Aircraft Identification Method, Device, Electronic Device, and Computer-Readable Medium

Abstract

Embodiments of the present disclosure disclose a method, apparatus, electronic device, and computer-readable medium for aircraft identification. A specific implementation of the method includes: capturing radio signals within a restricted flight area; determining the signal source type corresponding to the target signal source; in response to the signal source type being an aircraft type, performing the following processing steps: determining the track information of the target signal source; collecting directional sound signals according to the track information; collecting a first image set and a second image set corresponding to the target signal source according to the track information; respectively performing image enhancement on the first image set and the second image set; generating aircraft track features; generating aircraft voiceprint features; determining the aircraft shape features corresponding to the target signal source; determining the aircraft identification information for the target signal source. This implementation can accurately achieve aircraft identification.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technologies, and in particular, to methods and apparatuses for aircraft recognition, electronic devices, and computer-readable media. Background Art

[0002] An aircraft is a device capable of flying continuously off the ground. In particular, taking an unmanned aerial vehicle (UAV) as an example, due to its simple operation, it has been widely used in many fields at present. Currently, when identifying a UAV, radar detection, acoustic detection and other methods are usually adopted for identification.

[0003] However, when using the above methods, the following technical problems often exist:

[0004] Due to the flexible movement of the UAV and other characteristics, it is difficult to capture the UAV using a single detection method.

[0005] The above information disclosed in this background art section is only used to enhance the understanding of the background of the inventive concept, and thus, it may include information that does not form the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] This content section of the present disclosure is used to briefly introduce concepts that will be described in detail in the following detailed implementation section. This content section of the present disclosure is not intended to identify the key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0007] Some embodiments of the present disclosure propose methods and apparatuses for aircraft recognition, electronic devices, and computer-readable media to solve one or more of the technical problems mentioned in the above background art section.

[0008] In a first aspect, some embodiments of the present disclosure provide an aircraft identification method, which includes: capturing radio signals within a restricted flight area, where the restricted flight area is an area where untrusted aircraft are restricted from flying; determining the signal source type corresponding to the target signal source according to the radio signals, where the target signal source is the signal source that emits the radio signals, and the signal source type includes: non-aircraft type and aircraft type; in response to the signal source type being the aircraft type, perform the following processing steps: determining the flight track information of the target signal source, where the flight track information includes: a flight track, and the flight track represents the three-dimensional navigation track of the target signal source; collecting directional sound signals according to the flight track information to obtain sound signals; collecting a first image set and a second image set corresponding to the target signal source according to the flight track information, where the first image in the first image set is an image containing the target signal source collected by a high-resolution optoelectronic camera, and the second image in the second image set is an image containing the target signal source collected by a high-resolution infrared camera; performing image enhancement on the first image set and the second image set respectively to obtain a third image set and a fourth image set; generating an aircraft flight track feature according to the flight track information; generating an aircraft voiceprint feature according to the sound signals; determining the aircraft shape feature corresponding to the target signal source according to the third image set and the fourth image set; determining the aircraft identification information for the target signal source according to the aircraft flight track feature, the aircraft voiceprint feature, and the aircraft shape feature.

[0009] Second aspect, some embodiments of the present disclosure provide an aircraft identification device, the device comprising: a capture unit configured to capture radio signals within a restricted flight area, wherein the restricted flight area is an area where untrusted aircraft are restricted from flying; a determination unit configured to determine a signal source type corresponding to a target signal source according to the radio signals, wherein the target signal source is the signal source that emits the radio signals, and the signal source type includes: a non-aircraft type and an aircraft type; an execution unit configured to, in response to the signal source type being the aircraft type, perform the following processing steps: determining the track information of the target signal source, wherein the track information includes: a flight track, and the flight track represents the three-dimensional navigation track of the target signal source; collecting a directional sound signal according to the track information to obtain a sound signal; collecting a first image set and a second image set corresponding to the target signal source according to the track information, wherein the first image in the first image set is an image including the target signal source collected by a high-resolution optoelectronic camera, and the second image in the second image set is an image including the target signal source collected by a high-resolution infrared camera; respectively performing image enhancement on the first image set and the second image set to obtain a third image set and a fourth image set; generating an aircraft track feature according to the track information; generating an aircraft voiceprint feature according to the sound signal; determining an aircraft shape feature corresponding to the target signal source according to the third image set and the fourth image set; and determining aircraft identification information for the target signal source according to the aircraft track feature, the aircraft voiceprint feature, and the aircraft shape feature.

[0010] Third aspect, some embodiments of the present disclosure provide an electronic device, comprising: one or more processors; a storage device having stored thereon one or more programs, which when executed by the one or more processors cause the one or more processors to implement the method described in any implementation manner of the first aspect.

[0011] Fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having stored thereon a computer program, wherein the program, when executed by a processor, implements the method described in any implementation manner of the first aspect.

