Unmanned aerial vehicle detection method, device, equipment and medium

Abstract

The embodiment of the invention discloses an unmanned aerial vehicle detection method and device, equipment and a medium. The method comprises the following steps: acquiring original audio data generated when an unmanned aerial vehicle works in a transformer substation; performing transformation processing on the original audio data to obtain transformed audio data; inputting the converted audio data into a sound generator to generate pseudo audio data; determining first audio data according to the pseudo audio data and the converted audio data; determining an unauthenticated unmanned aerial vehicle according to the first audio data and a set audio database; wherein the set audio database is determined based on a passive audio generator, and the set audio database comprises authentication audio data for authenticating the unmanned aerial vehicle. According to the technical scheme, the non-certified unmanned aerial vehicle is recognized through the audio data generated by the passive audio generator, the detection efficiency of the non-certified unmanned aerial vehicle is improved, battery endurance is not needed by utilizing the passive audio generator, and the load weight is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the field of signal processing technology, and in particular to a drone detection method, device, equipment and medium. Background Art

[0002] A drone is an aircraft that is remotely controlled or autonomously managed by a control station. With the advancement of technology and the reduction of costs, drones have been widely used in many fields such as military, agriculture, power inspection, law enforcement, environmental monitoring and film and television production.

[0003] Although drone technology has brought convenience, insufficient management can easily lead to some safety issues. For example, "illegal" drones have caused safety hazards in sensitive areas such as airports, nuclear power plants and borders. At present, the main methods of drone detection include radar detection, radio signal detection, acoustic signal detection, photoelectric detection and coordinated detection of multiple means.

[0004] Currently, one challenge facing the sound-based drone detection and identification technology is that the sound characteristics of "enemy" and "friend" drones may be very similar because they belong to the same brand, which poses a great challenge to the drone's sound identification. In order to deal with this problem, an active sound device can be carried on the drone to distinguish the sound of the propeller airflow, so that the drone has a unique sound characteristic. However, this method has the following problems: first, the active sound device is relatively large in size, which affects the stable flight of the drone; second, it requires batteries for power supply, which will also cause problems with the endurance of the sound device. Summary of the invention

[0005] The present invention provides a drone detection method, device, equipment and medium, which generates audio data through a passive audio generator to identify non-certified drones, thereby improving the efficiency of non-certified drone detection. The passive audio generator does not require battery life, which greatly reduces the load weight.

[0006] According to one aspect of the present invention, a method for detecting a drone is provided, comprising:

[0007] Collect raw audio data generated by drones working in substations;

[0008] Performing transformation processing on the original audio data to obtain transformed audio data;

[0009] Inputting the transformed audio data into a sound generator to generate pseudo audio data;

[0010] determining first audio data according to the pseudo audio data and the transformed audio data;

[0011] A non-certified drone is determined according to the first audio data and a set audio database; wherein the set audio database is determined based on a passive audio generator, and the set audio database includes authentication audio data of authenticated drones.

[0012] Optionally, before collecting the raw audio data generated by the drone working in the substation, it also includes:

[0013] Collecting audio data of the authentication drone; wherein the authentication drone is equipped with a passive audio generator, and the audio data is generated by the passive audio generator;

[0014] Encoding the audio data to obtain encoded audio data;

[0015] The set audio database is constructed according to the encoded audio data.

[0016] Optionally, the audio data is ultrasonic data;

[0017] Encoding the audio data to obtain encoded audio data includes:

[0018] The ultrasonic data is encoded by setting a frequency to obtain encoded audio data.

[0019] Optionally, the set audio database includes at least one authentication audio data;

[0020] Determining a non-certified drone according to the first audio data and a set audio database includes:

[0021] Comparing the first audio data with the at least one authentication audio data, and taking the audio data which is different from the first audio data and the authentication audio data as the second audio data;

[0022] The drone corresponding to the second audio data is determined to be an uncertified drone.

