Lightweight unmanned aerial vehicle identification method and device, terminal and medium

Abstract

The invention discloses a lightweight unmanned aerial vehicle identification method and device, a terminal and a medium, and belongs to the technical field of unmanned aerial vehicle identification. When unmanned aerial vehicle identification request data is received, fusion grey-scale map data of an unmanned aerial vehicle in the unmanned aerial vehicle identification request data is acquired; and substituting the fused grey-scale map data into the optimal lightweight network model obtained by training, and carrying out unmanned aerial vehicle identification modeling by using the optimal lightweight network model. According to the lightweight unmanned aerial vehicle identification method and device, the terminal and the medium provided by the invention, the target micro-Doppler features are enhanced through singular value decomposition reconstruction, multiple micro-Doppler feature grey-scale maps are fused, the dynamic feature capture capability of the target is improved, and meanwhile, the model complexity is reduced by using a lightweight network structure, so that the recognition efficiency of the unmanned aerial vehicle is improved. The method is suitable for efficient deployment of embedded equipment, and accurate identification of an unmanned aerial vehicle target is realized.

Description

Technical Field

[0001] The invention discloses a lightweight unmanned aerial vehicle (UAV) identification method, device, terminal and medium, and belongs to the technical field of UAV identification. Background Art

[0002] In recent years, the research on combining micro-Doppler technology with deep learning has made rapid progress, and a variety of innovative algorithms and technologies have been proposed. In terms of micro-Doppler feature extraction, the main methods used are single time-frequency graph and multi-feature graph fusion; in terms of target classification technology, support vector machine (SVM) and deep learning technology have been widely used. However, the feature extraction method of a single time-frequency graph limits its ability to capture the dynamic characteristics of the target, and the SVM classification accuracy is low. Although multi-feature graph fusion and deep learning technology are used for target classification, most of the existing network models have too many parameters and high computational complexity, which makes it difficult to meet the real-time requirements of embedded devices.

[0003] Most current research focuses on the recognition of a single feature map, and most adopt complex network models, which leads to insufficient feature extraction and difficulty in deployment on embedded devices. Summary of the invention

[0004] In view of the defects of insufficient recognition ability and difficult embedded device deployment caused by single feature map and high-complexity network model in the prior art, the present invention is based on the combination of micro-Doppler technology and deep learning, and proposes a lightweight UAV identification method, device, terminal and medium based on singular value decomposition and micro-Doppler multi-feature grayscale image fusion.

[0005] The technical solution of the present invention is as follows: According to a first aspect of an embodiment of the present invention, a lightweight drone identification method is provided, comprising: When receiving the drone identification request data, obtaining the fused grayscale image data of the drone in the drone identification request data; The fused grayscale image data is introduced into the best lightweight network model obtained through training, and the best lightweight network model is used to perform drone recognition modeling.

[0006] Furthermore, when receiving the drone identification request data, before obtaining the fused grayscale image data of the drone in the drone identification request data, it also includes: Obtain multi-feature grayscale image fusion data and build an improved lightweight network model for drone recognition; The improved lightweight network model is trained using the multi-feature grayscale image fusion data to obtain the optimal lightweight network model.

[0007] Furthermore, the obtaining of multi-feature grayscale image fusion data includes: Acquire radar echo data, and perform short-time Fourier transform on the radar echo data to generate a time-frequency spectrum of the target; Performing singular value decomposition on the time-frequency spectrum and reconstructing an enhanced time-frequency spectrum; The enhanced time-frequency spectrum, rhythm-speed map and short-time cepstrum map are subjected to multi-dimensional grayscale image fusion to obtain multi-feature grayscale image fusion data.

[0008] Furthermore, the improved lightweight network model includes: Traditional convolutional layers, used to extract basic features from the input image; The bottleneck layer is used to model the dependencies between channels; The classification head is used to convert the extracted feature map into the final classification result.

[0009] Further, the generating a time-frequency spectrum of the target by performing a short-time Fourier transform on the radar echo data comprises: Determining a range cell of a target and extracting a plurality of chirp signals from the range cell; Arrange the plurality of chirp signals in sequence into a vector, define the length and step size of the window function, and determine the sampling frequency as the inverse of the time interval between the chirp signals; Perform a short-time Fourier transform on the constructed signal vector to generate a time-spectrogram.