[0012] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: Through the aircraft recognition method of some embodiments of the present disclosure, the recognition success rate and accuracy for drones are improved. Specifically, the reasons for the failure to recognize successfully and the low recognition accuracy are as follows: Due to the flexible movement of drones and other characteristics, it is difficult to capture them by using radar detection. Based on this, in the aircraft recognition method of some embodiments of the present disclosure, first, radio signals within a restricted flight area are captured, where the above-mentioned restricted flight area is an area where untrusted aircraft are restricted from flying. In practice, common drones often control the drones by combining wireless signals, or transmit the images collected by the drones to the control terminal through video transmission signals. Therefore, it is possible to determine whether there is a drone by connecting to the radio signals. Next, according to the above-mentioned radio signals, the signal source type corresponding to the target signal source is determined, where the above-mentioned target signal source is the signal source that emits the above-mentioned radio signals, and the above-mentioned signal source type includes: non-aircraft type and aircraft type. In practice, radio signals are relatively complex, so it is necessary to determine whether the radio signals are emitted by drones. Further, in response to the above-mentioned signal source type being the aircraft type, the following processing steps are executed: First step, determine the flight track information of the above-mentioned target signal source, where the above-mentioned flight track information includes: flight track. In practice, different drones often correspond to different flight track characteristics, so the flight track information can be collected for subsequent aircraft recognition. Second step, according to the above-mentioned flight track information, collect directional sound signals to obtain sound signals. In practice, different drones often correspond to different acoustic characteristics, so the sound signals can be collected for subsequent aircraft recognition. Third step, according to the above-mentioned flight track information, collect a first image set and a second image set corresponding to the above-mentioned target signal source, where the first image in the above-mentioned first image set is an image obtained by a high-resolution optoelectronic camera and includes the above-mentioned target signal source, and the second image in the above-mentioned second image set is an image obtained by a high-resolution infrared camera and includes the above-mentioned target signal source. By combining the flight track, accurate image collection can be achieved. Fourth step, perform image enhancement on the above-mentioned first image set and the above-mentioned second image set respectively to obtain a third image set and a fourth image set. In practice, drones often fly at high altitudes and the targets are small, so the image quality is improved by means of image enhancement. Fifth step, generate aircraft flight track characteristics according to the above-mentioned flight track information. Sixth step, generate aircraft voiceprint characteristics according to the above-mentioned sound signals. Seventh step, determine the aircraft shape characteristics corresponding to the above-mentioned target signal source according to the above-mentioned third image set and the above-mentioned fourth image set. Eighth step, determine the aircraft recognition information for the above-mentioned target signal source according to the above-mentioned aircraft flight track characteristics, the above-mentioned aircraft voiceprint characteristics and the above-mentioned aircraft shape characteristics. By combining the aircraft flight track characteristics, the aircraft voiceprint characteristics and the aircraft shape characteristics, aircraft recognition can be accurately achieved. Description of the Drawings

[0013] In combination with the accompanying drawings and with reference to the following specific embodiments, the above and other features, advantages, and aspects of the various embodiments of the present disclosure will become more apparent. Throughout the accompanying drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the elements and elements are not necessarily drawn to scale.

[0014] Figure 1 is a flowchart of some embodiments of a method for identifying an aircraft according to the present disclosure;

[0015] Figure 2 is a schematic structural diagram of some embodiments of an aircraft identification device according to the present disclosure;

[0016] Figure 3 is a schematic structural diagram of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed Description of the Embodiments

[0017] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the accompanying drawings and embodiments of the present disclosure are only for exemplary purposes and are not used to limit the protection scope of the present disclosure.

[0018] In addition, it should be noted that for the sake of convenience of description, only parts related to the relevant invention are shown in the accompanying drawings. Without conflict, the embodiments in the present disclosure and the features in the embodiments can be combined with each other.

[0019] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules, or units, and are not used to limit the order of functions performed by these devices, modules, or units or their interdependent relationships.

[0020] It should be noted that the modifications of "one" and "multiple" mentioned in the present disclosure are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly specified in the context, it should be understood as "one or more".

[0021] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0022] The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0023] Reference Figure 1, which shows the process 100 of some embodiments of the aircraft identification method according to the present disclosure. The aircraft identification method includes the following steps:

[0024] Step 101, capturing radio signals within a restricted flight area.

[0025] In some embodiments, the execution subject of the aircraft identification method (such as Figure 1 the computing device 101 shown) can capture radio signals within a restricted flight area. Among them, the restricted flight area is an area where untrusted aircraft are restricted from flying. In practice, the above execution subject can capture radio signals within the restricted flight area through a passive radio spectrum detection device. Among them, the detection frequency band of the passive radio spectrum detection device is 30 MHz - 6 GHz, the detection distance ≥ 5 Km, the detection range is 360°, the direction finding accuracy is 3°, and the operating temperature is -40°C - 70°C.

[0026] It should be noted that the above computing device can be hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster composed of multiple servers or terminal devices, or as a single server or a single terminal device. When the computing device is embodied as software, it can be installed in the above-listed hardware devices. It can be implemented as, for example, multiple software or software modules for providing distributed services, or as a single software or software module. No specific limitation is made here.

[0027] Step 102, determining the signal source type corresponding to the target signal source according to the radio signals.

[0028] In some embodiments, the above execution subject can determine the signal source type corresponding to the target signal source according to the radio signals. Among them, the above signal source types include: non-aircraft type and aircraft type. Specifically, the non-aircraft type indicates that the radio signal is emitted by a non-aircraft device. The aircraft type indicates that the radio signal is emitted by an aircraft. When the radio signal is of the aircraft type, the radio signal can be a control signal of a drone or a video transmission signal of a drone.