[0023] Optionally, transforming the original audio data to obtain transformed audio data includes:

[0024] Preprocessing the original audio data to obtain preprocessed audio data;

[0025] Performing a fast Fourier transform on the preprocessed audio data to obtain transformed audio data.

[0026] Optionally, inputting the transformed audio data into a sound generator to generate pseudo audio data includes:

[0027] The frequency characteristics of the transformed audio data are input into a sound generator to generate pseudo audio data.

[0028] Optionally, determining the first audio data according to the pseudo audio data and the transformed audio data includes:

[0029] Determine the audio data to be restored according to the pseudo audio data and the transformed audio data;

[0030] Performing an inverse fast Fourier transform on the audio data to be restored to obtain first audio data.

[0031] According to another aspect of the present invention, there is provided a drone detection device, comprising:

[0032] The data acquisition module is used to collect the original audio data generated by the drone working in the substation;

[0033] A data transformation processing module, used for transforming the original audio data to obtain transformed audio data;

[0034] A pseudo audio data generating module, used for inputting the transformed audio data into a sound generator to generate pseudo audio data;

[0035] A first audio data determining module, configured to determine first audio data according to the pseudo audio data and the transformed audio data;

[0036] The non-certified drone determination module is used to determine the non-certified drone according to the first audio data and a set audio database; wherein the set audio database is determined based on a passive audio generator, and the set audio database includes the authentication audio data of the authenticated drone.

[0037] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0038] at least one processor; and

[0039] a memory communicatively connected to the at least one processor; wherein,

[0040] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the drone detection method described in any embodiment of the present invention.

[0041] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the drone detection method described in any embodiment of the present invention when executed.

[0042] The technical solution of the embodiment of the present invention is to collect the original audio data generated when the drone is working in the substation; transform the original audio data to obtain transformed audio data; input the transformed audio data into a sound generator to generate pseudo audio data; determine the first audio data according to the pseudo audio data and the transformed audio data; determine the non-certified drone according to the first audio data and a set audio database; wherein the set audio database is determined based on a passive audio generator, and the set audio database includes the authenticated audio data of the authenticated drone. This technical solution generates audio data through a passive audio generator to identify non-certified drones, thereby improving the efficiency of non-certified drone detection, and the use of a passive audio generator does not require battery life, which greatly reduces the load weight.

[0043] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 is a flow chart of a drone detection method provided according to Embodiment 1 of the present invention;

[0046] Figure 2 is a flow chart of a drone detection method provided according to Embodiment 2 of the present invention;

[0047] Figure 3 is a circuit diagram of a passive audio generator provided according to Embodiment 2 of the present invention;

[0048] Figure 4 is a structural schematic diagram of a drone detection device provided according to Embodiment 3 of the present invention;

[0049] Figure 5 It is a schematic diagram of the structure of an electronic device provided according to Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0050] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0051] It should be noted that the terms "first", "second" and "target" in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0052] Embodiment 1

[0053] Figure 1 This is a flow chart of a drone detection method provided according to the first embodiment of the present invention. This embodiment is applicable to the case of detecting drones in a substation. The method can be performed by a drone detection device. The drone detection device can be implemented in the form of hardware and / or software. The drone detection device can be configured in an electronic device with data processing capabilities. Figure 1 As shown, the method includes:

[0054] S110, collecting original audio data generated when the drone is working in the substation.

[0055] The original audio data may be understood as the audio data generated when all drones in the substation are working. In this embodiment, the original audio data generated when the drone is working may refer to the sound data generated when the drone is flying. In this embodiment, the original audio data generated when the drone is working in the substation may be collected by a microphone array of the drone or a dedicated acoustic sensor.

[0056] It is understandable that the audio data of the drone is mainly generated by its rotor (for multi-rotor drones) or engine (for fixed-wing or hybrid drones). These sound characteristics vary depending on the type, size, design and operating mode of the drone. In order to identify the identity of the drone, in this embodiment, the substation’s own drones can be specially processed to emit ultrasonic audio information of a special frequency, and the identity of the drone can be authenticated to facilitate identity identification in the subsequent drone sound processing process. When an external drone approaches the defense range within the substation, the approach of the drone and the audio data without identity characteristics can be reliably identified, thereby triggering the operation of the directional countermeasure system. In this embodiment, the sound data generated by all current drones in the substation during flight can be collected.