[0010] Furthermore, performing singular value decomposition on the time-frequency spectrum and reconstructing the enhanced time-frequency spectrum includes: For the input time-frequency matrix Perform singular value decomposition: (1) in, is the left singular vector matrix, is a diagonal matrix of singular values; from Before extraction The largest singular values ​​and their corresponding left singular vectors and the right singular vectors ; before removal After the principal components are obtained, the matrix is ​​reconstructed according to the remaining singular values ​​and singular vectors to obtain the enhanced time-frequency spectrum: (2) in, are the remaining singular values; is the corresponding left singular vector, is the corresponding right singular vector, For the matrix rank, is the reconstructed matrix.

[0011] According to a second aspect of an embodiment of the present invention, a lightweight drone identification device is provided, comprising: A receiving module, configured to obtain the fused grayscale image data of the drone in the drone identification request data when receiving the drone identification request data; The recognition module brings the fused grayscale image data into the best lightweight network model obtained through training, and uses the best lightweight network model to perform drone recognition modeling.

[0012] According to a third aspect of an embodiment of the present invention, a terminal is provided, including: one or more processors; a memory for storing said one or more processor executable instructions; Wherein, the one or more processors are configured to: Execute the method described in the first aspect of the embodiment of the present invention.

[0013] According to a fourth aspect of an embodiment of the present invention, a non-temporary computer-readable storage medium is provided. When instructions in the storage medium are executed by a processor of a terminal, the terminal is enabled to execute the method described in the first aspect of the embodiment of the present invention.

[0014] According to a fifth aspect of an embodiment of the present invention, an application product is provided. When the application product is running on a terminal, the terminal executes the method described in the first aspect of the embodiment of the present invention.

[0015] The beneficial effects of the present invention are: The present invention provides a lightweight UAV identification method, device, terminal and medium, which enhance the target micro-Doppler characteristics through singular value decomposition reconstruction and fuse multiple micro-Doppler feature grayscale images to improve the target's dynamic feature capture capability. At the same time, a lightweight network structure is used to reduce the model complexity, making it suitable for efficient deployment of embedded devices and realizing accurate identification of UAV targets.

[0016] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The present invention is a flowchart of a lightweight UAV identification method according to an exemplary embodiment.

[0018] Figure 2 The present invention is a flowchart of a lightweight UAV identification method according to an exemplary embodiment.

[0019] Figure 3 The present invention is a flowchart of a lightweight UAV identification method according to an exemplary embodiment.

[0020] Figure 4 It is a time-frequency diagram obtained by performing short-time Fourier transform on radar echo signals of different targets in a lightweight UAV identification method shown according to an exemplary embodiment.

[0021] Figure 5 It is a time-frequency diagram of different targets after mean elimination in a lightweight UAV identification method according to an exemplary embodiment.

[0022] Figure 6 It is a time-frequency diagram of different targets reconstructed using singular value decomposition in a lightweight drone identification method according to an exemplary embodiment.

[0023] Figure 7 It is a schematic block diagram of the structure of a flight action recognition device based on a bidirectional recurrent neural network according to an exemplary embodiment.

[0024] Figure 8 The present invention is a schematic block diagram of a terminal structure according to an exemplary embodiment. DETAILED DESCRIPTION

[0025] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] In the description of the present invention, it should be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc. indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they cannot be understood as limitations on the present invention.

[0027] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0028] An embodiment of the present invention provides a lightweight drone identification method, which is implemented by a terminal. The terminal may be a desktop computer or a laptop computer, etc. The terminal at least includes a CPU, etc.

[0029] Embodiment 1: Figure 1 The present invention is a flowchart of a lightweight drone identification method according to an exemplary embodiment. The method is used in a terminal, and the method includes the following steps: Step 101, when receiving drone identification request data, obtaining fused grayscale image data of the drone in the drone identification request data; Step 102: bring the fused grayscale image data into the best lightweight network model obtained through training, and use the best lightweight network model to perform drone recognition modeling.

[0030] Furthermore, when receiving the drone identification request data, before obtaining the fused grayscale image data of the drone in the drone identification request data, it also includes: Obtain multi-feature grayscale image fusion data and build an improved lightweight network model for drone recognition; The improved lightweight network model is trained using the multi-feature grayscale image fusion data to obtain the optimal lightweight network model.