[0029] In some optional implementation manners of some embodiments, the above execution subject determines the signal source type corresponding to the target signal source according to the above radio signals, which may include the following steps:

[0030] The first step is to perform multi-scale radio signal downsampling on the above radio signals to obtain a set of downsampled radio signals.

[0031] In practice, the above-mentioned execution entity can perform radio signal downsampling on radio signals through an A radio signal downsampling network, a B radio signal downsampling network, and a C radio signal downsampling network respectively. Among them, the A radio signal downsampling network, the B radio signal downsampling network, and the C radio signal downsampling network all adopt a recurrent neural network model. The receptive field of the A radio signal downsampling network > the receptive field of the B radio signal downsampling network > the receptive field of the C radio signal downsampling network.

[0032] Second step, for each downsampled radio signal in the above-mentioned set of downsampled radio signals, perform clutter point localization to determine the clutter information group corresponding to the above-mentioned downsampled radio signal.

[0033] Among them, the clutter information in the clutter information group includes: clutter position and clutter characteristics. In practice, the above-mentioned execution entity can set a localization window, and then perform binary classification judgment on whether the local radio signal within the localization window is a clutter signal. The clutter position represents the relative position of the clutter signal in the radio signal. The clutter characteristic represents the clutter signal.

[0034] As an example, taking the downsampled radio signal A as an example, the downsampled radio signal A can be composed of signal values at times T1, T2, T3, T4, T5, T6... Tn. Assuming that the window size of the localization window is 3, perform binary classification judgment on whether the local signals corresponding to times T3, T4, and T5 are clutter signals. When classified as a clutter signal, the local signals corresponding to times T3, T4, and T5 can be used as clutter characteristics, and the positions of the local signals corresponding to times T3, T4, and T5 in the radio signal can be determined as the clutter position.

[0035] Third step, according to the clutter position and clutter characteristics included in the clutter information in the obtained set of clutter information groups, perform cross-clutter removal on the set of downsampled radio signals in the above-mentioned set of downsampled radio signals to generate clutter-removed radio signals, and obtain a set of clutter-removed radio signals.

[0036] In practice, since the downsampled radio signals adopt radio signal downsampling methods with different scales, that is, corresponding clutter signals under different receptive fields, therefore, the above-mentioned execution entity can adopt the method of taking the union to perform cross-clutter removal on the set of downsampled radio signals in the above-mentioned set of downsampled radio signals to generate clutter-removed radio signals, and obtain a set of clutter-removed radio signals.

[0037] As an example, the clutter information group corresponding to the radio signal A after clutter downsampling includes: clutter information A1, clutter information A2, and clutter information A3. The clutter information group corresponding to the radio signal B after clutter downsampling includes: clutter information B1, clutter information B2, and clutter information B3. Taking clutter information A1 and clutter information B1 as an example, first, the above-mentioned execution entity first determines whether the clutter positions included in clutter information A1 and the clutter positions included in clutter information B1 are adjacent. Specifically, the above-mentioned execution entity can use the distance formula to determine the distance between the clutter positions included in clutter information A1 and the clutter positions included in clutter information B1. Then, when the distance is less than the distance threshold, the feature similarity between the clutter features included in clutter information A1 and the clutter features included in clutter information B1 is determined. Specifically, the above-mentioned execution entity can determine the feature similarity between the clutter features included in clutter information A1 and the clutter features included in clutter information B1 by calculating the cosine similarity. When the feature similarity is less than the feature similarity threshold, it can be understood that the intersection of the clutter information group corresponding to the radio signal A after clutter downsampling and the clutter information group corresponding to the radio signal B after clutter downsampling is clutter information A1 and clutter information A2. Therefore, the above-mentioned execution entity can extract the clutter signal included in clutter information A1 from the radio signal A after clutter downsampling to obtain the radio signal after clutter removal. And extract the clutter signal included in clutter information B1 from the radio signal B after clutter downsampling to obtain the radio signal after clutter removal.

[0038] In the fourth step, multi-scale radio signal feature extraction is performed on the above-mentioned radio signal set after clutter removal to generate radio signal features.

[0039] In practice, the above-mentioned execution entity can perform multi-scale radio signal feature extraction on the above-mentioned radio signal set after clutter removal through a Feature Pyramid Network (FPN) to generate radio signal features.

[0040] In the fifth step, according to the above-mentioned radio signal features, the signal source type corresponding to the above-mentioned target signal source is classified.

[0041] Among them, the above-mentioned execution entity can use a binary classification model to classify the signal source type corresponding to the above-mentioned target signal source according to the above-mentioned radio signal features.

[0042] Step 103, in response to the signal source type being the aircraft type, perform the following processing steps:

[0043] Step 1031, determine the track information of the target signal source.

[0044] In some embodiments, the above-mentioned execution entity can determine the track information of the above-mentioned target signal source.

[0045] Among them, the above-mentioned track information includes: flight track, and the above-mentioned flight track represents the three-dimensional navigation track of the above-mentioned target signal source. In practice, the above-mentioned execution entity can use a Doppler three-coordinate radar to real-time locate the position of the above-mentioned target signal source and obtain the above-mentioned track information. Specifically, the Doppler three-coordinate radar is of type YFP-01, where the detection range ≥ 5 Km, the pitch coverage angle is 0° - 40°, and the scanning method is azimuth mechanical scanning + pitch frequency scanning. Specifically, the above-mentioned execution entity can draw the above-mentioned flight track by locating the continuous three-dimensional coordinate positions of the target signal source corresponding to different moments.