[0057] S120: transform the original audio data to obtain transformed audio data.

[0058] The transformation process can be understood as a transformation process by converting the original audio data from the time domain to the frequency domain. The transformation process in this embodiment can include preprocessing and transformation operations of the sound signal.

[0059] Specifically, in this embodiment, since the collected original audio data may contain noise and unnecessary information, it is necessary to first perform a preprocessing operation on the original audio data, and then perform a fast Fourier transform FFT on the original audio data to convert the original audio data from the time domain to the frequency domain to obtain the transformed audio data.

[0060] In this embodiment, optionally, the original audio data is transformed to obtain transformed audio data, including: preprocessing the original audio data to obtain preprocessed audio data; and performing fast Fourier transform on the preprocessed audio data to obtain transformed audio data.

[0061] Among them, preprocessing can include filtering, denoising and segmentation of sound signals. The preprocessed audio data can be understood as data processed by filtering, denoising and segmentation of sound signals. Fast Fourier transform (FFT) is an efficient algorithm of discrete Fourier transform (DFT), which can be used to convert the time domain signal of the sound signal into the frequency domain signal, thereby realizing the analysis and processing of the frequency components of the audio signal. The transformed audio data can be understood as the frequency domain sound signal obtained by fast Fourier transform.

[0062] Specifically, in this embodiment, the original audio data can be preprocessed by filtering, denoising, and segmenting the sound signal to obtain preprocessed audio data, and then the preprocessed audio data can be subjected to a fast Fourier transform (FFT) to convert the audio data from the time domain to the frequency domain to obtain the transformed audio data. The FFT algorithm can quickly calculate the spectrum of the sound signal, that is, the amplitude and phase of different frequency components. In the frequency domain signal of the audio, the characteristics of the sound can be extracted, such as frequency distribution, energy distribution, frequency peak and other characteristics, which can be used to distinguish different types of drone sounds.

[0063] Through such a setting in this embodiment, the time-dependent audio data can be converted into frequency-dependent audio data, so as to facilitate the subsequent recognition of the drone sound.

[0064] S130. Input the transformed audio data into a sound generator to generate pseudo audio data.

[0065] Among them, the sound generator can be understood as a pre-trained neural network for generating sound. In this embodiment, the sound generator can be a trained generative adversarial network (GAN). The generative adversarial network GAN consists of two neural networks (NN): a generator NN model (called a generator) and a discriminator NN (called a discriminator). GAN realizes the generation and simulation of transformed audio data through adversarial training of two neural networks. Pseudo audio data can be understood as simulated sound data imitating a drone. In this embodiment, the transformed audio data can be input into the sound generator to generate simulated sound data imitating a drone.

[0066] In this embodiment, optionally, the transformed audio data is input into a sound generator to generate pseudo audio data, including: inputting frequency characteristics of the transformed audio data into the sound generator to generate pseudo audio data.

[0067] The transformed audio data packet contains a frequency feature. Specifically, the frequency feature may include frequency data and a level value at the frequency time point. In this embodiment, the frequency data and the level value at the frequency time point may be input into a sound generator, so that the sound generator generates pseudo audio data of the drone.

[0068] In this embodiment, through such a setting, the purpose of using a sound generator trained by a generative adversarial network to generate pseudo drone sounds is to process the audio captured by the actual microphone and improve the accuracy of the drone audio data.

[0069] S140: Determine first audio data according to the pseudo audio data and the transformed audio data.

[0070] The first audio data may be understood as audio data obtained by processing based on the pseudo audio data and the transformed audio data. In this embodiment, a specific method for determining the first audio data based on the pseudo audio data and the transformed audio data may be to perform a subtraction operation on the absolute value of the sound data of each frequency of the transformed audio data and the sound level data of each frequency of the pseudo audio data to obtain the audio data to be restored, and the audio data obtained by restoring the audio data to be restored is the first audio data.