[0031] Furthermore, the obtaining of multi-feature grayscale image fusion data includes: Acquire radar echo data, and perform short-time Fourier transform on the radar echo data to generate a time-frequency spectrum of the target; Performing singular value decomposition on the time-frequency spectrum and reconstructing an enhanced time-frequency spectrum; The enhanced time-frequency spectrum, rhythm-speed map and short-time cepstrum map are subjected to multi-dimensional grayscale image fusion to obtain multi-feature grayscale image fusion data.

[0032] Furthermore, the improved lightweight network model includes: Traditional convolutional layers, used to extract basic features from the input image; The bottleneck layer is used to model the dependencies between channels; The classification head is used to convert the extracted feature map into the final classification result.

[0033] Further, the generating a time-frequency spectrum of the target by performing a short-time Fourier transform on the radar echo data comprises: The range cell of the target is determined and each chirp signal is extracted from that range cell.

[0034] These signals are arranged sequentially into a vector, and then the length and step size of the window function are defined, and the sampling frequency is determined as the inverse of the time interval between the chirp signals.

[0035] Perform a short-time Fourier transform on the constructed signal vector x[n] to generate a time-spectrogram.

[0036] Furthermore, performing singular value decomposition on the time-frequency spectrum and reconstructing the enhanced time-frequency spectrum includes: For the input time-frequency matrix Perform singular value decomposition: (1) in, is the left singular vector matrix, is a diagonal matrix of singular values, is the right singular vector matrix; from Before extraction The largest singular values ​​and their corresponding left singular vectors and the right singular vectors ; before removal After the principal components are obtained, the matrix is ​​reconstructed according to the remaining singular values ​​and singular vectors to obtain the enhanced time-frequency spectrum: (2) in, are the remaining singular values; is the corresponding left singular vector, is the corresponding right singular vector, For the matrix rank, is the reconstructed matrix.

[0037] Embodiment 2: Figure 2 and Figure 3 The present invention is a flowchart of a lightweight drone identification method according to an exemplary embodiment. The method is used in a terminal, and the method includes the following steps: Step S201, obtain multi-feature grayscale image fusion data and construct an improved lightweight network model for drone identification. The specific contents are as follows: Acquire radar echo data and perform short-time Fourier transform on the radar echo data to generate the target's time-frequency spectrum. First, the range bin of the target is determined and each chirp signal is extracted from the range bin.

[0038] Then, these signals are arranged sequentially into a vector, the length and step size of the window function are defined, and the sampling frequency is determined as the inverse of the time interval between the chirp signals.

[0039] Finally, a short-time Fourier transform is performed on the constructed signal vector x[n] to generate a time-frequency spectrum. Through these operations, the time-frequency spectrum of various types of drones can be obtained, such as Figure 4 (a) Figure 4 (b) Figure 4 (c) and Figure 4 As shown in (d) in .

[0040] Since the main velocity of the target often masks the micro-Doppler characteristics of the rotor blades, thus affecting the accuracy of target classification, and the traditional mean cancellation method will lead to incomplete removal of the main velocity and affect the micro-Doppler component, such as Figure 5 (a) Figure 5 (b) Figure 5 (c) and Figure 5 As shown in (d) in the figure. In order to solve this problem, a new method is proposed to remove the main velocity of the target time-frequency graph through singular values, and the principal component corresponding to the maximum singular value is removed by performing singular value decomposition on the time-frequency spectrum, and the matrix is ​​reconstructed. The proposed method significantly improves the micro-Doppler characteristics of the target, as shown in Figure 6 (a) Figure 6 (b) Figure 6 (c) and Figure 6 The specific algorithm is as follows: for the input time-frequency matrix Perform singular value decomposition: (1) in, is the left singular vector matrix, is a diagonal matrix of singular values, is the right singular vector matrix; from Before extraction The largest singular values ​​and their corresponding left singular vectors and the right singular vectors ; before removal After principal components, the matrix is ​​reconstructed according to the remaining singular values ​​and singular vectors: (2) in, are the remaining singular values; is the corresponding left singular vector, is the corresponding right singular vector, For the matrix The rank of the reconstructed matrix This is the time-frequency matrix after removing the main velocity component. Figure 5 (a) Figure 5 (b) Figure 5 (c) and Figure 5 As shown in (d), it is the time-frequency diagram of different targets reconstructed by singular value decomposition. From the figure, it can be seen that the target micro-Doppler is enhanced after removing the main body speed.