[0046] Step 1032: Perform directional sound signal acquisition according to the track information to obtain a sound signal.

[0047] In some embodiments, the above-mentioned execution entity can perform directional sound signal acquisition according to the track information to obtain a sound signal. In practice, the above-mentioned execution entity can control the orientation of the directional audio acquisition device according to the latest coordinates in the flight track included in the track information to achieve directional sound signal acquisition and obtain the above-mentioned audio signal. Among them, the audio acquisition device can be a directional pick-up microphone with noise reduction function.

[0048] Step 1033: Acquire a first image set and a second image set corresponding to the target signal source according to the track information.

[0049] In some embodiments, the above-mentioned execution entity can acquire a first image set and a second image set corresponding to the target signal source according to the track information. Among them, the first image in the above-mentioned first image set is an image containing the above-mentioned target signal source acquired by a high-resolution optoelectronic camera. The second image in the above-mentioned second image set is an image containing the above-mentioned target signal source acquired by a high-resolution infrared camera. In practice, the above-mentioned execution entity can control the image acquisition directions of the high-resolution optoelectronic camera and the high-resolution infrared camera according to the latest coordinates in the flight track included in the track information to achieve simultaneous acquisition of the first image and the second image and obtain the above-mentioned first image set and second image set.

[0050] Step 1034: Perform image enhancement on the first image set and the second image set respectively to obtain a third image set and a fourth image set.

[0051] In some embodiments, the above-mentioned execution entity can perform image enhancement on the first image set and the second image set respectively to obtain a third image set and a fourth image set.

[0052] In some optional implementation manners of some embodiments, the above-mentioned execution entity performs image enhancement on the first image set and the second image set respectively to obtain a third image set and a fourth image set, which may include the following steps:

[0053] First step, perform initial image feature extraction on each first image in the above-mentioned first image set to generate first initial image features.

[0054] Among them, the first initial image features include: global image features and local image features.

[0055] In practice, the above-mentioned execution entity may set up 2 image feature extraction networks, namely image feature extraction network A and image feature extraction network B. Among them, the receptive field of image feature extraction network A is smaller than the receptive field of image feature extraction network B. Specifically, the above-mentioned execution entity may input the first image into the above-mentioned image feature extraction network A and image feature extraction network B respectively to obtain the first initial image features corresponding to the first image.

[0056] Second step, for each first image in the above-mentioned first image set, in response to the image position of the first image in the first image set not being the last position, perform the following first image enhancement steps:

[0057] First sub-step, determine the image feature similarity between the first initial image features corresponding to the first image and the first image features corresponding to each fifth image in the fifth image set.

[0058] Among them, the fifth images in the above-mentioned fifth image set are the first images in the first image set whose corresponding image positions are after the image position of the first image.

[0059] Second sub-step, screen out the fifth images in the above-mentioned fifth image set whose corresponding image feature similarity is greater than the preset similarity and whose corresponding image positions are continuous as the sixth images to obtain a sixth image set.

[0060] As an example, the fifth image set includes: fifth image A, fifth image B, fifth image C, fifth image D. Among them, the image similarity corresponding to fifth image A is greater than the preset similarity, the image similarity corresponding to fifth image B is greater than the preset similarity, the image similarity corresponding to fifth image C is less than the preset similarity, and the image similarity corresponding to fifth image D is greater than the preset similarity. Therefore, the above-mentioned execution entity may use fifth image A, fifth image B, and fifth image C as the sixth image set.

[0061] Third sub-step, perform deep image feature extraction on the first initial image features corresponding to the sixth images in the above-mentioned sixth image set to generate first deep image features, and obtain a first deep image feature set.

[0062] Among them, the above-mentioned execution entity uses the image feature extraction network C to perform deep image feature extraction on the first initial image features corresponding to the sixth image in the above-mentioned sixth image set to generate first deep image features. Among them, the network depth of the image feature extraction network C > the network depth of the image feature extraction network A, and the network depth of the image feature extraction network C > the network depth of the image feature extraction network B.

[0063] The fourth sub-step is to perform image enhancement on the above-mentioned first image according to the above-mentioned first deep image feature set to obtain the third image corresponding to the above-mentioned first image in the above-mentioned third image set.

[0064] In practice, for example, the above-mentioned execution entity can adopt the method of image superposition to perform feature superposition on the first deep image feature set and the first image to obtain the third image corresponding to the above-mentioned first image in the above-mentioned third image set. Another example is that since the target signal source has a small proportion in the image, therefore, the above-mentioned execution entity can adopt the method of local image enhancement to perform image enhancement on the above-mentioned first image according to the above-mentioned first deep image feature set to obtain the third image corresponding to the above-mentioned first image in the above-mentioned third image set.

[0065] The third step is that for each second image in the above-mentioned second image set, in response to the image position of the second image in the above-mentioned second image set not being the last position, perform the following second image enhancement steps:

[0066] The first sub-step is to determine the image feature similarity between the second initial image features corresponding to the above-mentioned second image and the second image features corresponding to each seventh image in the seventh image set.

