy change information as the target drone category information; Generate a drone type recognition result based on the target drone category information.
8. A drone identification device, characterized in that: include: A generation module, used to generate a time-frequency diagram to be identified based on the collected radio frequency signal; A signal discovery module, used to perform signal discovery based on the time-frequency graph to be identified, and obtain a target time-frequency graph corresponding to the signal discovery information; The identification module is used to identify the type of drone based on the target time-frequency diagram to obtain a drone type identification result.
9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor is used to implement the drone identification method described in any one of claims 1 to 7 when executing the program stored in the memory.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the drone identification method according to any one of claims 1 to 7 is implemented.
Citation Information
Cited By
Unmanned aerial vehicle radio frequency signal target detection and identification method and system based on time-frequency diagram
CN122634532A