[0041] For the time-frequency spectrum after removing the main speed, the rhythm speed spectrum is used to extract the frequency information of the periodic components in the time-frequency spectrum. The spectrum is obtained by performing Fourier transform along the time dimension of the signal's spectrum diagram. The formula can be expressed as: (3) in: is the original frequency of the signal, is the rhythm frequency, is the signal vector constructed from the radar echo, The time-frequency spectrum generated by performing a short-time Fourier transform on it, For The time-frequency spectrum reconstructed by singular value decomposition is Fourier transform is performed to reflect the change of micro-Doppler frequency and provide the motion state, size information and frequency information of the micro-moving components.

[0042] In the spectrum, the harmonic components of the signal appear as product relationships between frequencies in the frequency domain. By taking the logarithm of the spectrum, these product relationships can be converted into additive relationships, thereby simplifying the analysis process. Therefore, the cepstrum is introduced. For targets with rotors, under long-term observation, it can reveal the periodic characteristics in the spectrum. These characteristics are closely related to the flashing frequency of the blades, and the flashing frequency of the blades directly reflects the angular velocity of the rotor. In the case of multi-rotors, the cepstrum can even be further used to determine whether the target is a single rotor or a multi-rotor. The formula is as follows: (4) in: is the time index, is the frequency index, The calculation means calculating the power spectrum density of the time-frequency spectrum after removing the main body speed, which represents the energy distribution of the signal at different frequencies. It means taking the logarithm of the power spectrum density and converting the multiplication relationship into an addition relationship, which is convenient for separating the convolution components in the signal. IFFT means performing an inverse Fourier transform on the power spectrum after taking the logarithm, converting the frequency domain signal back to the time domain to obtain the cepstrum. The result of IFFT is modulo to obtain the real-valued cepstrum, which represents the intensity of the periodic components in the signal.

[0043] The improved lightweight network model is the improved MobileNetV3-Small, and its architecture can be divided into three main parts. The first part is the initial traditional convolution layer, which is responsible for extracting basic features from the input image. The convolution kernel size is 3×3 and the step size is 2. The second part consists of bottleneck layers. These layers use depthwise separable convolution to effectively reduce the computational complexity and model size. By stacking multiple modules and using the ECA attention mechanism module instead of the original SE module, these operations further reduce the network parameters and improve the recognition accuracy. The third part is the classification head, which reduces the number of fully connected neurons in the original network and reduces the number of modules. The bottleneck layer uses an inverted residual structure and uses depth-wise separable convolution, namely depth-wise convolution and point-wise convolution. The depth-wise convolution modifies the spatial dimension of the feature map, while the point-wise convolution adjusts the number of channels. The advantage of depth-wise separable convolution is that it can significantly reduce the computational complexity and model parameters without affecting the accuracy. The SE module is replaced by the ECA module in the bottleneck layer. This module only uses 1D convolution to model the dependencies between channels, thereby avoiding the computational overhead introduced by the fully connected layer and the dimensionality reduction and dimensionality increase process of the original SE module, thus realizing an efficient and lightweight channel attention mechanism.

[0044] Step S202: Use the multi-feature grayscale image fusion data to train the improved lightweight network model to obtain the optimal lightweight network model.

[0045] The multi-feature grayscale image fusion data is divided into training set and test set in a ratio of 8:2. The improved MobileNetV3-Small network is trained with the training set to obtain the trained network model, and the network model is tested with the test set to obtain the optimal lightweight network model.

[0046] Step S203, when receiving the drone identification request data, obtaining the fused grayscale image data of the drone in the drone identification request data; Step S204, bringing the fused grayscale image data into the best lightweight network model obtained through training, and using the best lightweight network model to perform drone recognition modeling.

[0047] This application enhances the target micro-Doppler characteristics through singular value decomposition reconstruction, integrates multiple micro-Doppler feature grayscale images, improves the target's dynamic feature capture capability, and uses a lightweight network structure to reduce the model complexity, making it suitable for efficient deployment of embedded devices and achieving accurate identification of drone targets.

[0048] Embodiment three: Figure 7 : is a schematic block diagram of a lightweight drone identification device according to an exemplary embodiment, the device comprising: The receiving module 310 is used to obtain the fused grayscale image data of the drone in the drone identification request data when receiving the drone identification request data; The recognition module 320 brings the fused grayscale image data into the best lightweight network model obtained through training, and uses the best lightweight network model to perform drone recognition modeling This application enhances the target micro-Doppler characteristics through singular value decomposition reconstruction, integrates multiple micro-Doppler feature grayscale images, improves the target's dynamic feature capture capability, and uses a lightweight network structure to reduce the model complexity, making it suitable for efficient deployment of embedded devices and achieving accurate identification of drone targets.