[0067] Among them, the seventh image in the above-mentioned seventh image set is the second image in the above-mentioned second image set whose corresponding image position is after the image position of the second image.

[0068] The second sub-step is to screen out the seventh images in the above-mentioned seventh image set whose corresponding image feature similarity is greater than the preset similarity and whose corresponding image positions are continuous as the eighth images to obtain an eighth image set.

[0069] The third sub-step is to perform deep image feature extraction on the second initial image features corresponding to the eighth images in the above-mentioned eighth image set to generate second deep image features to obtain a second deep image feature set.

[0070] The fourth sub-step is to perform image enhancement on the above-mentioned second image according to the above-mentioned second deep image feature set to obtain the fourth image corresponding to the above-mentioned second image in the above-mentioned fourth image set.

[0071] In practice, the implementation of the above second image enhancement step can refer to the first image enhancement step, which will not be elaborated here.

[0072] Step 1035: Generate the flight track features of the aircraft according to the track information.

[0073] In some embodiments, the above execution entity can generate the flight track features of the aircraft according to the track information. In practice, the above execution entity can use a recurrent neural network model to extract features from the flight track included in the track information to generate the flight track features of the aircraft.

[0074] In some optional implementation manners of some embodiments, the above execution entity generating the flight track features of the aircraft according to the above track information may include the following steps:

[0075] First step: Voxelize the above flight track to generate the voxelized flight track.

[0076] Among them, the above execution entity can voxelize the above flight track through an octree to generate the voxelized flight track.

[0077] Second step: Use the main flight track feature extraction network to perform an initial extraction of the flight track on the above voxelized flight track to generate the initially extracted flight track features.

[0078] Among them, the above execution entity can use the main flight track feature extraction network to perform an initial extraction of the flight track on the above voxelized flight track to generate the initially extracted flight track features. In practice, the main flight track feature extraction network uses a Bi-LSTM model.

[0079] Third step: Use the flight track direction transformation feature extraction network to extract the flight track direction transformation features from the above initially extracted flight track features to generate the flight track direction transformation features.

[0080] Among them, the flight track direction transformation feature extraction network uses a convolutional neural network including 5 serially connected convolutional layers to extract the features of the flight track direction transformation angle.

[0081] Fourth step: Use the flight track altitude transformation feature extraction network to extract the flight track altitude transformation features from the above initially extracted flight track features to generate the flight track altitude transformation features.

[0082] Among them, the flight track altitude transformation feature extraction network has the same model structure as the flight track direction transformation feature extraction network.

[0083] Fifth step: Determine the above flight track direction transformation features and the above flight track altitude transformation features as the above flight track features of the aircraft.

[0084] Among them, the above-mentioned main flight path feature extraction network, flight path direction transformation feature extraction network, and flight path altitude transformation feature extraction network are included in a pre-trained aircraft flight path feature extraction model.

[0085] Step 1036: Generate an aircraft voiceprint feature according to the voice signal.

[0086] In some embodiments, the above-mentioned execution subject can generate an aircraft voiceprint feature according to the voice signal. In practice, the above-mentioned execution subject can combine an LSTM model to generate an aircraft voiceprint feature according to the voice signal.

[0087] Step 1037: Determine the aircraft shape feature corresponding to the target signal source according to the third image set and the fourth image set.

[0088] In some embodiments, the above-mentioned execution subject can determine the aircraft shape feature corresponding to the target signal source according to the third image set and the fourth image set.

[0089] In some optional implementation manners of some embodiments, the above-mentioned execution subject determines the aircraft shape feature corresponding to the target signal source according to the third image set and the fourth image set, including:

[0090] Perform aircraft shape feature extraction on the above-mentioned third image set and the above-mentioned fourth image set through an aircraft shape feature extraction model to obtain the above-mentioned aircraft shape feature.

[0091] Among them, the above-mentioned aircraft shape feature extraction model includes: an aircraft positioning network, an aircraft main structure feature extraction network, and an aircraft drive structure feature extraction network. In practice, the aircraft positioning network adopts an EfficientDet network. The aircraft main structure feature extraction network and the aircraft drive structure feature extraction network adopt the same network structure, both of which adopt a Tiny YOLO network without a locator.

[0092] Step 1038: Determine the aircraft identification information for the target signal source according to the aircraft flight path feature, the aircraft voiceprint feature, and the aircraft shape feature.

[0093] In some embodiments, the above-mentioned execution subject can determine the aircraft identification information for the target signal source according to the aircraft flight path feature, the aircraft voiceprint feature, and the aircraft shape feature. Among them, the above-mentioned execution subject can use a pre-trained model (GPT) based on a Transformer structure as a prediction head, and combine the aircraft flight path feature, the aircraft voiceprint feature, and the aircraft shape feature to determine the aircraft identification information for the target signal source.

[0094] Optionally, the above method further includes:

[0095] First, match the above-mentioned aircraft identification information with the aircraft information in the aircraft information database to generate a matching result.

[0096] Among them, the above-mentioned matching result indicates whether the above-mentioned target signal source is an untrusted aircraft, and the above-mentioned aircraft information database is used to store the aircraft information of trusted aircraft. The aircraft information database can use the database that stores the pre-collected aircraft information.

[0097] Second, in response to the above-mentioned matching result indicating that the above-mentioned target signal source is an untrusted aircraft, drive away the above-mentioned target signal source as an aircraft.