[0071] S150: Determine a non-authenticated drone based on the first audio data and a set audio database.

[0072] Among them, the audio database is set based on the passive audio generator, and the audio database is set to include the authentication audio data of the authentication drone. The authentication drone can be understood as our drone that has been authenticated. In this embodiment, the authentication drone refers to a drone that can be trusted. The unauthenticated audio data can be the audio data generated by the authentication drone. The non-authenticated drone can be understood as a drone that has not been authenticated. The passive audio generator can be a sound generator installed on our authentication drone, which is used to generate corresponding audio data for our authentication drone. In this embodiment, the passive audio generator can provide power for the entire device through a passive design.

[0073] In this embodiment, the audio data of our certified drones in the substation are collected and processed to establish an audio database, so as to distinguish and identify uncertified drones. It can be understood that the audio database is set with each certified audio data corresponding to each certified drone. In this embodiment, the specific method of determining the uncertified drone based on the first audio data and the set audio database can be to compare the first audio data with the certified audio data in the set audio database. If the first audio data is consistent with any of the certified audio data in the set audio database, the drone corresponding to the first audio data is considered to be a certified drone; if the first audio data is inconsistent with any of the certified audio data in the set audio database, the drone corresponding to the first audio data can be considered to be a uncertified drone.

[0074] The technical solution of the embodiment of the present invention collects the original audio data generated by the UAV working in the substation; transforms the original audio data to obtain transformed audio data; inputs the transformed audio data into a sound generator to generate pseudo audio data; determines the first audio data according to the pseudo audio data and the transformed audio data; and determines the non-certified UAV according to the first audio data and a set audio database. This technical solution generates audio data to identify non-certified UAVs through a passive audio generator, thereby improving the efficiency of non-certified UAV detection. The use of a passive audio generator does not require battery life, which greatly reduces the load weight.

[0075] Embodiment 2

[0076] Figure 2 This is a flow chart of a drone detection method provided in accordance with the second embodiment of the present invention. This embodiment is optimized based on the above embodiment. The specific optimization is as follows: before collecting the original audio data generated by the drone working in the substation, it also includes: collecting the audio data of the authentication drone; wherein a passive audio generator is installed on the authentication drone, and the audio data is generated by the passive audio generator; encoding the audio data to obtain the encoded audio data; and constructing and setting an audio database based on the encoded audio data. Figure 2 As shown, the method includes:

[0077] S210. Collect audio data of the certified drone.

[0078] Among them, a passive audio generator is installed on the certification drone, and the audio data is generated by the passive audio generator. The audio data of the certification drone can be a sound signal generated by the certification drone when it is flying. In this embodiment, the audio data of the current certification drone during flight can be generated according to the passive audio generator installed on the certification drone. The audio data of the certification drone in this embodiment can be ultrasonic audio data.

[0079] Exemplarily, the circuit diagram of the passive audio generator in this embodiment is as follows: Figure 3 As shown in the figure, the entire passive audio generator includes a piezoelectric vibration energy capture module, an ultrasonic emission module and an ultrasonic frequency encoding module. Among them, the electric vibration energy capture module is used to generate electricity by capturing the airflow during the rotation of the drone propeller and causing the film to vibrate according to the captured airflow. The ultrasonic emission module can be used to emit ultrasonic waves. Therefore, ultrasonic waves are less likely to be imitated for drone identification. Figure 3 The circuit in the figure uses the 555 timer to generate a frequency above 20kHz and drive the buzzer to generate ultrasonic waves. The ultrasonic frequency encoding module can be used to change the resistance value of the resistor Rx to change the frequency generated by the 555 timer; Figure 3 As shown, different occurrence frequencies can be selected by dialing the external switch, and ultrasonic coding with different frequencies can be realized for different drones. In this embodiment, a passive audio generator with a passive design is used, and no additional battery module is required, which greatly reduces the load weight.