[0049] Embodiment 4: Figure 8 400 is a block diagram of a terminal provided in an embodiment of the present application, and the terminal may be a terminal in the above embodiment. The terminal 400 may be a portable mobile terminal, such as a smart phone or a tablet computer. The terminal 400 may also be referred to as a user equipment, a portable terminal, or other names.

[0050] Typically, the terminal 400 includes a processor 401 and a memory 402 .

[0051] The processor 401 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 401 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 401 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 401 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 401 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0052] The memory 402 may include one or more computer-readable storage media, which may be tangible and non-transitory. The memory 402 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 402 is used to store at least one instruction, which is used to be executed by the processor 401 to implement a lightweight drone identification method provided in the present application.

[0053] In some embodiments, the terminal 400 may further include: a peripheral device interface 403 and at least one peripheral device. Specifically, the peripheral device includes: at least one of a radio frequency circuit 404 , a touch display screen 405 , a camera 406 , an audio circuit 407 , a positioning component 408 and a power supply 409 .

[0054] The peripheral device interface 403 may be used to connect at least one peripheral device related to I / O (Input / Output) to the processor 401 and the memory 402. In some embodiments, the processor 401, the memory 402, and the peripheral device interface 403 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 401, the memory 402, and the peripheral device interface 403 may be implemented on a separate chip or circuit board, which is not limited in this embodiment.

[0055] The radio frequency circuit 404 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit 404 communicates with the communication network and other communication devices through electromagnetic signals. The radio frequency circuit 404 converts the electrical signal into an electromagnetic signal for transmission, or converts the received electromagnetic signal into an electrical signal. Optionally, the radio frequency circuit 404 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, and the like. The radio frequency circuit 404 can communicate with other terminals through at least one wireless communication protocol. The wireless communication protocol includes, but is not limited to: the World Wide Web, a metropolitan area network, an intranet, various generations of mobile communication networks (2G, 3G, 4G and 5G), a wireless local area network and / or a WiFi (Wireless Fidelity) network. In some embodiments, the radio frequency circuit 404 may also include circuits related to NFC (Near Field Communication), which is not limited in this application.

[0056] The touch display screen 405 is used to display a UI (User Interface). The UI may include graphics, text, icons, videos, and any combination thereof. The touch display screen 405 also has the ability to collect touch signals on the surface or above the surface of the touch display screen 405. The touch signal may be input as a control signal to the processor 401 for processing. The touch display screen 405 is used to provide virtual buttons and / or virtual keyboards, also known as soft buttons and / or soft keyboards. In some embodiments, the touch display screen 405 may be one, and the front panel of the terminal 400 is set; in other embodiments, the touch display screen 405 may be at least two, which are respectively set on different surfaces of the terminal 400 or are folded; in still other embodiments, the touch display screen 405 may be a flexible display screen, which is set on a curved surface or a folded surface of the terminal 400. Even, the touch display screen 405 may be set to a non-rectangular irregular shape, that is, a special-shaped screen. The touch display screen 405 may be made of materials such as LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode), etc.

[0057] The camera assembly 406 is used to capture images or videos. Optionally, the camera assembly 406 includes a front camera and a rear camera. Typically, the front camera is used to realize video calls or selfies, and the rear camera is used to realize photo or video shooting. In some embodiments, there are at least two rear cameras, which are any one of a main camera, a depth of field camera, and a wide-angle camera, so as to realize the fusion of the main camera and the depth of field camera to realize the background blur function, and the fusion of the main camera and the wide-angle camera to realize panoramic shooting and VR (Virtual Reality) shooting functions. In some embodiments, the camera assembly 406 may also include a flash. The flash can be a monochrome temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm light flash and a cold light flash, which can be used for light compensation at different color temperatures.