[0098] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: Through the aircraft recognition method of some embodiments of the present disclosure, the recognition success rate and accuracy for drones are improved. Specifically, the reasons for the failure to recognize successfully and the low recognition accuracy are as follows: Due to the flexible movement of drones and other characteristics, it is difficult to capture them by using radar detection. Based on this, in the aircraft recognition method of some embodiments of the present disclosure, first, radio signals within a restricted flight area are captured, where the above-mentioned restricted flight area is an area where untrusted aircraft are restricted from flying. In practice, common drones often control the drone by combining wireless signals or transmit the images collected by the drone to the control end through video transmission signals. Therefore, it is possible to determine whether there is a drone by connecting to the radio signals. Next, according to the above-mentioned radio signals, the signal source type corresponding to the target signal source is determined, where the above-mentioned target signal source is the signal source that emits the above-mentioned radio signals, and the above-mentioned signal source type includes: non-aircraft type and aircraft type. In practice, radio signals are relatively complex, so it is necessary to determine whether the radio signals are emitted by drones. Further, in response to the above-mentioned signal source type being the aircraft type, the following processing steps are executed: First step, determine the flight path information of the above-mentioned target signal source, where the above-mentioned flight path information includes: flight track. In practice, different drones often correspond to different flight path characteristics, so the flight path information can be collected for subsequent aircraft recognition. Second step, according to the above-mentioned flight path information, directional sound signals are collected to obtain sound signals. In practice, different drones often correspond to different acoustic characteristics, so the sound signals can be collected for subsequent aircraft recognition. Third step, according to the above-mentioned flight path information, a first image set and a second image set corresponding to the above-mentioned target signal source are collected, where the first image in the above-mentioned first image set is an image containing the above-mentioned target signal source collected by a high-resolution optoelectronic camera, and the second image in the above-mentioned second image set is an image containing the above-mentioned target signal source collected by a high-resolution infrared camera. By combining the flight path, accurate image collection can be achieved. Fourth step, image enhancement is performed on the above-mentioned first image set and the above-mentioned second image set respectively to obtain a third image set and a fourth image set. In practice, drones often fly at high altitudes and the targets are small, so the image quality is improved by means of image enhancement. Fifth step, according to the above-mentioned flight path information, an aircraft flight path characteristic is generated. Sixth step, according to the above-mentioned sound signals, an aircraft voiceprint characteristic is generated. Seventh step, according to the above-mentioned third image set and the above-mentioned fourth image set, the aircraft shape characteristic corresponding to the above-mentioned target signal source is determined. Eighth step, according to the above-mentioned aircraft flight path characteristic, the above-mentioned aircraft voiceprint characteristic, and the above-mentioned aircraft shape characteristic, the aircraft recognition information for the above-mentioned target signal source is determined. By combining the aircraft flight path characteristic, the aircraft voiceprint characteristic, and the aircraft shape characteristic, aircraft recognition can be accurately achieved.

[0099] Further reference is made to Figure 2 , as an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of an aircraft identification device, and these device embodiments correspond to Figure 1 the method embodiments shown, and the aircraft identification device can be specifically applied to various electronic devices.

[0100] As shown in Figure 2 , some embodiments of the aircraft identification device 200 include: a capture unit 201, a determination unit 202, and an execution unit 203. Among them, the capture unit 201 is configured to capture radio signals within a restricted flight area, where the restricted flight area is an area where untrusted aircraft are restricted from flying; the determination unit 202 is configured to determine the signal source type corresponding to the target signal source according to the radio signals, where the target signal source is the signal source that emits the radio signals, and the signal source type includes: non-aircraft type and aircraft type; the execution unit 203 is configured to, in response to the signal source type being the aircraft type, perform the following processing steps: determine the flight path information of the target signal source, where the flight path information includes: a flight path, and the flight path represents the three-dimensional navigation trajectory of the target signal source; collect directional sound signals according to the flight path information to obtain sound signals; collect a first image set and a second image set corresponding to the target signal source according to the flight path information, where the first image in the first image set is an image including the target signal source collected by a high-resolution optoelectronic camera, and the second image in the second image set is an image including the target signal source collected by a high-resolution infrared camera; perform image enhancement on the first image set and the second image set respectively to obtain a third image set and a fourth image set; generate an aircraft flight path feature according to the flight path information; generate an aircraft voiceprint feature according to the sound signals; determine the aircraft shape feature corresponding to the target signal source according to the third image set and the fourth image set; determine the aircraft identification information for the target signal source according to the aircraft flight path feature, the aircraft voiceprint feature, and the aircraft shape feature.

[0101] It can be understood that the various units described in the aircraft identification device 200 correspond to the respective steps in the method described with reference to Figure 1 . Thus, the operations, features, and beneficial effects described above for the method also apply to the aircraft identification device 200 and the units included therein, and will not be repeated here.

[0102] Next, reference is made to Figure 3, which shows a schematic structural diagram of an electronic device (e.g., a computing device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The illustrated electronic device is merely an example and should not impose any limitations on the functions and usage scope of the embodiments of the present disclosure.

[0103] As Figure 3 shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which may perform various appropriate actions and processes according to the program stored in the read-only memory 302 or the program loaded from the storage device 308 into the random access memory 303. In the random access memory 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the read-only memory 302, and the random access memory 303 are connected to each other through a bus 304. The input / output interface 305 is also connected to the bus 304.