[0080] In addition, the passive audio generator in this embodiment uses the piezoelectric effect caused by the airflow of the drone to vibrate the film and supply power to the ultrasonic generating circuit. By using ultrasonic frequency coding, the entire ultrasonic emission frequency can be changed to play the role of coding the drone. Therefore, only when the drone is flying will it vibrate to generate electricity and emit ultrasonic waves. When it is not taking off, there is no audio feature, which is difficult to be cracked by the enemy.

[0081] S220: Encode the audio data to obtain encoded audio data.

[0082] Among them, encoding can be understood as the encoding operation of setting different frequencies of ultrasound for different audio data. The encoded audio data can be understood as the sound signal obtained by setting different frequencies of ultrasound for the audio data corresponding to different certified drones. The audio data is ultrasonic data, which can be generated by a passive audio generator. Among them, ultrasound usually refers to sound waves with a frequency greater than 20kHz, which cannot be heard by human ears.

[0083] In this embodiment, optionally, encoding the audio data to obtain the encoded audio data includes: encoding the ultrasonic data by setting the frequency to obtain the encoded audio data.

[0084] The set frequency may be a pre-set ultrasonic frequency. The set frequency in this embodiment may be one or more, and may be set according to actual needs. In this embodiment, by encoding ultrasonic data by setting the frequency, the same frequency may be set for ultrasonic data corresponding to different certified drones, or different ultrasonic frequencies may be set for ultrasonic frequencies corresponding to different certified drones to obtain encoded audio data.

[0085] For example, if there are three certified drones in the current substation, namely, certified drone No. 1, certified drone No. 2, and certified drone No. 3; the set frequency can have three different ultrasonic frequencies. In this embodiment, the ultrasonic data corresponding to the three certified drones can be encoded with different frequencies by setting the frequency. The 25kHz ultrasonic frequency can be selected for certified drone No. 1, the 30kHz ultrasonic frequency can be selected for certified drone No. 2, and the 40kHz ultrasonic frequency can be selected for certified drone No. 3. In this embodiment, since ultrasonic waves are not easily heard by human ears, they can play an encryption role, and by encoding the ultrasonic data, only the drone of the party knows the sound wave encryption frequency, and the unverified drone does not have this ultrasonic frequency feature.

[0086] Through such a setting in this embodiment, the audio data of the authentication drone can be encoded through different set frequencies, which is more conducive to identifying one's own drone.

[0087] S230: construct a set audio database according to the encoded audio data.

[0088] In this embodiment, the encoded audio data corresponding to all authenticated drones can be established to obtain a set audio database.

[0089] S240, collect the original audio data generated by the drone working in the substation.

[0090] S250: transform the original audio data to obtain transformed audio data.

[0091] S260: Input the transformed audio data into a sound generator to generate pseudo audio data.

[0092] In this embodiment, a conditional generative adversarial network (CGAN) method can be used to learn the sound of the drone. Among them, the generative adversarial network consists of two neural networks (NN): a generator NN model (called the generator) and a discriminator NN (called the discriminator). The training process of the initial sound generator can be: the training data sample set can include two dimensions: the frequency of the audio data and the level value at the frequency time point. In addition, the correlation can be determined by authenticating the drone sound characteristics. Therefore, the use of CGAN in this study is effective in terms of the amount of training data and the required training time. However, in order to learn the changes in the drone sound generated by the motor according to the changing flight state, a model based in part on the CGAN method called image-to-image translation is established. In the CGAN method, the input layer or input data of the generator can be pre-conditioned to produce arbitrary output results according to the conditions. The objective function of GAN is given by (1) below.