[0058] The audio circuit 407 is used to provide an audio interface between the user and the terminal 400. The audio circuit 407 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, and convert the sound waves into electrical signals and input them into the processor 401 for processing, or input them into the radio frequency circuit 404 to achieve voice communication. For the purpose of stereo acquisition or noise reduction, there may be multiple microphones, which are respectively arranged at different parts of the terminal 400. The microphone may also be an array microphone or an omnidirectional acquisition microphone. The speaker is used to convert the electrical signal from the processor 401 or the radio frequency circuit 404 into sound waves. The speaker may be a traditional film speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can not only convert the electrical signal into sound waves audible to humans, but also convert the electrical signal into sound waves inaudible to humans for purposes such as ranging. In some embodiments, the audio circuit 407 may also include a headphone jack.

[0059] The positioning component 408 is used to locate the current geographical location of the terminal 400 to implement navigation or LBS (Location Based Service).

[0060] The power supply 409 is used to power various components in the terminal 400. The power supply 409 can be an alternating current, a direct current, a disposable battery, or a rechargeable battery. When the power supply 409 includes a rechargeable battery, the rechargeable battery can be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery charged through a wired line, and a wireless rechargeable battery is a battery charged through a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0061] In some embodiments, the terminal 400 further includes one or more sensors 410 , including but not limited to: an acceleration sensor 411 , a gyroscope sensor 412 , a pressure sensor 413 , a fingerprint sensor 414 , an optical sensor 415 , and a proximity sensor 416 .

[0062] The acceleration sensor 411 can detect the magnitude of acceleration on the three coordinate axes of the coordinate system established by the terminal 400. For example, the acceleration sensor 411 can be used to detect the components of gravity acceleration on the three coordinate axes. The processor 401 can control the touch display screen 405 to display the user interface in a horizontal view or a vertical view according to the gravity acceleration signal collected by the acceleration sensor 411. The acceleration sensor 411 can also be used for collecting game or user motion data.

[0063] The gyro sensor 412 can detect the body direction and rotation angle of the terminal 400. The gyro sensor 412 can cooperate with the acceleration sensor 411 to collect the user's 3D (3 Dimensions) actions on the terminal 400. The processor 401 can implement the following functions based on the data collected by the gyro sensor 412: motion sensing (such as changing the UI according to the user's tilt operation), image stabilization during shooting, game control, and inertial navigation.

[0064] The pressure sensor 413 can be set in the side frame of the terminal 400 and / or the lower layer of the touch display screen 405. When the pressure sensor 413 is set in the side frame of the terminal 400, it can detect the user's holding signal of the terminal 400, and perform left and right hand recognition or shortcut operation according to the holding signal. When the pressure sensor 413 is set in the lower layer of the touch display screen 405, it can control the operability control on the UI interface according to the user's pressure operation on the touch display screen 405. The operability control includes at least one of a button control, a scroll bar control, an icon control, and a menu control.

[0065] The fingerprint sensor 414 is used to collect the user's fingerprint to identify the user's identity based on the collected fingerprint. When the user's identity is identified as a trusted identity, the processor 401 authorizes the user to perform relevant sensitive operations, including unlocking the screen, viewing encrypted information, downloading software, paying, and changing settings. The fingerprint sensor 414 can be set on the front, back, or side of the terminal 400. When a physical button or a manufacturer logo is set on the terminal 400, the fingerprint sensor 414 can be integrated with the physical button or the manufacturer logo.

[0066] The optical sensor 415 is used to collect the ambient light intensity. In one embodiment, the processor 401 can control the display brightness of the touch display screen 405 according to the ambient light intensity collected by the optical sensor 415. Specifically, when the ambient light intensity is high, the display brightness of the touch display screen 405 is increased; when the ambient light intensity is low, the display brightness of the touch display screen 405 is reduced. In another embodiment, the processor 401 can also dynamically adjust the shooting parameters of the camera assembly 406 according to the ambient light intensity collected by the optical sensor 415.

[0067] The proximity sensor 416, also called a distance sensor, is usually arranged on the front of the terminal 400. The proximity sensor 416 is used to collect the distance between the user and the front of the terminal 400. In one embodiment, when the proximity sensor 416 detects that the distance between the user and the front of the terminal 400 is gradually decreasing, the processor 401 controls the touch display screen 405 to switch from the screen-on state to the screen-off state; when the proximity sensor 416 detects that the distance between the user and the front of the terminal 400 is gradually increasing, the processor 401 controls the touch display screen 405 to switch from the screen-off state to the screen-on state.

[0068] Those skilled in the art will understand that Figure 8 The structure shown in the figure does not constitute a limitation on the terminal 400, and the terminal 400 may include more or less components than those shown in the figure, or combine some components, or adopt a different component arrangement.