[0104] Generally, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or wiredly to exchange data. Although Figure 3 the electronic device 300 with various devices is shown, it should be understood that it is not required to implement or have all the shown devices. Instead, more or fewer devices may be implemented or had. Figure 3 Each block shown in

[0105] Specifically, according to some embodiments of the present disclosure, the process described above with reference to the flowchart may be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the method shown in the flowchart. In such some embodiments, the computer program may be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the read-only memory 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0106] It should be noted that the computer-readable media described in some embodiments of the present disclosure may be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program, and the program may be used by or in combination with an instruction execution system, apparatus, or device. In some embodiments of the present disclosure, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than the computer-readable storage medium, and the computer-readable signal medium may send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted by any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0107] In some embodiments, the client and the server may communicate using any currently known or future-developed network protocol such as HTTP (Hyper Text Transfer Protocol), and may be interconnected with digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include local area networks ("LANs"), wide area networks ("WANs"), the Internet (e.g., the Internet), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0108] The above computer-readable medium may be included in the above electronic device; or it may exist independently and not be assembled into the electronic device. The above computer-readable medium carries one or more programs. When the above one or more programs are executed by the electronic device, the electronic device is caused to: capture radio signals within a restricted flight area, where the restricted flight area is an area where untrusted aircraft are restricted from flying; determine the signal source type corresponding to the target signal source according to the above radio signals, where the target signal source is the signal source that emits the above radio signals, and the signal source type includes: non-aircraft type and aircraft type; in response to the signal source type being the aircraft type, perform the following processing steps: determine the flight track information of the above target signal source, where the flight track information includes: a flight track, and the flight track represents the three-dimensional navigation track of the above target signal source; collect directional sound signals according to the above flight track information to obtain sound signals; collect a first image set and a second image set corresponding to the above target signal source according to the above flight track information, where the first image in the first image set is an image containing the above target signal source collected by a high-resolution optoelectronic camera, and the second image in the second image set is an image containing the above target signal source collected by a high-resolution infrared camera; perform image enhancement on the above first image set and the above second image set respectively to obtain a third image set and a fourth image set; generate an aircraft flight track feature according to the above flight track information; generate an aircraft voiceprint feature according to the above sound signals; determine the aircraft shape feature corresponding to the above target signal source according to the above third image set and the above fourth image set; determine the aircraft identification information for the above target signal source according to the above aircraft flight track feature, the above aircraft voiceprint feature, and the above aircraft shape feature.

[0109] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages or combinations thereof. The above programming languages include object-oriented programming languages - such as Java, Smalltalk, C++; and also include conventional procedural programming languages - such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network - including a local area network (LAN) or a wide area network (WAN) - or may be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0110] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0111] The units described in some embodiments of the present disclosure can be implemented in software or in hardware. The described units can also be provided in a processor. For example, it can be described as: a processor includes a capture unit, a determination unit, and an execution unit. Among them, the names of these units do not constitute a limitation on the unit itself in some cases. For example, the capture unit can also be described as "a unit that captures radio signals within a restricted flight area, where the above-mentioned restricted flight area is an area where untrusted aircraft are restricted from flying".

[0112] The functions described above can be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that can be used include: Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), Systems on Chip (SOCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0113] The above description is only some preferred embodiments of the present disclosure and an explanation of the technical principles applied. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, technical solutions formed by mutually replacing the above features with technical features (but not limited to) having similar functions disclosed in the embodiments of the present disclosure.

Claims

1. An aircraft identification method, comprising: capturing a radio signal in a restricted flight area, wherein the restricted flight area is an area where untrusted aircraft are restricted from flying; Determine, according to the radio signal, a signal source type corresponding to a target signal source, wherein the target signal source is a signal source that sends out the radio signal, and the signal source type includes: a non-aircraft type and an aircraft type; In response to the signal source type being an aircraft type, the following processing steps are performed: Determine the track information of the target signal source, wherein the track information includes: a flight track, the flight track represents a three-dimensional navigation trajectory of the target signal source; According to the track information, directional sound signal collection is performed to obtain a sound signal; According to the track information, a first image set and a second image set corresponding to the target signal source are collected, wherein the first image in the first image set is an image obtained by a high-resolution optoelectronic camera and includes the target signal source, and the second image in the second image set is an image obtained by a high-resolution infrared camera and includes the target signal source; performing image enhancement on the first image set and the second image set respectively to obtain a third image set and a fourth image set; generating aircraft track characteristics according to the track information; Generating an aircraft voiceprint feature according to the sound signal; Determining, according to the third image set and the fourth image set, an aircraft shape feature corresponding to the target signal source; Aircraft identification information for the target signal source is determined according to the aircraft track characteristics, the aircraft soundprint characteristics and the aircraft appearance characteristics.

2. The method according to claim 1, wherein: The method further comprises: Matching the aircraft identification information with aircraft information in an aircraft information library to generate a matching result, wherein the matching result indicates whether the target signal source is an untrusted aircraft, and the aircraft information library is used to store aircraft information of trusted aircraft; In response to the matching result indicating that the target signal source is an untrusted aircraft, the target signal source is driven away by an aircraft.