[0093] min G max D V(D, G) = E x～pdata(x) [logD(x,C)]+E x～pz(x) [log(1-D)(z,c)] (1);

[0095] Among them, E x～pdata(x) represents the expected value of the real drone audio data used for training, and represents the evaluation of the discriminator on these real data; on the other hand, D(x) represents the evaluation of the discriminator on these real data, and E x～pz(x) Represents the expected value when the pseudo drone sound (synthetically generated by the generator) is input to the discriminator. G(z) refers to the pseudo drone sound created by the generator from the noise input z, and D(G(z)) represents the discriminator's evaluation of the generated sound. In essence, D(x) evaluates the real audio data, while D(G(z)) evaluates the authenticity of the pseudo drone sound generated by the generator. Furthermore, in this embodiment, in order to ensure that the drone audio output generated by the generator is very close to the true value. The loss function of the generator is expressed in formula (2). Among them, G(x) represents the output of the generator when given noise x as input, and D(G(x)) represents the output of the discriminator given G(x), that is, the output of the generator.

[0096] L(G)=E x～pz(x) [log(xD(G(x)))] (2);

[0097] In this embodiment, the sound generator is obtained by training the initial sound generator, and the frequency in the transformed audio data and the level value at the corresponding time point of the frequency are input into the sound generator to obtain pseudo audio data. In this embodiment, the adversarial generative network GAN is used to suppress the audio noise of the drone, and only the sound waves of specific ultrasonic frequencies are retained, which can greatly improve the accuracy of drone detection and identification.

[0098] S270: Determine first audio data according to the pseudo audio data and the transformed audio data.

[0099] In this embodiment, optionally, determining the first audio data based on the pseudo audio data and the transformed audio data includes: determining the audio data to be restored based on the pseudo audio data and the transformed audio data; and performing an inverse fast Fourier transform on the audio data to be restored to obtain the first audio data.

[0100] The audio data to be restored may be obtained by performing a subtraction operation on the pseudo audio data and the transformed audio data. Inverse Fast Fourier Transform (IFFT) is the inverse operation of Fourier Transform, which is a way to remap the signal in the frequency domain to the time domain. The first audio data may be audio data obtained by performing an inverse fast Fourier transform operation on the audio data to be restored. It is understandable that the first audio data is audio data based on the time domain.

[0101] Specifically, in this embodiment, the microphone captures audio data every second and obtains the transformed audio data after processing. The absolute value of the sound level data of each frequency can be expressed as |X n |=[X0,X1,X2,...,X n ,]. Similarly, the absolute value of the sound level data of the pseudo audio data of each frequency per second generated by the sound generator is expressed as |Z n |=[Z0,Z1,Z2,...,Z n ,]. From |X n |minus|Z n The audio data to be restored is represented by |ω n |=[ω0,ω1,ω2,...,ω n ,], the subtraction process is shown in the following formula (3):

[0102]

[0103] The purpose of using a sound generator trained by a generative adversarial network to generate pseudo audio data in this embodiment is to process the audio captured by the actual microphone. The purpose of the subtraction operation obtained by performing a subtraction operation on the pseudo audio data and the transformed audio data in this embodiment is to suppress surrounding noise, thereby enhancing the clarity of the drone sound in the recording.

[0104] In this embodiment, after suppressing other drone sounds and surrounding noise from the frequency data, the audio data to be recovered is subjected to an inverse fast Fourier transform. Since the transformed audio data is time-based audio data converted to frequency-based data through a fast Fourier transform (FFT). In order to convert these frequency-based data back to the time domain, an inverse fast Fourier transform (IFFT) is used. This conversion is crucial for reconstructing the original audio characteristics while excluding the noise components that were identified and suppressed earlier. In this embodiment, the process of performing an inverse fast Fourier transform IFFT on the audio data to be recovered is shown in formula (4):

[0105]

[0106] Wherein, F(t) represents the discrete FFT; Z(t) represents the complex frequency domain representation of the audio data signal to be restored obtained from the subtraction result.

[0107] Furthermore, since the audio data to be restored ω n The value of is taken as an absolute value during the processing, and the positive and negative signs of the original audio data are lost. In this case, it is difficult to restore the sound through IFFT. In the sound restoration process, it is very important to maintain the correct sign (positive or negative) of each audio data point. In order to overcome this problem, the U-Net technology is also used in this embodiment. n The integration process is performed to obtain Z(t). Specifically, the U-Net method is able to reintegrate certain information at the input and output stages, bypassing the training and learning phase of the GAN neural network. This feature is crucial to prevent the loss of important data during the input process. Therefore, the U-Net method plays a key role in ensuring the integrity of the sign, phase, and volume of the data.