[0069] Embodiment 5: In an exemplary embodiment, a computer-readable storage medium is also provided, on which a computer program is stored. When the program is executed by a processor, a lightweight drone identification method provided in all the inventive embodiments of the present application is implemented.

[0070] Any combination of one or more computer-readable media may be used. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with 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 thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or device.

[0071] Computer-readable signal media may include a data signal propagated in baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including, but not limited to, electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0072] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0073] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate 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 (e.g., via the Internet using an Internet service provider).

[0074] Embodiment 6: In an exemplary embodiment, an application product is also provided, including one or more instructions, which can be executed by the processor 401 of the above-mentioned device to complete the above-mentioned lightweight drone identification method.

[0075] Although the embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and the embodiments. It can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily realized. Therefore, without departing from the general concept defined by the claims and equivalent scope, the present invention is not limited to the specific details and the illustrations shown and described here.

Claims

1. A lightweight UAV identification method, characterized in that: include: When receiving the drone identification request data, obtaining the fused grayscale image data of the drone in the drone identification request data; The fused grayscale image data is introduced into the best lightweight network model obtained through training, and the best lightweight network model is used to perform drone recognition modeling.

2. The lightweight UAV identification method according to claim 1, characterized in that: When receiving the drone identification request data, before obtaining the fused grayscale image data of the drone in the drone identification request data, the method further includes: Obtain multi-feature grayscale image fusion data and build an improved lightweight network model for drone recognition; The improved lightweight network model is trained using the multi-feature grayscale image fusion data to obtain the optimal lightweight network model.

3. The lightweight UAV identification method according to claim 2, characterized in that: The step of obtaining multi-feature grayscale image fusion data includes: Acquire radar echo data, and perform short-time Fourier transform on the radar echo data to generate a time-frequency spectrum of the target; Performing singular value decomposition on the time-frequency spectrum and reconstructing an enhanced time-frequency spectrum; The enhanced time-frequency spectrum, rhythm-speed map and short-time cepstrum map are subjected to multi-dimensional grayscale image fusion to obtain multi-feature grayscale image fusion data.

4. The lightweight UAV identification method according to claim 3, characterized in that: The improved lightweight network model includes: Traditional convolutional layers, used to extract basic features from the input image; The bottleneck layer is used to model the dependencies between channels; The classification head is used to convert the extracted feature map into the final classification result.

5. The lightweight UAV identification method according to claim 4, characterized in that: The step of performing short-time Fourier transform on the radar echo data to generate a time-frequency spectrum of the target includes: Determining a range cell of a target and extracting a plurality of chirp signals from the range cell; Arrange the plurality of chirp signals in sequence into a vector, define the length and step size of the window function, and determine the sampling frequency as the inverse of the time interval between the chirp signals; Perform a short-time Fourier transform on the constructed signal vector to generate a time-spectrogram.

6. The lightweight UAV identification method according to claim 5, characterized in that: The step of performing singular value decomposition on the time-frequency spectrum and reconstructing the enhanced time-frequency spectrum comprises: For the input time-frequency matrix Perform singular value decomposition: （1） in, is the left singular vector matrix, is a diagonal matrix of singular values, is the right singular vector matrix; before removing After the principal components are obtained, the matrix is ​​reconstructed according to the remaining singular values ​​and singular vectors to obtain the enhanced time-frequency spectrum: （2） in, are the remaining singular values; is the corresponding left singular vector, is the corresponding right singular vector, For the matrix rank, is the reconstructed matrix.

7. A lightweight drone identification device, characterized in that: include: A receiving module, configured to obtain the fused grayscale image data of the drone in the drone identification request data when receiving the drone identification request data; The recognition module brings the fused grayscale image data into the best lightweight network model obtained through training, and uses the best lightweight network model to perform drone recognition modeling.

8. A terminal, characterized in that: include: one or more processors; a memory for storing said one or more processor executable instructions; Wherein, the one or more processors are configured to: Execute the lightweight drone identification method as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium, characterized in that: When the instructions in the storage medium are executed by the processor of the terminal, the terminal is enabled to execute the lightweight drone identification method as described in any one of claims 1 to 6.

10. An application product, when the application product is running on a terminal, enables the terminal to execute the lightweight drone identification method according to any one of claims 1 to 6.

Images