3. The method according to claim 2, wherein: The step of determining the signal source type corresponding to the target signal source according to the radio signal includes: Performing multi-scale radio signal downsampling on the radio signal to obtain a downsampled radio signal set; Performing clutter point positioning for each downsampled radio signal in the downsampled radio signal set to determine a clutter information group corresponding to the downsampled radio signal, wherein the clutter information in the clutter information group includes: clutter position and clutter characteristics; According to the clutter position and clutter characteristics included in the clutter information in the obtained clutter information group set, cross clutter removal is performed on the downsampled radio signal set in the downsampled radio signal set to generate clutter-removed radio signals, thereby obtaining a clutter-removed radio signal set; Performing multi-scale radio signal feature extraction on the radio signal set after clutter removal to generate radio signal features; According to the radio signal characteristics, a signal source type corresponding to the target signal source is classified.

4. The method according to claim 3, wherein: The performing image enhancement on the first image set and the second image set respectively to obtain a third image set and a fourth image set comprises: Performing initial image feature extraction on each first image in the first image set to generate first initial image features, wherein the first initial image features include: global image features and local image features; For each first image in the first image set, in response to the image position of the first image in the first image set not being the last position, performing the following first image enhancement step: determining an image feature similarity between a first initial image feature corresponding to the first image and a first image feature corresponding to each fifth image in a fifth image set, wherein the fifth image in the fifth image set is a first image in the first image set whose corresponding image position is located after the image position of the first image; Selecting fifth images whose corresponding image feature similarity is greater than a preset similarity and whose corresponding image positions are continuous from the fifth image set as sixth images, thereby obtaining a sixth image set; Performing depth image feature extraction on a first initial image feature corresponding to a sixth image in the sixth image set to generate a first depth image feature, thereby obtaining a first depth image feature set; According to the first depth image feature set, the first image is enhanced to obtain a third image in the third image set corresponding to the first image; initial image feature extraction is performed on each second image in the second image set to generate a second initial image feature, wherein the second initial image feature includes: a global image feature and a local image feature; For each second image in the second image set, in response to the image position of the second image in the second image set not being the last position, performing the following second image enhancement step: determining an image feature similarity between a second initial image feature corresponding to the second image and a second image feature corresponding to each seventh image in a seventh image set, wherein the seventh image in the seventh image set is a second image in the second image set whose corresponding image position is located after the image position of the second image; Selecting, from the seventh image set, seventh images whose corresponding image feature similarity is greater than a preset similarity and whose corresponding image positions are continuous as eighth images, thereby obtaining an eighth image set; Performing depth image feature extraction on the second initial image features corresponding to the eighth image in the eighth image set to generate second depth image features, thereby obtaining a second depth image feature set; The second image is enhanced according to the second depth image feature set to obtain a fourth image in the fourth image set that corresponds to the second image.

5. The method according to claim 4, wherein: Generating the aircraft track characteristics according to the track information includes: voxelize the flight track to generate a voxelized flight track; Performing initial track extraction on the voxelized flight track through a main track feature extraction network to generate initial extracted track features; Performing track direction change feature extraction on the initial extracted track features through a track direction change feature extraction network to generate a track direction change feature; Extracting the track altitude change feature from the initial extracted track feature through a track altitude change feature extraction network to generate a track altitude change feature; The track direction change feature and the track altitude change feature are determined as the aircraft track feature, wherein the main track feature extraction network, the track direction change feature extraction network and the track altitude change feature extraction network are included in a pre-trained aircraft track feature extraction model.

6. The method according to claim 5, wherein: The determining, based on the third image set and the fourth image set, the aircraft shape feature corresponding to the target signal source comprises: The aircraft shape feature extraction model is used to extract the aircraft shape features of the third image set and the fourth image set to obtain the aircraft shape features, wherein the aircraft shape feature extraction model includes: an aircraft positioning network, an aircraft trunk structure feature extraction network and an aircraft drive structure feature extraction network.

7. An aircraft identification device, comprising: a capture unit configured to capture a radio signal in a restricted flight area, wherein the restricted flight area is an area where untrusted aircraft are restricted from flying; a determining unit configured to determine a signal source type corresponding to a target signal source according to the radio signal, wherein the target signal source is a signal source that sends the radio signal, and the signal source type includes: a non-aircraft type and an aircraft type; The execution unit is configured to execute the following processing steps in response to the signal source type being an aircraft type: determining the track information of the target signal source, wherein the track information includes: a flight track, wherein the flight track represents the three-dimensional navigation trajectory of the target signal source; performing directional sound signal acquisition according to the track information to obtain a sound signal; acquiring a first image set and a second image set corresponding to the target signal source according to the track information, wherein the first image in the first image set is an image obtained by a high-resolution optoelectronic camera and contains the target signal source, and the second image in the second image set is an image obtained by a high-resolution infrared camera and contains the target signal source; performing image enhancement on the first image set and the second image set respectively to obtain a third image set and a fourth image set; generating an aircraft track feature according to the track information; generating an aircraft soundprint feature according to the sound signal; determining an aircraft shape feature corresponding to the target signal source according to the third image set and the fourth image set; determining aircraft identification information for the target signal source according to the aircraft track feature, the aircraft soundprint feature and the aircraft shape feature.

8. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

9. A computer readable medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images