[0108] It should be noted that in this embodiment, due to the linear symmetry of frequency values ​​around the 0 axis in FFT, the values ​​on the positive axis of the frequency spectrum are trained. Then, the results are linearly copied to the negative axis, which simplifies the process while maintaining the accuracy of the audio data. Therefore, only half of the total training data set can be used in the training phase, greatly reducing the training time and processing requirements for training.

[0109] Through such a setting in this embodiment, the mixing of drone and surrounding noise in the audio data captured by the drone microphone can be effectively reduced, thereby improving the reliability and accuracy of the audio data.

[0110] S280: Determine a non-authenticated drone based on the first audio data and a set audio database.

[0111] The audio database is set based on a passive audio generator, and the audio database is set to include authentication audio data of the authentication drone.

[0112] In this embodiment, optionally, the audio database is set to include at least one authentication audio data; the unauthenticated drone is determined based on the first audio data and the set audio database, including: comparing the first audio data with at least one authentication audio data, and taking the audio data that is different from the first audio data and the authentication audio data as the second audio data; and determining the drone corresponding to the second audio data as an unauthenticated drone.

[0113] The audio database is set to include at least one authenticated audio data, and the specific number of authenticated audio data can be determined according to the number of authenticated drones. The second audio data can be understood as audio data that does not exist in the audio database, that is, audio data that has not been authenticated, that is, non-authenticated audio data.

[0114] In this embodiment, the first audio data is compared with at least one authentication audio data contained in the set audio database. If the first audio data is the same as the at least one authentication audio data contained in the set audio database, the first audio data is the authentication audio data corresponding to the authenticated drone; if the first audio data is different from the at least one authentication audio data contained in the set audio database, the audio data that is different from the first audio data and the authentication audio data is taken as the second audio data, indicating that the second audio data is the non-authenticated audio data corresponding to the non-authenticated drone. Therefore, the drone corresponding to the second audio data can be determined as a non-authenticated drone.

[0115] Through such a setting in the present embodiment, by comparing the authentication audio data with the collected drone audio data, the approach of the drone and the audio information without identity characteristics can be reliably identified, thereby triggering the operation of the directional countermeasure system.

[0116] Embodiment 3

[0117] Figure 4 Schematic diagram of the structure of a drone detection device provided according to Embodiment 3 of the present invention. Figure 4 As shown, the device comprises:

[0118] The data acquisition module 410 is used to collect the original audio data generated when the drone is working in the substation;

[0119] The data transformation processing module 420 is used to transform the original audio data to obtain transformed audio data;

[0120] A pseudo audio data generating module 430, configured to input the transformed audio data into a sound generator to generate pseudo audio data;

[0121] A first audio data determining module 440, configured to determine first audio data according to the pseudo audio data and the transformed audio data;

[0122] The non-certified drone determination module 450 is used to determine the non-certified drone according to the first audio data and the set audio database; wherein the set audio database is determined based on the passive audio generator, and the set audio database includes the authentication audio data of the authenticated drone.

[0123] Optionally, the device further comprises:

[0124] The audio data collection module is used to collect the audio data of the certification drone before collecting the original audio data generated by the drone when it is working in the substation; wherein a passive audio generator is installed on the certification drone, and the audio data is generated by the passive audio generator;

[0125] An encoding module, used for encoding the audio data to obtain encoded audio data;

[0126] The database construction module is used to construct a set audio database according to the encoded audio data.

[0127] Optionally, the audio data is ultrasonic data;

[0128] The encoding module is specifically used to encode the ultrasonic data by setting the frequency to obtain the encoded audio data.

[0129] Optionally, the audio database is set to include at least one authentication audio data;

[0130] The non-certified drone determination module 450 is specifically used to compare the first audio data with at least one certified audio data, and use the audio data that is different from the first audio data and the certified audio data as the second audio data; and determine the drone corresponding to the second audio data as a non-certified drone.

[0131] Optionally, the data transformation processing module 420 is specifically used to preprocess the original audio data to obtain preprocessed audio data; and perform fast Fourier transform on the preprocessed audio data to obtain transformed audio data.

[0132] Optionally, the pseudo audio data generating module 430 is specifically configured to input the frequency characteristics of the transformed audio data into a sound generator to generate pseudo audio data.

[0133] Optionally, the first audio data determination module 440 is specifically configured to determine the audio data to be restored according to the pseudo audio data and the transformed audio data; and perform an inverse fast Fourier transform on the audio data to be restored to obtain the first audio data.

[0134] A drone detection device provided in an embodiment of the present invention can execute a drone detection method provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0135] Embodiment 4

[0136] Figure 5 1 is a schematic diagram of the structure of an electronic device provided according to Embodiment 4 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0137] like Figure 5 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, ROM 12 and RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0138] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0139] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as the drone detection method.

[0140] In some embodiments, the drone detection method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the drone detection method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the drone detection method in any other appropriate manner (e.g., by means of firmware).

[0141] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0142] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0143] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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 foregoing.

[0144] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0145] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0146] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0147] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0148] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A drone detection method, characterized in that: include: Collect raw audio data generated by drones working in substations; Performing transformation processing on the original audio data to obtain transformed audio data; Inputting the transformed audio data into a sound generator to generate pseudo audio data; determining first audio data according to the pseudo audio data and the transformed audio data; A non-certified drone is determined according to the first audio data and a set audio database; wherein the set audio database is determined based on a passive audio generator, and the set audio database includes authentication audio data of authenticated drones.

2. The method according to claim 1, characterized in that Before collecting the raw audio data generated by the drone working in the substation, it also includes: Collecting audio data of the authentication drone; wherein the authentication drone is equipped with a passive audio generator, and the audio data is generated by the passive audio generator; Encoding the audio data to obtain encoded audio data; The set audio database is constructed according to the encoded audio data.

3. The method according to claim 2, characterized in that The audio data is ultrasonic data; Encoding the audio data to obtain encoded audio data includes: The ultrasonic data is encoded by setting a frequency to obtain encoded audio data.

4. The method according to claim 1, characterized in that: The set audio database includes at least one authentication audio data; Determining a non-certified drone according to the first audio data and a set audio database includes: Comparing the first audio data with the at least one authentication audio data, and taking the audio data which is different from the first audio data and the authentication audio data as the second audio data; The drone corresponding to the second audio data is determined to be an uncertified drone.

5. The method according to claim 1, characterized in that The original audio data is transformed to obtain transformed audio data, including: Preprocessing the original audio data to obtain preprocessed audio data; Performing a fast Fourier transform on the preprocessed audio data to obtain transformed audio data.

6. The method according to claim 5, characterized in that Inputting the transformed audio data into a sound generator to generate pseudo audio data, comprising: The frequency characteristics of the transformed audio data are input into a sound generator to generate pseudo audio data.

7. The method according to claim 1, characterized in that Determining first audio data according to the pseudo audio data and the transformed audio data includes: Determine the audio data to be restored according to the pseudo audio data and the transformed audio data; Performing an inverse fast Fourier transform on the audio data to be restored to obtain first audio data.

8. A drone detection device, characterized in that: include: The data acquisition module is used to collect the original audio data generated by the drone working in the substation; A data transformation processing module, used for transforming the original audio data to obtain transformed audio data; A pseudo audio data generating module, used for inputting the transformed audio data into a sound generator to generate pseudo audio data; A first audio data determining module, configured to determine first audio data according to the pseudo audio data and the transformed audio data; The non-certified drone determination module is used to determine the non-certified drone according to the first audio data and a set audio database; wherein the set audio database is determined based on a passive audio generator, and the set audio database includes the authentication audio data of the authenticated drone.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the drone detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the drone detection method according to any one of claims 1 to 7 when executed.

Images